MÉMOIRE D’ HABILITATION À DIRIGER DES RECHERCHES

UNIVERSITÉ D’ANTANANARIVO ÉCOLE SUPÉRIEURE POLYTECHNIQUE

LABORATOIRE DE SYSTÈMES ET INGÉNIERIE DE PROJETS INDUSTRIELS

Année : 2013 N° 001

MÉMOIRE D’ HABILITATION À DIRIGER DES RECHERCHES

présenté et soutenu publiquement par :

Dr. RAVALISON Andrianaivomalala François

devant le Jury composé de :

Antananarivo, 14 Mars 2013

Président : Professeur ANDRIANARY Philippe Ecole Supérieure Polytechnique Antananarivo

Rapporteur Interne : Professeur RAVELOSON Elisé (Titulus) Ecole Supérieure Polytechnique Antananarivo

Rapporteur Externe : GAZERIAN Joëlle, Maître de Conférences Hors Classe avec HDR Ecole Centrale de Marseille

Examinateur s : Professeur RAKOTOMARIA Etienne (Emeritus) Ecole Supérieure Polytechnique Antananarivo

: Professeur RAKOTOVAO José Denis (Titulus) Ecole Supérieure Polytechnique Antananarivo

DDDÉÉÉVVVEEELLLOOOPPPPPPEEEMMMEEENNNTTT DDD’’’UUUNNN MMMOOODDDÈÈÈLLLEEE DDDEEE SSSYYYSSSTTTÈÈÈMMMEEE

DDD’’’IIINNNGGGÉÉÉNNNIIIEEERRRIIIEEE ÀÀÀ PPPAAARRRTTTIIIRRR DDDEEE LLLAAA SSSYYYNNNTTTHHHÈÈÈSSSEEE DDDEEESSS

TTTRRRAAAVVVAAAUUUXXX DDDEEE RRREEECCCHHHEEERRRCCCHHHEEE

II

REMERCIEMENTS

Cette synthèse de travaux de recherche a fait l’objet de convergence de collaborations et de contributions de plusieurs personnes. Tout au long du processus, imposé par cette synthèse de travaux de recherche, elles ont montré une sympathie sans précédente et une persévérance remarquable. Je leur adresse tous mes sincères remerciements.

Mais ces remerciements et ces reconnaissances vont aussi à l’endroit des personnes suivantes qui ont joué un rôle déterminant à l’accomplissement du processus :

au Président de l’Université d’Antananarivo pour son dévouement pour le changement au sein de notre chère Université,

au Professeur ANDRIANARY Philippe, Directeur de l’Ecole Supérieure Polytechnique d’Antananarivo, d’avoir agréé la présidence de la soutenance,

au Professeur RAVELOSON Elisé, d’avoir accepté de donner son avis scientifique sur la synthèse des travaux de recherche,

à Madame GAZERIAN Joëlle, Maître de Conférences Hors Classe avec HDR, d’avoir fait un rapport scientifique sur la synthèse des travaux de recherche,

au Professeur RAKOTOMARIA Etienne, d’avoir accepté à siéger en tant qu’examinateur,

au Professeur RAKOTOVAO José Denis, d’avoir accepté d’examiner le Mémoire,

Je remercie sincèrement tous mes collègues de la filière Génie Industriel de l’Ecole Supérieure Polytechnique pour la démarche de collaboration dynamique ayant contribué à mon développement. Je voudrais aussi remercier l’équipe du Génie Electrique sous la houlette du Dr Rakotoniaina Solofo Hery pour les échanges fructueux. Et mes remerciements sont aussi formulés à l’endroit du Professeur Andrianaharison Yvon pour son encouragement. A travers lui, j’adresse, à tous mes collègues enseignant(e)s, mes remerciements.

Mes remerciements s’adressent aussi au Dr. Bhimsen Soragaon, Professeur au Departement d’Ingénierie Industriel et Management de JSS Academy of Technical Education, Bangalore-Inde, pour ses conseils et ses encouragements. Au Professeur Ricardo Valerdi, d’Arizona University, je le remercie de m’avoir orienté dans la filière « revue des articles pour publications internationales ». Je suis aussi reconnaissant au Professeur Timothy Simpson, du Pensylvania State University, de m’avoir initié dans le « learning factory ». Et je ne peux pas oublier de remercier mes amis le Professeur Suresh Sethi de l’University of Texas-Dallas et le Professeur Vijay Arora de l’Universiti Teknologi Malaysia.

Je voudrais également remercier mes collègues de l’équipe du Portland International Center for the Management of Technology (PICMET), de l’Institute of Industrial Engineers (IIE), de Southern African Institute for Industrial Engineering (SAIIE), de l’International Conference on Industrial Engineering and Operations Management (IEOM) et de International Council on Systems Engineering (INCOSE). Les échanges avec eux ont été des intrants précieux dans mon travail.

Enfin, je voudrais formuler mes remerciements les plus amicaux à Rabeariveloarisoa Aro et Soloarivelo Harinaina pour leur conseil en matière de duplication de cette synthèse.

A tous, merci !

III

A mon épouse Miarozò A mon fils Aaron et à mes deux filles Francia et Mirana A Dada et à la mémoire de Neny

IV

SOMMAIRE

INTRODUCTION GENERALE

CHAPITRE 1 : LES ARTICLES PUBLIÉS OU EN COURS DE PUBLICATION EN LIGNE

INTRODUCTION

CHAPITRE 2 : SYNTHÈSE DES ACTIVITÉS DE RECHERCHE

INTRODUCTION

MATERIELS

METHODOLOGIE

RÉSULTATS

DISCUSSION

CONCLUSION

CHAPITRE 3 : CURRICULUM VITAE

CONCLUSION GENERALE, PERSPECTIVES ET POSTULATION A L’HABILITATION A

DIRIGER DES RECHERCHES

BIBLIOGRAPHIE

V

TABLE DES MATIÈRES

INTRODUCTION GENERALE ................................................................................................. 1

CHAPITRE 1 : LES ARTICLES PUBLIÉS OU EN COURS DE PUBLICATION EN LIGNE .... 2

INTRODUCTION ..................................................................................................................... 2

CHAPITRE 2 : SYNTHÈSE DES ACTIVITÉS DE RECHERCHE .......................................... 95

INTRODUCTION ................................................................................................................... 95

MATERIELS .......................................................................................................................... 96

Scientométrie ..................................... ............................................................ 97

Degré de Collaboration ............................ ..................................................... 97

Revue des Pairs ................................... .......................................................... 98

Occurrence des mots clés .......................... .................................................. 98

Système d’Ingénierie .............................. ....................................................... 99

METHODOLOGIE ................................................................................................................. 99

RÉSULTATS ....................................................................................................................... 100

1. Contribution par format d’auteurs ................. ....................................... 100

2. Contribution dans le sens institutionnel .......... .................................... 101

3. Contribution internationale ....................... ............................................ 102

4. Revue des Pairs et Indexation ..................... ......................................... 103

5. Réseau de mots clés ............................... ............................................... 104

6. Synthèse des Travaux de Recherche ................. .................................. 105

DISCUSSION ...................................................................................................................... 111

Par rapport au Modèle de Système d’Ingénierie développé par Zarczynski ........................ 111

Par rapport au Modèle de Système d’Ingénierie proposé par Gonzalez et al. ..................... 111

Par rapport au Modèle de Système d’Ingénierie de Friedman et Sage ............................... 112

Par rapport au Modèle de Système d’Ingénierie de Crider et DeRosa ................................ 112

Par rapport au Modèle de Système d’Ingénierie d’Estefan .................................................. 113

Par rapport au Modèle de Système d’Ingénierie de Sheard ................................................ 113

Par rapport au Modèle de Système d’Ingénierie de Schoening et Miller ............................. 113

Par rapport au Modèle de Système d’Ingénierie de Honour ................................................ 113

Par rapport au Modèle de Système d’Ingénierie de Fugle ................................................... 114

CONCLUSION ..................................................................................................................... 115

CHAPITRE 3 : CURRICULUM VITAE ................................................................................. 117

VI

CONCLUSION GENERALE, PERSPECTIVES ET POSTULATION A L’HABILITATION A

DIRIGER DES RECHERCHES ........................................................................................... 124

BIBLIOGRAPHIE ................................................................................................................. 129

1

INTRODUCTION GENERALE

Le présent mémoire intitulé « Développement d’un Modèle de Système d’Ingénierie à partir de la

Synthèse des Travaux de Recherche » rentre dans le cadre de mémoire d’Habilitation à Diriger des

Recherches ou HDR à l’Ecole Supérieure Polytechnique de l’Université d’Antananarivo. Comme

travaux de recherche, il y a dix publications internationales et neuf encadrements de mémoire pour

l’obtention du Diplôme d’Etudes Approfondies. Le mémoire concerne, en effet, la synthèse de ces

travaux.

Ce mémoire comprend trois chapitres :

- Le Chapitre 1 est intitulé « Les Articles Publiés ou en cours de Publication en Ligne ». Ils

sont présentés tels qu’ils ont été soumis aux éditeurs. Et leur format ont suivi les guides

recommandées par ces derniers.

- Le Chapitre 2 est intitulé « Synthèse des Activités de Recherche ». Il s’agit de synthétiser

les travaux de recherche mentionnés ci-dessus. C’est l’aboutissement à un Modèle de

Système d’Ingénierie.

- Le Chapitre 3 est intitulé « Curriculum Vitae ». Il retrace l’éducation, les activités

d’enseignement, les activités de recherche et les activités administratives.

2

CHAPITRE 1 : LES ARTICLES PUBLIÉS OU EN COURS DE PUBLICATION EN LIGNE

INTRODUCTION Ce chapitre présente dix travaux de recherche, dont six sont des articles publiés et quatre en cours

de publication en ligne.

N° TRAVAUX DE RECHERCHE SITUATION

1 Assessment of Technology Management in a Context of Sustainable Development-The Case of a Paper Mill Industry in a Developing Country

Publié sur IEEE/PICMET

2 Are Converging Technologies Tools of Competitiveness?-The Case of a Paper Mill Industry in Madagascar Publié sur IEEE/PICMET

3 How does Reengineering Sustain Economy?-The Case of a Paper Mill Industry in Developing Country


4 Developing a Mathematical Concept and Process Technologies to Accompany Firms from Rut to Change


5 Towards “Fair Globalization”: Critical Success Factors for Partnering Project of Development and Enterprise Project Publié sur IEEE/PICMET

6 Using Patent Statistics and Principal Component Analysis to Predict Global Competition

Publié sur International Journal of Industrial

Engineering and Management (IJIEM)

7

Critical Success Factors are the Essence to Improve Productivity: A Technical Note on the Article “Jig Design, Assembly Line Design and Work Station Design and their Effect to Productivity”

Publication en cours sur Jordan Journal of

Mechanical and Industrial Engineering (JJMIE)

8 Hadamard Matrix to Improve Enterprise’s Activities: An Exploratory Research

Publication en cours sur International Journal of

Industrial Engineering and Management (IJIEM)

9 Using Principal Component Analysis to Determine Key Factor of Rural Electrification Development Investment



10 Quality Control Mapping to Assess Quality Assurance Performance in Textile and Garment Factories



Trois journaux de rang international ont servi de support de publication. Il s’agit de :

- L’Institute of Electrical and Electronics Engineers/Portland International Center for the

Management of Engineering and Technology ou IEEE/PICMET,

3

- L’International Journal of Industrial Engineering and Management ou IJIEM,

- Le Jordan Journal of Mechanical and Industrial Engineering ou JJMIE.

L’Institute of Electrical and Electronics Engineers /Portland International Center for the

Management of Engineering and Technology ou IEEE/PI CMET

L’Institute of Electrical and Electronics Engineers est une association professionnelle la plus grande et

la plus organisée dans le monde. Elle est dédiée pour promouvoir l’innovation technologique et

l’excellence au profit de l’humanité. IEEE et ses membres inspirent une communauté globale à

travers ses publications de haut niveau et hautement citées, ses conférences, ses standards

technologiques, et ses activités professionnelles et éducationnelles.

IEEE donne accès à des articles, à la littérature technique de plus haute qualité du monde, dans le

domaine de l’ingénierie et technologie.

Quant au Portland International Center for the Management of Engineering and Technology, elle est

une organisation à but non lucrative pour disséminer les informations sur le management de la

technologie à travers des publications internationales. Ces dernières font l’objet d’un « en double

aveugle »* et d’un « processus de revue des pairs »†.

IEEE et PICMET collaborent ensemble dans le cadre de publications internationales.

L’International Journal of Industrial Engineering a nd Management ou IJIEM

L’International Journal of Industrial Engineering and Management est un journal de l’University of Novi

Sad-Serbie. C’est un journal de rang international qui est publié quatre fois dans l’année. Il

sélectionne rigoureusement les articles à travers du processus « en double aveugle » et d’un

« processus de revue des pairs ».

Les domaines de l’IJIEM sont : systèmes de production, automatismes-robotiques-mechatronics,

qualité-maintenance-logistique, management de l’ingénierie, système de communication et

d’information, sécurité et exactitude, organisation et ressources humaines, entrepreunariat et

innovation, management de projet, marketing et commerce, investissement-finance-comptabilité,

ingénierie de l’assurance et management, ingénierie de média et management, éducation et pratique

en ingénierie industrielle et management.

Le Jordan Journal of Mechanical and Industrial Engi neering ou JJMIE

Le Jordan Journal of Mechanical and Industrial Engineering est un journal de haute qualité publié par

le Ministère Jordanien de l’Education Nationale et de la Recherche Scientifique en collaboration avec

* Traduction libre de « double blind » † Traduction libre de « peer review process »

4

le Hashemite University-Jordanie. Ce journal est consacré à l’ingénierie industrielle et mécanique

avec quatre parutions par an.

JJMIE est un journal de référence internationale avec « un processus de revue des pairs ». Son

objectif est de mettre à la disposition des chercheurs des produits de recherche hautement lisible et à

valeur ajoutée à la littérature qui servira d’outil de référence indispensable. Le journal couvre toutes

nouvelles théories et résultats expérimentaux dans le domaine de la mécanique et de l’ingénierie

industrielle ou tout autre domaine s’y rapportant. Le journal encourage les articles commentant les

recherches récentes ou des notes techniques.

Les articles publiés ou en cours de publication sont présentés ci-après :

5

Assessment of the Technology Management in a Context of Sustainable Development: The Case of Paper Mill Industry in a Developing Country

Ravalisonj, F.A.; Raveloson, E.A.; Rakotomaria, E.; Mananjean, D.; Randrianasolo, A.; Ecole Superieure Polytech., Univ. of Antananarivo This paper appears in: Technology Management for the Global Future, 2006. PICMET 2006

Issue Date: 8-13 July 2006 On page(s): 2372 - 2381

Location: Istanbul

Print ISBN: 1-890843-14-8 References Cited: 8 INSPEC Accession Number: 9376680

Digital Object Identifier: 10.1109/PICMET.2006.296824 Date of Current Version: 29 January 2007

ABSTRACT

The problems dealt in this assessment concern the quality of present and future life

system threatened and social and environmental system dragged by the industrial

option. The achievements are to assess the impact of the technology management. Quality management is the proposed methodology. It is a global strategy by which

enterprises manage the entire organization so that they excel on all dimensions of products that are important to the customer. Achievements are about evaluating the advantages and the disadvantages on the industry application of Technology

Management. The study is about appraising what happen while a Computer Integrated Manufacturing keeps on running and assessing what is the actual situation without that

innovative technology. From the above studies the following results are pointed out; the technology remedies contrasts between industrial and environmental strategies and the innovative technology brings luck to restart the competitiveness in a sustainable

development context. So technology improves not only the competitiveness of the enterprise and the national economy but also reconciles the binomial industrial

development and sustainable development

INDEX TERMS Available to subscribers and IEEE members.

REFERENCES Available to subscribers and IEEE members.

CITING DOCUMENTS Available to subscribers and IEEE members.

6

ASSESSMENT OF THE TECHNOLOGY MANAGEMENT IN A CONTEXT OF SUSTAINABLE DEVELOPMENT

The Case of Paper Mill Industry in a Developing Cou ntry François A. Ravalison1, Elisé A. Raveloson1, Etienne Rakotomaria1,

Diana Mananjean2, Albert Randrianasolo2 1University of Antananarivo, Ecole Supérieure Polytechnique, Bloc Technique Ankatso,

101 Antananarivo Madagascar 2Les Papeteries de Madagascar-Océan Indien, Ambohimanambola, POB 1756,

101 Antananarivo Madagascar

ABSTRACT The problems dealt in this assessment concern the q uality of present and future life system threatened and social and environmental syst em dragged by the industrial option. The achievements are to assess the impact o f the Technology Management. Quality Management is the proposed methodology. It is a global strategy by which enterprises manage the entire organization so that they excel on all dimensions of products that are important to the customer. Achiev ements are about evaluating the advantages and the disadvantages on the industry ap plication of Technology Management. The study is about appraising what happ en while a Computer Integrated Manufacturing keeps on running and assessing what i s the actual situation without that innovative technology. From the above studies the following results are pointed out; the technology remedies contrasts between indu strial and environmental strategies and the innovative technology brings luc k to restart the competitiveness in a sustainable development context. So technology im proves not only the competitiveness of the enterprise and the national economy but also reconciles the binomial industrial development and sustainable dev elopment.

KEYWORDS Productivity

Process technology

Dampness Waste Paper mill Computer Integrated Manufacturing Product technology Automated Control System Steam Weight

INTRODUCTION AND PROBLEMATICS “Industrialization and Sustainable Development”, is a binomial where on one hand, the first hinders the second and on the other hand the second penalizes the first. The African continent is confronted to this conflict of interests. And sometimes its competitiveness, in the concert of globalization, is at stake. As other countries, Madagascar meets this contradictory in paper mill industry. In 1997, the latter got equipped with a new technology to better control the produced paper quality. Unfortunately, in December 2004, a fire damaged that technology. Indeed currently, the problematics of this paper mill industry are:

- a continuous financial difficulty,

7

- the products do not comply with ISO standards, - an excessive consumption of forest resources, - precarious situation of the employees.

This being, what are the impacts of the technology failing in a paper mill industry? For that, we are going to do a comparative analysis of the produced paper in term of grammage and dampness profiles. Then, we are going to situate, by process flow charts: the consumption of the utilities as wood for steam production and the evolution of the grammage loss. From these analyzes, we are going to elaborate an economic study of the resulting externalities of the lack of the technology management. Finally some recommendations will be formulated in the axis of the sustainable development.

MATERIAL AND RESEARCH METHODOLOGY Material The study is going to carry on a quality control technology. It is a Computer Integrated Machine, Measurex MXOpen version 2.0. This technology is used fluently by the paper mill industries and has an impact on the continuous improvement of the quality of the production. We are going to assess the economic consequences of the externalities of the abandonment of use of the Measurex. The evaluation criteria are about:

- the grammage - the dampness - the source of energy

Figure 1: Measurex Research methodology We are going to use the Quality Management in our assessment of the industrial application of the technology. The Quality Management is a global strategy for managing the entire organization so that it excels on all dimensions of product that are important to the customer. This definition is more applicable than another used by P. Crosby: “conformance to specifications”.

MOTOR

CALCULATOR ROOM

CG

CG

To damp parts

DR

UM

SC

AN

NE

R

HE

AD

BO

X

SG

SG

8

The generic tools of Quality Management used in this research are: - Statistical Quality Control or SQC - Statistical Process Control or SPC

For the data collecting, we have chosen a type of paper " Cover: grammage = 250 g/m2, dampness = 5%" It concerns 35% of the Malagasy demand. Grammage and the dampness data, during 2002 and 2005, have been collected at “Les Papeteries de Madagascar-Indian Ocean” or PAPMAD-OI laboratory. During the same periods, the quantities of the Transformation Workshop output and the quantity of woods used to produce steams have been gathered. The software EXCEL has been used to stock data and to do a good part of the treatment. The rest has been treated on SPSS.

RESULTS ANALYSIS OF GRAMMAGE AND DAMPNESS PROFILES WHEN THE MEASUREX WAS FUNCTIONAL

Cross Direction Grammage Profile Measurex has been set up on Paper Machine. And data related to Grammage have been collected along cross direction and a sample is represented in the Fig. below.

Figure 2: Grammage Profile "Cross Direction" Source : the authors

- the variations of the grammage according to the " cross direction " are sinusoids and are between the upper and lower limits,

- in comparison with standard, a maximum grammage and a minimum grammage have been observed respectively to the 2nd and 5th measures,

- grammage profile is not regular compared with the standard and is not symmetrical to the median measure corresponding to 5th measure,

- three measures have revealed grammage superior to the standard while five have have marked grammage lower to the standard.

- a measure coincided with the standard.

220

225

230

235

240

245

250

255

260

265

1 2 3 4 5 6 7 8 9

Gram

mag

e

Measure rank

Grammage Standard Upper limit Lower limit

9

Dampness Profile Then data related to dampness have been collected along the same cross direction and the corresponding sample is given below.

Figure 3: Dampness Profile "Cross Direction" Source : the authors

- a maximal and a minimal dampness have been identified respectively at 2nd and 5th measures,

- three points have been included in the limits, three others were out of the maximal limit and the rests were out of the minimal limit,

- dampness profile is not regular compared with the standard and is not symmetrical to the median measure corresponding to 5th measure

Machine Direction Grammage Profile Along machine direction, data correlated to grammage have been gathered and the Fig. below represents a sample.

Figure 4: Grammage Profile "Machine Direction" Source : the authors

- the variations of the grammage according to the " machine direction " are included between the upper and lower limits,

- in comparison with standard, two minima grammage and a maximum grammage have been observed respectively at 1st – 3rd and 8th measures,

0

1

2

3

4

5

6

7

1 2 3 4 5 6 7 8 9

Damp

ness

[%

]

Measure rank

Dampness Standard Upper limit Lower limit

220

225

230

235

240

245

250

255

260

265

1 2 3 4 5 6 7 8

Gram

mag

e

Measure rank


10

- grammage profile is regular compared with the standard, - four measures showed a grammage superior to the standard and four others

presented a grammage lower to the standard. Dampness Profile Then, data corresponded to dampness along machine direction have been collected. A chart representing a sample is given below.

Figure 5: Dampness Profile "Cross Direction" Source : the authors

- six points have been consisted in the limits, a point was out of the maximal limit and another point was out of the minimal limit,

- dampness profile has a sinusoid pace in running direction ANALYSIS OF GRAMMAGE AND DAMPNESS PROFILES WHEN THE MEASUREX IS NOT ANYMORE FUNCTIONAL Grammage When Measurex is not anymore functional, we have processed data from laboratory. Measure rank is corresponding to data collecting date.

Figure 6: Grammage Profile Source : the authors

- the grammage profile is regular and superior to the standard at the 2nd until 7th dates,

- two points are between the upper and lower limits,

4.3

4.4

4.5

4.6

4.7

4.8

4.9

5

5.1

5.2

5.3

5.4

1 2 3 4 5 6 7 8

Dam

pnes

s %

Measure rank


210

220

230

240

250

260

270

280

1 2 3 4 5 6 7 8

Gram

mag

e

Measure rank


11

- the majority of points are out of the limits. Dampness In that case, our data are those collected by the laboratory team. The following chart shows the variations of a sample.

Figure 7: Dampness Profile Source : the authors

- two points have been consisted in the limits while six points were out of the minimal limit,

- from the 1st to the 4th day, the dampness is nearly constant but below the minimal limit,

- from the 5th day until the 8th day, another regular shape appears, but the position of the points below the minimum limit dominates

ANALYSIS OF LOSS IN GRAMMAGE Outputs of the Transformation Workshop have been observed. The data related to the grammage of sold paper and the one of corresponding standard paper have been collected and analyzed. Two periods are concerned: 2002 while the Measurex was functional and 2005 when it is not anymore functional. The following chart has been gotten.

Figure 8: Grammage Loss Source : the authors

- the variations of the grammage losses are very irregular, - the average grammage loss of 2005 is superior to the one of 2002,

4

4.2

4.4

4.6

4.8

5

5.2

5.4

1 2 3 4 5 6 7 8

Dam

pnes

s %

Measure rank


0.00

20.00

40.00

60.00

80.00

100.00

120.00

140.00

160.00

Jan Feb Mar Apr May Jun Jul Aug Sep

Loss

[Kg]

Month

2005 grammage loss:index weight based on 1000 kg

2002 grammage loss:index weight based on 1000 kg

12

- the maximum differential is noticed on January, - the minimum differential is marked on August, - on March there is not grammage loss

ANALYSIS OF THE CONSUMPTION OF WOOD The same periods as the analysis above have been considered. The histogram below represents the consumption of wood.

Figure 9: Consumption of Wood Source: the authors

The consumption of wood in 2005: - recorded a sharp decrease from January to February followed by a sharp growth

from February to March, - was on average constant from March until May, - marked a sharp rise from May to June

As for the consumption of wood in 2002: - she showed a quick rise from January to February followed by a sharp decrease

from February to March, - she was on average constant from March until June

DISCUSSIONS The technology management can contribute to solve the present distress of the developing country industrial system while aiming the sustainable development’s objectives: the integrity of the environment, the social fairness and the economic efficiency. The measures of appreciation appear through four meaningful parameters of manufacturing performances [1].

FEATURES AND QUALITY When the Measurex is not anymore functional, the automatic correction and adjustment operations of pressure of steam and outflow of pulp become manual. Manual operations related to grammage and dampness control must take place in laboratory. And after, the corrections and corresponding regulations intervene. Indeed, the paper features do not reach the conformity to the required quality [4]. The grammage is over the upper limit and the

0.00 2.00 4.00 6.00 8.00 10.00 12.00 14.00

Jan

Feb

Mar

Apr

May

Jun

Wood consumption (stere/ton reel)

Mon

th

2005 2002

13

dampness is below the lower limit. The paper becomes heavier and drier. The first case generates an additional cost [1] and the second one an extra cost of energy.

QUALITY AT SOURCE A paper mill industry consists of a Manufacture Unit and a Transformation Workshop. The first must assure a quality at source [1] so that inputs, that will supply the second, are in conformity with the required specifications. Quality at source determines Just-In-Time Production as in textile domain [6].

COSTS Salary cost When the technology is not anymore functional, the manual operations get back into practice, a frequent case in the developing country industries. The employees, accustomed to the use of the technology, require not only a certain time of restarting but also reinforcement. So, number of workers has risen and salary cost has increased but production has decreased. The trilogy of the total quality [4] is unbalanced. Energy cost Wood is used as source of energy for the boiler. Since no device permits to control anymore automatically the pressure of steam, the produced paper becomes drier and does not respect the standard required. There is a waste of energy which will overcharge the cost. That case penalizes the competitiveness. Environmental cost Anticipated annual use of 38 ha of forest is not pay for itself. A negative externality whose cost has not been considered by the enterprise nor in her selling price nor in her economic cost. This is the consequence of lack of technology management. Competitiveness The costs increased and the quality is not anymore in conformity with the standard. The competitiveness of the paper mill industry, opposite to the paper importing enterprises, is meeting increased difficulties.

ASSESSMENT OF THE ECONOMIC IMPACTS OF THE EXTERNALITIES The above results have highlighted many important parameters. The latter have guided the analysis of the externalities and have permitted to get the chart below.

Table 1: Analysis of Externalities

Source : the authors

Mm : Man x month st/tb : stere per ton reel

QUANTITATIVE SITUATION

PAPER MILL INDUSTRY SUSTAINABLE DEVELOPMENT

2002 2005 positive

externality negative

externality positive

externality negative

externality

Human Resources 36 Mm 84 Mm � �

Forest Resources 6,42 st/tr 8,13 st/tr � �

Grammage Loss 41,04 kg/t 99,21 kg/t � �

14

kg/t : kilogram per ton

The negative externalities generated by the failing of technology management in the paper mill industry influenced the sustainable development. Indeed, the capacity of the future generations is compromised according to Brundland Report. And a set of terrestrial natural ecosystems will be disrupted. Starting from two hypotheses:

- yearly production of the Manufacture Unit = 3000 tons - yearly production of Transformation Workshop = 2000 tons

And while knowing that : - the average salary cost is US $50.00 per month - the delivered on site price of a stere of wood is US $7.50 - the middle selling price of a ton of paper is US $1000.00

Considering these data, the assessment of the economic impacts of the externalities will be summed up in the following chart:

Tableau 2: Assessment of the Economic Impacts of th e Externalities Source : the authors

The paper mill industry has undergone a loss of US $118,740.00 and has created an obvious consequence on the ecological environment estimated at US $38,475.00. The survey could not value the other consequences that are certainly valuable. The unavailability of data did not permit to make as much. This assessment is very meaningful. And to this stage, we can already appreciate the damages and that it is not more useful to proceed by the calculation of the Economic Rate On Return. This last requires some details of investment data. But this is not the objective of this study. In the studied case, the failing of technology management has not only handicapped the paper mill industry but also had as tendency to accelerate the rhythm of use of the environment resources.

RECOMMENDATIONS

- international competition in foreign and domestic markets is one of the main factors motivating the introduction of microelectronics technologies [2], so the Measurex MXOpen version 2.0 should be reset rapidly and upgraded to improve the economic efficiency of the enterprise,

CONSEQUENCES OF THE EXTERNALITIES QUANTITY COST (US $)

Extra salary cost 48 Mm 2,400.00

Sale loss 116 t 116,340.00

Anticipated use of forest resources 38 ha 38,475.00

15

- a part of the profit made by the use of this technology should assure the maintenance of the integrity of the environment,

- still in the economic improvement of the enterprise, an optimal human resource management by Employee Counseling [3], while opening out the overstaffing in other stations, should be hired while putting on again the Measurex. It contributes to the improvement of the social fairness,

- the environmental costs and its social effects should be internalized [5] in the selling price of papers,

- some ad hoc environmental critical indicators should be elaborated and should be managed to observe the rhythm of use of the forest resources,

- the choice of the technology should have an impact on the cost, the delivery time, the quality and flexibility [1]. Success Critical Factors for each of these entities should be elaborated to observe the impact of the technology management on the economic efficiency of the enterprise,

- benchmarking is a tool of quality management [7], such tool should be set upamong the top management as a kaizen.

CONCLUSION The survey has presented that the absence of the Technology Management implies: a non competitive cost, a limited delivery speed, a neglected quality and an unknown flexibility. It has credited an economic loss for our national champion [8] in paper mill industry. Besides, such absence has not failed to mark print in the environment. Some negative externalities have been identified. The most important is the cutting down rhythm of the forest resources proved by their non valorized anticipated use. For the developing countries, some lessons deserve to be learned:

- it is sufficient to put on again and to make pay the existing technology : to increase the outputs in a sensitive manner in term of quality and quantity with the same quantity of inputs and to attenuate the negative externalities on the environment,

- once again the tools of Quality Management could demonstrate that systematic reinvestment in technology is not always useful, but adapting know how and expertise with the existing technology should be committed to solve the major part of industrial and environmental difficulties,

- the local industry must know how to take maximum advantage of the existing technologies but obviously in the long-term, a meaningful investment in advanced technology is inescapable to consolidate the regional and international competitiveness in the sustainable development context

- the obstacle is not poverty but the will of change not only to the level of the Top Management but until the one of the workers, in the case of the paper mill industry.

These steps of the paper mill industry are transferable to other industrial units of the developing countries where competitiveness restarting and improving depend on prospecting all available capacities, tools and technologies at their disposal.

16

In fact, another strong point of the survey is to have a new and efficient vision of the diffusion of the modern Production Management tools. If these last have been used judiciously, a relative gain of 63% would be credited to the advantages of the paper mill industries. If the Technology Management announces itself like tool of management to solve the problems while coupling “competitiveness” with “sustainable development”, would the strategic reorientation of the latter constitute a new approach for a world industrial new order?

REFERENCES [1] Aquilano N. and Chase R.,Production and Operations Management, 7th editon,

Irwin, Chicago, 1996, 821 pages [2] Duruiz L. and Yenturk N., Facing the Challenge, 1st edition, Iletisin Yayinlari,

Istanbul, 1992, 192 pages [3] Herzberg F., How do you Motivate your Employees, Harvard Business Review,

2003, volume 81, number 1, p.87 [4] Kélada J., Qualité Totale: amélioration continue et réingénierie, Edition Quafec,

Québec, 2000, 480 pages [5] Rakoto D., Etude de la Pollution du Marais Masay par la Modélisation

Mathématique et par la Méthode Intégrée d’Ingénierie de Projet, Thèse d’Etat, Université d’Antananarivo-Ecole Supérieure Polytechnique, Madagascar, 2005, 200 pages

[6] Ravalison A., Joromanantsoa H., Rarivoson A., Raveloson E., Rakotomaria E., Randriaherindrainy S., Evaluation de la Production Juste-A-Temps dans les Industries Textiles Malagasy à Vocation Exportation, Communication Scientifique, Toamasina Madagascar, 2005

[7] Raveloson E., Méthodologie de la Gestion de Projet dans des Pays En Développement-Application de la Méthode du Cadre Logique-Champs d’Etude Madagascar, Thèse d’Etat, Université d’Antananarivo-Ecole Supérieure Polytechnique, Madagascar, 1998, 249 pages

[8] Reich R., The Work of Nations, 1st edition, New York, 1991, 330 pages

17

Y SETTINGS

Are Converging Technology Tools of Competitiveness?

Ravalison, F.A.; Rajaonary, P.; Raveloson, P.E.A.; Rakotomaria, P.E.; Univ. of Antananarivo, Antananarivo

This paper appears in: Management of Engineering and Technology, Portland International Center

for Issue Date: 5-9 Aug. 2007

On page(s): 241 - 245 Location: Portland, OR

Print ISBN: 978-1-8908-4315-1

References Cited: 10

INSPEC Accession Number: 9870473 Digital Object Identifier: 10.1109/PICMET.2007.4349337 Date of Current Version: 15 October 2007

ABSTRACT

Madagascar is the only country of the Indian Ocean Commission possessing a paper industry. Since three years, that paper industry has confronted major fiber waste due to major

technological problems. The latter affects the competitiveness at national and regional level. The objective is to prove that solving technological problems solves competitiveness

problems. To approach this problem, the Skinner's competitiveness concept will be utilized.

And the methodologies will be the value engineering and the benchmarking. The purpose of value engineering is to simplify products and processes. Its objective is to achieve equivalent or better performance at a lower cost while maintaining all functional requirements defined

by the customer. The objective of process benchmarking is to understand and evaluate the current position of an industry in relation to "best practice" and to identify areas and means

of performance. The survey is going to analyze fiber waste, and then she will explore its

effect on competitiveness. A focal result appears: the rehabilitation of the existing technology and its combination with information technology improve competitiveness. That permits to conclude that the converging technologies are competitive quality process.

INDEX TERMS

Available to subscribers and IEEE members. REFERENCES

Available to subscribers and IEEE members. CITING DOCUMENTS

Available to subscribers and IEEE members.

18

ARE CONVERGING TECHNOLOGIES TOOLS OF COMPETITIVENESS?

François A. RAVALISON1, Elisé A. RAVELOSON1, Etienne RAKOTOMARIA1, Patrick RAJAONARY2 1University of Antananarivo, ESPA, Industrial Project Engineering, 101 Antananarivo, Madagascar

2Les Papeteries de Madagascar-PAPMAD-OI, POB 1756, 101 Antananarivo, Madagascar

ABSTRACT

Madagascar is the only country of the Indian Ocean Commission that possesses a paper industry. Since three years, that paper indus try has confronted major fiber waste due to major technological problems. The latter aff ects the competitiveness at national and regional level. The objective is to prove that solving technological problems solves competitiveness problems. To approach this p roblem, the Skinner’s competitiveness concept will be utilized. And the m ethodologies will be the Value Engineering and the Benchmarking. The purpose of Va lue Engineering is to simplify products and processes. Its objective is to achieve equivalent or better performance at a lower cost while maintaining all functional requi rements defined by the customer. The objective of Process Benchmarking is to underst and and evaluate the current position of an industry in relation to “best practi ce” and to identify areas and means of performance. The survey is going to analyze fiber w aste, and then it will explore its effect on competitiveness. A focal result appears: the rehabilitation of the existing technology and its combination with information tec hnology improve competitiveness. That permits to conclude that the converging techno logies are competitive quality process. KEYWORDS Benchmarking Paper Competitiveness Paper machine Competitive quality Process Converging technologies Production capacity Cost Product Fiber waste Productivity Hydrapulper Recuperator Information technology Value engineering

INTRODUCTION "We must re-establish our technological pre-eminence" confirmed an American civil servant. And according to the president of one company associated in Sematech, "It is our last luck. If we don't succeed in achieving this technology, we can go home". Those hard sentences on the technology have been mentioned by Robert Reich in “The Work of Nations". This same technology constitutes industrialization restarting luck of the developing countries in the actual globalized context. The case of Madagascar is marked by its integration in three regional organizations: Indian Ocean Commission or IOC, Common Market of Eastern and Southern Africa or COMESA

19

and Southern African Development Community or SADC. The correspondent regional competition threatens the Malagasy paper industry. This last one has some competitiveness problem but underneath it is a technological problem. The “recuperator” is no more functional. It is the cause of fiber waste. There is no suitable information technology that allows managing, in time, all production parameters simultaneously and to act consequently. The paper machine technology and raw materials are also other problem faces. Group of Technology and Competitiveness seem to be coupled. Are Converging Technologies Tools of Competitiveness?

APPROACH AND MATERIAL APPROACH We are going to use C. Wickham Skinner approach on measure manufacturing performance. From his work and others, four basic operations strategies were identified: cost, quality, speed of delivery and flexibility. Those four strategies translate directly into characteristics used to appreciate competitiveness.

Cost Every industry use cost to compete in a market or a segment of the market. Generally strategy makers base their decision on low cost to successfully compete in this niche.

Quality Quality can be divided into product quality and process quality. Product quality’s goal is to focus on the requirements of the customers. And the process quality’s target is to produce error-free products through total quality management.

Speed of Delivery It is the ability of an industry to provide dependable and fast delivery allows it to charge a premium price for its products.

Flexibility It refers to the ability of an industry to offer a wide variety of products to its customers[1]. MATERIAL We are going to study two paper industries: a South African paper industry and a Malagasy paper industry. Cost, Quality, Speed of Delivery, Flexibility and Non Recycled Fiber have been evaluated. Judgment criteria are based on competitivity.

METHODOLOGY BENCHMARKING Benchmarking is the process of identifying who is the best, who sets the standard, and what that standard is. The objective of Benchmarking is to understand and to evaluate the current position of a business in relation to “best practice” and to identify areas and means of

20

performance improvement. It can also be used to gain information regarding a company’s relative position in key business process and core competencies. Performance or Competitive Benchmarking is a type of Benchmarking. Industry considers its position in relation to performance characteristics of key products. To support continuous process, four phased methodology is used: plan, collect, analyze and adapt.

Fig. 1: Four phased Methodology of Benchmarking Plan During this phase, the study area, and key measures are clearly defined. The data collection tools and methods are established, tested and finalized. And research is conducted to identify the best practice company to study.

Collect A data-gathering template must be elaborated. It has too specific objectives: collect qualitative or quantitative data and learn from the selected company above.

Analyze Key activities during this phase include: analyzing trends, identifying practices that enable and hinder superior performance, and elaborating decisions elements. An action plan document is an output of that phase.

Adapt Implementation of new process is institutionalized and monitored. VALUE ENGINEERING The purpose of Value Engineering is to simplify products and processes. Its objective is to achieve equivalent or better performance at a lower cost while maintaining all functional requirements defined by the customer. Value Engineering does this by identifying and eliminating unnecessary costs. Technically, it deals with products already in production and is used to analyze product specifications and requirements as shown in production documents and purchase requests.

PLAN

COLLECT

ANALYZE

ADAPT

The Value Engineering analysis approach involverelated to the product.

COST AND QUALITY

Fig. 2 : Malagasy and South African Paper Industries' cost s

Two paper products have been observed: Office paper They are produced at the same time by the two according to their cost in USproducts are more expensive For the Office paper A4 and the African product and Malagasy product, is respectively 1.30 and 1.31. The South African products are 30% dearer than

Fig. 3 : Malagasy and South African

21

The Value Engineering analysis approach involves brainstorming based on some questions

RESULTS

: Malagasy and South African Paper Industries' cost s

have been observed: Office paper A4 and Generalare produced at the same time by the two paper industries. They have been valued

in US$ per packet. We have noticed that the expensive than the Malagasy ones.

A4 and the General printing application, the costAfrican product and Malagasy product, is respectively 1.30 and 1.31. The South African

than those of the Malagasy ones.

: Malagasy and South African Paper Industries' Process Quality

ISO

900

1/2

s brainstorming based on some questions

: Malagasy and South African Paper Industries' cost s

eneral printing application. industries. They have been valued

the South African paper

, the cost ratio, between South African product and Malagasy product, is respectively 1.30 and 1.31. The South African

Paper Industries' Process Quality

On the first hand, the South African paper free products through total quality industry practices no process qualit

Tab. 1: Malagasy and South African Paper industries' Prod uct Quality

Standards

H.S.

H.S.

H.S.

H.S.: Home Standard

The paper quality is observed according to three properties: Bendtsen Smoothness. They are unit of each property. The extreme columns industries. The South African paper industryto the ISO standards, for each of the three properties. home standards.

SPEED OF DELIVERY

Fig. 4 : Malagasy and South African Paper Industries' Spee d of Delivery

The above stick diagram representdelivery is the number of dayquantity, the South African paper days. The first is forty times faster

Malagasy paper industry

22

South African paper industry follows a process quality to produce errorfree products through total quality management. And in the second hand, t

practices no process quality.

: Malagasy and South African Paper industries' Prod uct Quality

Property Units of Measure

Standards

Basis Mass g/m2 ISO 536

Thickness µm ISO 534

Bendtsen Smoothness

ml/min ISO 8791/2

quality is observed according to three properties: Basis. They are presented in the second column. The third column gives the

property. The extreme columns give the standards practiced by the two

industry, which is the reference, produces productfor each of the three properties. Her Malagasy homol

: Malagasy and South African Paper Industries' Spee d of Delivery

diagram represents the speed of delivery of the two industrdelivery is the number of days necessary to deliver 1000 tons of paper

South African paper industry takes 2 days while the one of Madagascar takes 85 faster than the second.

South african

process quality to produce error-And in the second hand, the Malagasy paper

: Malagasy and South African Paper industries' Prod uct Quality

Standards

ISO 536

ISO 534

ISO 8791/2

asis Mass, Thickness and in the second column. The third column gives the

the standards practiced by the two paper

produces product quality according er Malagasy homologous practices

: Malagasy and South African Paper Industries' Spee d of Delivery

industries. The speed of to deliver 1000 tons of paper. To deliver this

industry takes 2 days while the one of Madagascar takes 85

South african paper industry

FLEXIBILITY

Fig. 5 : Malagasy and South

The number of varieties of products values were not specified. During this period, the varieties decreased while the 110 paper varieties in 1990. Currently varieties per year. The second produced 5 paper per year. In 2005, the South African paper Malagasy homologous can only present half

FIBER WASTE

Fig.

The curve gives the paper production from fiber as raw materials. These two quantities have been valued in kg. The assessment takes place The paces of the production and the raw materials are proportional. February, these paces have

23

: Malagasy and South African Paper Industries' Flexibility

The number of varieties of products was observed in 1990 and in 2005. The intermediate t specified. During this period, the Malagasy paper industry’s

s decreased while the South African paper’s one increased. The s in 1990. Currently it produces approximately ten, a regular decrease of 7

5 paper varieties in 1990 and now it produces South African paper industry offers a range of 14 products while her

Malagasy homologous can only present half of it.

Fig. 6: Malagasy Paper Industry's Fiber Waste

The curve gives the paper production from fiber as raw materials. These two quantities have been valued in kg. The assessment takes place from January to June of

of the production and the raw materials are proportional. have recorded a rise. Then, from February until April, a continuous

African Paper Industries' Flexibility

observed in 1990 and in 2005. The intermediate Malagasy paper industry’s number of

increased. The first one produced ten, a regular decrease of 7

now it produces 14, a rise of 1 variety ffers a range of 14 products while her

The curve gives the paper production from fiber as raw materials. These two quantities have June of 2005.

of the production and the raw materials are proportional. From January to recorded a rise. Then, from February until April, a continuous

decrease has been observed. After a place from May to June. The difference, between the quantity of rthe non recycled waste. This March-April. This waste became important in Aprilobserved in January-March. On waste.

MULTICRITERIA ANALYSIS

Fig. 7 : Multicriteria Analysis of Malagasy and South Afri can Paper Industries

The studied criteria are: quality, cost, speed of delivery, flexibility fibers. Each criterion is a processSouth African paper industry ISO standards’ requirements. Then, determines the speed of delivery. This same production system of paper products and to reduce the quantity of the non recycledsystems have been noted 5 out ofAnd proportionally or inversely proportional to Malagasy paper industry have been deducted. and has been scored 5 out of score has been deducted. The curves of profile show well that the is vast and balanced. While the The first’s points relating to quality, fibers are reference points. They

Converging Technologies areindustry to be competitive. But case of Malagasy paper industry

24

decrease has been observed. After a new increase from April to May, a new decrease takes

The difference, between the quantity of raw materials and the quantity of production, gives . This waste was important between January-became important in April-June but while being lower t

On the average, there was 70,000 kg per month of non recycled

: Multicriteria Analysis of Malagasy and South Afri can Paper Industries

: quality, cost, speed of delivery, flexibility and quantity of non recycledcriterion is a process’ output.

has a quality system allowing it to take account previously the requirements. Then, it has a production system whose productivity

es the speed of delivery. This same production system has permitted to get and to reduce the quantity of the non recycled fibers. On a

out of 5, except for the cost. r inversely proportional to South African paper industry’s

have been deducted. The latter has a less expensive 5. According to this last, the South African paper industry’s c

The curves of profile show well that the South African paper industry is a referenceis vast and balanced. While the Malagasy paper industry’s area is narrow

quality, speed of delivery, flexibility and quantity of non recycledpoints. They spread on the whole plan.

DISCUSSIONS are tools of competitiveness. They allow

to be competitive. But lack of them has handicapped the competitivenesscase of Malagasy paper industry.

April to May, a new decrease takes

aw materials and the quantity of production, gives -March. It decreased in

June but while being lower than the one average, there was 70,000 kg per month of non recycled

: Multicriteria Analysis of Malagasy and South Afri can Paper Industries

and quantity of non recycled

to take account previously the has a production system whose productivity

permitted to get variety fibers. On a 0-5 scale, these

South African paper industry’s scores, those of has a less expensive cost system South African paper industry’s cost

is a reference: her area is narrow and unbalanced.

delivery, flexibility and quantity of non recycled

allow South African paper handicapped the competitiveness, as the

25

During the period of 1975-1985, Malagasy domestic market was in a difficult situation. Importation was very restricted and imported paper products were considered as deluxe products. Malagasy customers did not require quality products anymore. So, strategic decision makers, in paper industry, decided to reduce the variety of paper products. And focusing on low quality product at low cost was the strategy to master that market. Some elements of manufacturing and information technologies had been given up. But from 1990, Malagasy government has changed his policy and has adopted open free market. Malagasy customers’ behavior has been remodelled by imported quality products in general and by imported quality paper products in particular. Besides, the office technologies, as printer-photocopier-fax, require standardized papers. So, low quality products, even low cost, have been automatically neglected. Reducing flexibility and decreasing quality have been maintained till now and have affected Malagasy paper industry’s competitiveness. Behind those two basic operations strategies[6] are converging technologies mentioned above. In addition, speed of delivery, affected by manufacturing and recycling technologies, does not motivate purchasing decisions. So, charging a premium price for her products, even low cost, has not been realized. Besides, South African paper industry has increased her flexibility. She has implemented process quality to produce quality products with zero fiber waste. So, even high cost paper products practiced, she dominates Malagasy market by economy of scale. In the automobile area, for example, Japanese automobile firms increased their international competitiveness by producing a wider range of cars[7]. They have implemented Just-In-Time as approach to productivity and to quality[2]. They have assured, through Value Engineering, that cost will be the lowest[5]. So, mastering the four basic operations strategies of Skinner with zero fiber waste, by converging technologies, allows Japanese firms to be competitive.

The speed of delivery depends on manufacturing technology, recycling technology and information technology. For Malagasy paper industry: the manufacturing technology is frequently breakdown[3], the recycling[3] and information technologies are not functional anymore[4]. Those technology problems generate not only productivity decrease but also quality defective. The competitiveness of the Malagasy paper industry has been weakened by those three converging technologies. On the opposite, the synergy of those three technologies strengthens the competitiveness of South African paper industry.

The South African paper industry is competitive on the Malagasy market. She has technologies allowing her to master quality, speed of delivery, flexibility and fiber waste. And with the economy of scale, she could sell her products at a competitive price in Madagascar. In addition to the technologies, she adopts ISO assurance quality. The local paper industry seems to be in good position concerning cost. That position is vulnerable in front of technological strength coupled by an economy of scale and process quality. Hard technology and soft technology have shaped the South African paper industry’s competitiveness.

Value engineering is entirely neglected by Malagasy paper industry.

26

Fig. 8: Proposed Value Engineering for Malagasy Pap er Industry

The total cost before Value Engineering is $388.39. It s obtained from Machine Paper’s operating cost and Transformation’s one[3]. According to Lawrence Miles in « Value Analysis », 25% to 75% of operating cost are unnecessary costs. So, if 25% is considered, and if quality paper is the target, the following cost review would be operated : Outside Service, Fixed Costs and Structure Costs reduced by 25%, Other Raw Materials increased by 100%, and Raw Materials maintained. Actually, the first operation is conducted. Target is fixed at 50%. The second is not in decision makers’ strategy. Use of chlorine is indicated to obtain 90% white paper. But chlorine is opposite to environment. So, increasing other Raw Materials, especcially in order to obtain white paper and to eliminate chlorine’s effects, is a better performance. The third operation is to maintain a standard quantity for a tonne of paper. The total cost after Value Engineering is US$360.53. The paper quality is improved and is tending to ISO standard but the cost is becoming lower. So, Malagasy paper industry could offer very competitive quality paper if she achieves Value Engineering process.

CONCLUSION The comparative survey between South African paper industry and Malagasy paper industry has been conducted by benchmarking. The assessment and the comparison have permitted to put in evidence that the first paper industry mobilizes a group of hard technology coupled with soft technology based on management quality. This has built its competitiveness on the Malagasy market. But this comparative survey has also permitted to appreciate the competitiveness of the Malagasy paper industry. This competitiveness has been weakened by the failing of technology composed of manufacturing technology, recycling technology and information technology. Her non competitiveness on her own market is due to the failing of her converging technologies. Here, we would like to point out some of our findings and to share some lessons learned, to restructure technology management and industrialization policies in the developing countries in responding to competition requirements:

Before Value Engineering

After Value Engineering

(US$170.30)Raw Materials = Chemical Pulp

(US$42.68)Other Raw Materials

(US$55.84)Outside Service

(US$30.45)Fixed Costs

(US$61.26)Structure Costs

Raw Materials = Old Paper(US$170.30)

Other Raw Materials(US$21.34)

Outside Service(US$74.46)

Fixed Costs(US$40.60)

Structure Costs(US$81.69)

27

Developing countries’ paper industry is aware of the importance of converging technologies in the world competition. But they are limited in renewing them, so repairing and restarting the actual converging technologies may be a way and a chance to restart their competitiveness.

Paper manufacturing technology is what we call « hard technology ». It needs to be accompanied by total quality management, what we call « soft technology ». This is important in order to avoid underutilization, non quality and above all non competitiveness.

Paper manufacturing technology should be coupled with information technology to get much key information on time.

In the actual globalization context, and according to « Washington consensus », doors are opened for foreign investments. The latter will introduce converging technologies and will significantly master local and regional markets. The Government considerable strategic role is to encourage « foreign investments » and so « hard converging technologies ». How will be developing countries’ paper industry, with their converging technologies that need repairing and restarting, accompanied in front of « foreign investments » with « hard converging technologies »?

BIBLIOGRAPHY [1] Aquilano N. and Chase R., Production and Operations Management, 7th editon,

Irwin, Chicago, 1996, 821 pages [2] Kenneth W., The Japanese Approach to Productivity, Southfield, MI: Bendix

Corporation, 1983 [3] Ravalison A., Contribution à l’Amélioration du Système de Management par la

Gestion de Production-Le cas des Papeteries de Madagascar Océan Indien, DEA dissertation, University of Antananarivo-Ecole Supérieure Polytechnique, 2005, 63 pages

[4] Ravalison A., Randrianasolo A., Pr Raveloson E., Pr Rakotomaria E., Assessment of Technology Management in a Context of Sustainable Development-The case of a paper mill industry in a developing country, Portland State University-PICMET, 2006

[5] Reich R., The Work of Nations, 1st edition, New York, 1991, 330 pages

[6] Skinner W., Manufacturing: The Formidable Competitive Weapon, New York, John Wiley and Sons, 1985

[7] Urata S., The Development of the Motor Vehicle Industry in Post-second-world-war in Japan, Industry and Development n°24, UNIDO

WEBSITES www.fao.org www.paper.org.uk www.unido.org www.paperloop.com www.mondi.com

28

How does reengineering sustain economy? The case of a paper

industry in a developing country

Ravalison, F.; Rajaonary, P.; Raveloson, E.; Rakotomaria, E.; Gazerian, J.; Loubet, C.; Ruiz, J.M.;

Ecole Super. Polytech., Antananarivo Univ., Antananarivo

This paper appears in: Management of Engineering & Technology, 2008. PICMET 2008. Portland

International Conference on

Issue Date: 27-31 July 2008 On page(s): 210 - 219

Location: Cape Town Print ISBN: 978-1-890843-17-5

References Cited: 25 INSPEC Accession Number: 10151968

Digital Object Identifier: 10.1109/PICMET.2008.4599626 Date of Current Version: 15 August 2008

ABSTRACT

Madagascar has a paper mill industry that is not competitive with regional paper industries

which sell on the Malagasy market. Lack of quality process and product quality disadvantages

the paper mill industrypsilas products. Non reliability of production capacity and under utilization are factors that decrease the speed of delivery. Its varieties of product are very limited compared to the local market needs. The objectives of study are to improve

dramatically the paper industrypsilas performance in production and competitiveness. Reengineering methodology will be utilized. Reengineering is "the fundamental rethinking

and radical redesign of business processes to achieve dramatic improvements in critical,

contemporary measures of performance, such as cost, quality, service and speed". Study takes place on the paper transformation unit or PTU process of a paper mill industry. Outputs of the PTU are observed and analyzed. The process is then reengineered and a new design

based on new organization of process is obtained. When applying reengineering, some dramatic improvement in the speed of delivery, quality, flexibility and dependability are

obtained. Reengineering sustains paper industry by restarting its competitiveness in the

paper industry sector, helping to sustain the economy.

INDEX TERMS




29

HOW DOES REENGINEERING SUSTAIN ECONOMY? The Case of a Paper Industry in a Developing Countr y

François Ravalison1, Patrick Rajaonary2, Elisé Raveloson1, Etienne Rakotomaria1, Joëlle Gazérian3, Cécile Loubet3, Jean Michel Ruiz3

1University of Antananarivo, Ecole Supérieure Polytechnique, 101 Antananarivo, Madagascar 2Les Papeteries de Madagascar-Océan Indien or PAPMAD-OI, 101 Antananarivo, Madagascar

3Ecole Centrale de Marseille, Marseille, France

ABSTRACT Madagascar has a paper mill industry that is not co mpetitive with regional paper industries which sell on the Malagasy market. Lack of quality process and product quality disadvantages the paper mill industry’s pro ducts. Non reliability of production capacity and under utilization are factors that dec rease the speed of delivery. Its varieties of product are very limited compared to t he local market needs. The objectives of study are to improve dramatically the paper industry’s performance in production and competitiveness. Reengineering metho dology will be utilized. Reengineering is “the fundamental rethinking and ra dical redesign of business processes to achieve dramatic improvements in criti cal, contemporary measures of performance, such as cost, quality, service and spe ed”. Study takes place on the Paper Transformation Unit or PTU process of a paper mill industry. Outputs of the PTU are observed and analyzed. The process is then reen gineered and a new design based on new organization of process is obtained. When ap plying reengineering, some dramatic improvement in the speed of delivery, qual ity, flexibility and dependability are obtained. Reengineering sustains paper industry by restarting its competitiveness in the paper industry sector, helping to sustain the e conomy. KEYWORDS Process Speed of delivery Quality Paper Dependability Competitiveness Flexibility Reengineering Dramatic Improvement

INTRODUCTION – PROBLEMATICS – HYPOTHESIS – STEP Madagascar is suffering the impacts of globalization like many of the other developing countries. The majority of local production of goods is impacted and paper products are not spared. In 1985, while the Malagasy economy was being restructured, local products were disadvantaged by open market strategy and exchange rate liberalization. For the Malagasy Paper Industry or MPI, liberalization of the exchange rate has made her debt situation worse. The MPI is obliged to pay again its debt. Open market strategy has disoriented local paper products. The paper industry has not carried out in a timely manner necessary fundamental rethinking of its business process. As a result, sales are dropping, which generally tends to happen under the globalization process. Two key questions arise: “How Does Reengineering Face that Process? And How Does Reengineering Sustain Developing Country Economy?”

30

To answer these important questions, here are some hypothesizes: - The Malagasy paper industry is a system - That system is defined by five parameters such as quality, flexibility, speed of

delivery and dependability. Problem approach is based, first on part of Skinner’s metrics such as quality, speed of delivery and flexibility; and secondly on dependability issued from real time system. The process for reengineering will be identified. Then a new organization process will be suggested. Discussion relating to that organization process’s key outputs is presented.

I. METHODOLOGY: “REENGINEERING”

a. Definition Reengineering is a process of fundamental rethinking and radical redesign of business processes to achieve dramatic improvements in critical, contemporary measures of performance, such as cost, quality, service and speed [9].

b. Principles of Reengineering and Process Redesign Techniques and Tools Reengineering is about achieving a significant improvement in processes so that contemporary customer requirements are met [1]. This entails seven principles of doing work and six techniques and tools of processing innovation [10].

Table 1: Seven Principles of Reengineering and Six Techniques and Tools

THE SEVEN PRINCIPLES OF REENGINEERING THE SIX TECHNIQUES AND TOOLS

PRINCIPLE 1: Organize around outcomes not tasks 1: Inductive

PRINCIPLE 2: Have those who use the output of the process perform the process

2: Flowcharting

PRINCIPLE 3: Merge information-processing work into the real work that produces the information

3: Creative Process Redesign

PRINCIPLE 4: Treat geographically dispersed resources as though they were centralized

4: Process Benchmarking

PRINCIPLE 5: Link parallel activities instead of integrating their results 5: Simulation

PRINCIPLE 6: Put the decision point where the work is performed and build control into the process

6: Reengineering Software

PRINCIPLE 7: Capture information once at source

c. The Reengineering Process Reengineering requires innovation, which must be done in six steps.

Figure 1: The Reengineering Process

STEP 1 State a case for action

STEP 2

Identify the process

for reengineering

STEP 3

Evaluate enablers of

reengineering

STEP 6

Implement the

reengineered process

STEP 5

Create a new process

design

STEP 4

Understand the

current process

31

II. RESULTS

a. Identification of process for reengineering

Figure 2: MPI process

Figure 3: The process for reengineering

The Paper Machine or PM is supplied pulp from imported old paper or local waste of cardboard. PM then supplies Paper Transformation Unit or PTU by reels. PM transfers to PTU the technological problems of quality, speed of delivery, flexibility and dependability. Measurex, equipment for quality control, is out of service. So, reels of paper, delivered by PM to PTU, are of poor quality. Breakdown and pulp waiting times affect PM’s speed of delivery and production variety, and adversely affect PTU also. Breakdown waiting time shows a failure in maintenance process and influences PTU’s dependability.

Producing superheating steam

Supplying electricity

Paper transformation unit Selling Paper machine

Importing old papers

Collecting cardboard

PROBLEM SCOPE

PAPER MACHINE PAPER

TRANSFORMATION UNIT

Drop in Dependability

Drop in Speed of delivery

Drop in Quality

Drop in Production variety

Failure in Maintenance process

Drop in Speed of delivery

Drop in Quality

PULP

Drop in Flexibility

Pulp waste

Pulp waiting time Breakdown waiting time

32

b. Evaluation of enablers of reengineering Table 2: SWOT of human resources

Strengths Weaknesses 1 Highly qualified staff and workers 1 Some staff and workers discouraged 2 Serious staff and workers 2 Departure of some highly qualified staff and workers 3 Innovation ignored

Opportunities Threats 1 Qualified vocational training centers 1 Many production units have been set up Source: Authors from brainstorming workshop with MPI staff

Table 3: SWOT of organization Strengths Weaknesses

1 Open-mindedness 1 Non participatory organisation 2 Non flexible organisation 3 Innovation ignored 4 Quality process ignored

Opportunities Threats 1 Collaboration with the University of Antananarivo 1 Competitors from Southern Africa and Asia Source: Authors from brainstorming workshop with MPI staff

There are seven weaknesses and three strengths. Two opportunities face two threats. c. Understanding of the current process

Paper product quality of the Paper Transformation Unit before Reengineering

Source: Authors from Production Department’s data

Figure 4: Basis Mass profile

Basis mass is one of Product Quality’s characteristics. Basis Mass is observed in a chart line: abscissa is “rank of measure” and ordinate is “basis mass”. Measures have been charted through eleven equidistant points. Standard basis mass value is 80 (g/m2). Maximum and minimum control limits are respectively (92g/m2) and (67g/m2). MPI’s laboratory technicians fix the latter values. They take into account limits permitted by a standard step. Three basis masses are included in the control limits. Eight points are over the maximum control limit. During the first through the third measures and during the seventh through the eleventh measures, basis mass lines have sinusoidal profile. From the third to the seventh measure, they are constant.

0

20

40

60

80

100

120

1 2 3 4 5 6 7 8 9 10 11

Rank of measure

Bas

is m

ass

(g/m

2)

Basis Mass Standard Maximum Control Limit Minimum Control Limit

33

The basis mass profile is generally poor. Papers produced are heavier than maximum control limit. Technology for such control is out of service. Therefore, automatic control of basis mass is not effective.


Figure 5: Moisture profile

Moisture is the second of Product Quality’s characteristics. It is observed in a chart line: abscissa is “rank of measure” and ordinate is “moisture”. Measures have been collected through eleven equidistant points along paper. Standard moisture value is 5%, maximum and minimum control limits are respectively 5.25% and 4.75%. MPI’s laboratory technicians fix the latter values. They take into account limits permitted by a standard step. Two moistures fall within the control limits and nine data are under the minimum control limit. During the first through the fourth measures and during the eighth through the eleventh measures, moisture lines are constant. From the fourth to the eighth measures, they have a sinus profile. Because electronic equipment is broken down and not functional anymore, papers produced are too dry and therefore of poor quality.

Speed of delivery before Reengineering

Table 4: Speed of delivery comparison PAPER ENTERPRISE NUMBER OF DAYS TO DELIVER 1000 TONS OF PAPER

MPI 85 MONDI 2 SAPPI 2 APP 1

Source: Authors

This table shows the speed of delivery profile for each paper industry. The speed of delivery unit is the number of days needed to supply 1000 tons of paper. There are four paper industries and four speeds of delivery. To deliver 1000 tons of paper products, on Malagasy market, MPI needs 85 days. The same quantity is delivered in two days by MONDI and SAPPI. APP delivered it in one day. Only 16% of MPI’s capacity is utilized [20]. That underutilized capacity is often broken down [20].

4

4.2

4.4

4.6

4.8

5

5.2

5.4

1 2 3 4 5 6 7 8 9 10 11

Rank of measure

Moi

stur

e (%

)

Moisture Standard Maximum Control Limit Minimum Control Limit

34

Flexibility through varieties of products on Malagasy market before Reengineering

Source: Authors Figure 6: Number of varieties of value added paper products sold on Malagasy market

The number of varieties of value added paper products is presented in a chart line. Months are in abscissa. Four paper industries are concerned: APP, SAPPI, MONDI and MPI. On October 2006, MPI had four varieties, MONDI three varieties, APP and SAPPI two varieties. However, on November 2006, MPI had closed. And it stopped producing and selling value added paper products. In March 2007, the three competitors still had the same varieties as in the beginning of the study. MPI was still not producing.

Dependability through paper reel waiting time before Reengineering

Source: Authors Figure 7: Paper reel waiting time

Paper reel waiting time is input waiting time. PTU inputs are PM outputs which are paper reels. Waiting time is being observed from July 05 through August 06. There are 13 measures. Minimum value is 0 hour and maximum is 350 hours. A continuous trendline is shown. Mean values are 100 to 150 hours. Waiting time is constant during the first four months. From December, it increases. Paper reel waiting time is due to technological problems of PM. That problem is transferred to PTU.

0

50

100

150

200

250

300

350

400

Jul-05 Sep-05 Oct-05 Dec-05 Feb-06 Mar-06 May-06 Jul-06 Aug-06 Oct-06

Month

Hou

r

Paper Reel Waiting Time

0

1

2

3

4

5

Oct "06 Nov "06 Dec "06 Jan "07 Feb "07 Mar "07

Month

Num

ber

of V

arie

ties

of V

alue

A

dded

Pap

er P

rodu

cts

sold

on

Mal

agas

y M

arke

t

APP SAPPI MONDI MPI

35

Creation of a new process design

TPM: Total Productive Maintenance JIT : Just In Time

Source: Authors by benchmarking and inductive thinking

Figure 8: New process design

PTU will be definitely separated from the actual PM. They will become two independent production units. PM can keep on with its old parameters. It continues importing old papers and collecting waste cardboard. PM may maintain its technological situation: pulp waste, breakdown waiting time and pulp waiting time. The status quo does not create any consequence to PTU’s inputs. Inputs are standard imported paper reels. With two soft technologies, Total Productive Maintenance and Just In Time, dramatic improvements have been achieved in quality, speed of delivery, flexibility and dependability. Product quality and flexibility are the result of imported paper reels with soft technology based on JIT. Speed of delivery and dependability has been obtained from implementation of TPM. Implementing the reengineered process

Table 4: Action plan ACTIONS MONTH1 MONTH2 MONTH3

Building team and assign a senior executive to be responsible Creating an importing department Training all PTU’s employees about JIT and TPM Training all importing department’s employees about JIT Starting the reengineered process ►

Source: Authors from MPI staff document

TPM JIT

PAPER TRANSFORMATION

UNIT (PTU)

Importing paper reels respecting

international standards

Quality

Speed of delivery

Flexibility

Dependability

PULP PAPER MACHINE (PM)

Pul

p w

aste

Pul

p w

aitin

g tim

e

Bre

akdo

wn

wai

ting

time

36

An action plan based on team building and assigning a senior executive to be responsible for implementation has been conducted. Once conducted, TPM training has been organized for PTU’s employees. Both the latter and the importing department’s ones have been trained about JIT. Effective reengineered process implementation has begun in the third month.

III. RESULTS AFTER REENGINEERING Basis Mass profile

Source: Authors from Production Department’s data Figure 9: Basis mass profile after Reengineering

Moisture profile


Figure 10: Moisture profile after Reengineering

Speed of delivery

Table 5: Number of days to deliver 100 tons of A4 c opy paper

BEFORE REENGINEERING AFTER REENGINEERING

8 3


0102030405060708090

100

1 2 3 4 5 6 7 8 9 10 11

Rank of measure

Bas

is m

ass

(g/m

2)

Basis Mass Standard Maximum Control Limit Minimum Control Limit

4.5

4.6

4.7

4.8

4.9

5

5.1

5.2

5.3

1 2 3 4 5 6 7 8 9 10 11

Rank of measure

Moi

stur

e (%

)

Moisture Standard Maximum Control Limit Minimum Control Limit

37

Dependability

Source: Authors from Production Department’s data Figure 11: Paper reel waiting time after Reengineer ing

Studies focused on A4 copy paper have been conducted. After Reengineering, A4 copy paper meets ISO 536 specifications for basis mass and ISO 534 for moisture. Speed of delivery becomes twice rapid than before. And waiting time becomes zero.

IV. DISCUSSION Reengineering can sustain the Malagasy economy through restarting and sustaining Malagasy Paper Industry’s performance and competitiveness.

Process for Reengineering PTU is currently problematic. Its production is disturbed by maintenance failure, lack of production variety, slow speed of delivery and poor quality. PM transfers those technological problems to PTU and drops the latter’s performance. This situation does not permit either maximization of outputs or minimization of poor quality based on existing resources [1]. These facts do not show basic operations strategy [21] such as quality, flexibility and speed of delivery. So, the actual process disadvantages competitiveness for PTU and, of course, of MPI [19] in a globalizing competition. Continuous process will be problematic if technology management is not carried out at any of the levels of that process. In the present case, PTU follows PM. Thus, lacks in adequate technology management, all technological problems from PM are transferred to PTU.

Enablers of Reengineering Employees are determinant enablers. Their social cases interfere with the approach of productivity [24]. The special conditions required by qualified employees are not satisfied which generates a drop in productivity [24]. Therefore, translation of weakness into strength is difficult. As well, foreign companies setting up in Madagascar may be a valve of security to workers who have been enduring a problematic work situation. Organization includes some risks concerning quality management process or concurrent engineering concepts [6]. Some key points of weaknesses may be converted into strengths.

0

0.2

0.4

0.6

0.8

1

1.2

1.4

1.6

1.8

2

Oct-07 Nov-07 Dec-07 Jan-08 Feb-08

Month

Hou

r

38

Collaboration with universities will facilitate that translation through training upgrades. Southern African and Asian pressures are really important and may considerably affect Malagasy economy in general and Malagasy paper industry in particular.

The current process Paper products produced by MPI are too heavy and too dry. First, they waste fiber; and secondly, they waste energy. They are of poor quality [13]. The cost of fiber waste was estimated at $53,000.00 in 2005 [18]. The second needs $1,000,000.00 to purchase a steam accumulator [20]. There is no equipment for controlling the superheating steam production. At the PM level, pulp supply creates a considerable delay which costs $ 98,600.00 [18]. That delay has a consequence on the speed of delivery coordination [22]. A Skinner’s metric is disfavored. By extension, its effect on the speed of delivery does not permit a decrease of the output cost [11] which does not favor to competitiveness. Technology failure has generated a very important financial loss for MPI. That case, if benchmarked to other industries, will entail national financial loss.

The new process design Two independent units have been designed. PTU becomes autonomous. Its inputs are standardized and imported either from South Africa or Asia. The first is economically interesting, value chain analysis shows that importing from South Africa is cheaper than importing from Asia [20]. Compared with final cost, the first is 11% to 13% and the second is 20% [20]. With such option, PTU has no more paper reel supply problem. Just In Time has been implemented to assure quality management so that there is quality at the source [1], in order to uniform plant loading [24], to minimize of setup times [25] and to respect people [13]. Carrying out that soft technology allows for good quality and quantity on time. Total Productive Maintenance has been implemented to assure continuous production while operating maintenance. Speed of delivery has been considerably improved, as well as flexibility [1]. Better outputs are obtained after dramatic improvement in quality, speed of delivery, flexibility, and dependability. These outputs have balanced the foreign competitors’ specifications and meet the local market’s requirements. Soft technology, such as JIT and TPM, can accompany developing countries’ industries to balance their competitors on the local market.

Reengineering can sustain paper industry developing country Reengineering has re established and has upgraded the competitiveness for the paper industry in a developing country. Quality is aligning with international standards. It meets customers’ requirements and technologies’ necessities. The number of those customers is increased. The local market is mastered. Speed of delivery has been also improved. A developing country paper industry can quickly deliver products on the local market. That dramatic improvement increases the number of customers.

39

Now paper industry in a developing country imports paper reels for inputs. Therefore, it is able to offer a wider variety of products to its customers [1]. Paper industry is more and more dependable. TPM is also implemented and regularly followed. Reengineering has created an advantage of US$ 1,768,000.00 per year for a paper mill industry [20].

Industry sustains economy Industry is an important sector of economy. According to Robert Reich, industry shapes economy. It made British economy thriving. And it has fashioned Asia and South East Asian economies and has caused their countries to be ranked as new industrialized countries. The case of MPI is a current developing country’s case. MPI has 500 employees which supplies 80% of Malagasy customers. These are important data in the case of a developing country. Thus, the absence of MPI on the Malagasy market would entail a negative impact on economy. MPI has been reengineered. Not only MPI but also other national industries operating in Madagascar have to be reengineered to face world competition. Examples are: biscuit industries, soap factories, and the airline company. If they are reengineered, they will considerably sustain Malagasy economy.

CONCLUSION The present survey concerns a paper mill industry in a developing country. Problem scope is regarding two processes: Paper Machine and Paper Transformation Unit. The latter’s output face four problems relating to quality, speed of delivery, flexibility and dependability. So, Paper Transformation Unit is identified for reengineering application. Assessment of enablers of that reengineering shows many weaknesses for human resources and organization. The Paper Transformation Unit’s product quality is out of standard. It has a limited variety of paper product that it delivers forty times slower than its competitors on the local market. Most of time, it spends time waiting for paper reels as inputs. Dramatic redesign has been achieved and conducted. Product quality changes and becomes standard. Speed of delivery increases and gains five points than before reengineering. Inputs supply becomes just in time. So, reengineering has restarted and sustained paper mill industry competitiveness.

But what have we learned? Although reengineering succeeds in the above case, this paper has exposed four lessons relating to the conditions in which reengineering can restart and sustain a paper mill industry in a developing country.

1. Implement reengineering in a small scope, for example in two consecutive processes, so that only small problems appear. Generally “small scope” has “small problems”. And small problems are a bit easy to solve. Reengineering involves radical change where many difficult problems may appear. Sometimes those difficult problems ask top level solutions which are not easy to find. Focus on two consecutive processes, as developed in the present survey, is indicated to simplify problems and to get maximum result.

40

2. Shape reengineering to the parameters of industry in a developing country Many specific parameters need to be taken into account. Employees and current organization are important. All reengineering actions have to consider them.

3. Structure reengineering by Skinner’s metrics and Real Time System. Skinner’s metrics are quality, cost, speed of delivery and flexibility. And one of Real Time System’s parameters is dependability. Organize reengineering around those allows us showing competitiveness and performance to all stakeholders. That facilitates radical change culture driven.

4. Success in a small scope can be benchmark to another small scope in the same system. For developing country, operating reengineering in large scale may be hard to master. The first objective is “to have built a small critical mass of believers”, according to Caron R.. Break system into sub systems and then operate reengineering step by step, sub system by sub system, are an indicated way.

If Reengineering can sustain the economy of a developing country, how could we generalize such methodology so that developing countries’ industries can insert in globalizing competition?

REFERENCES [1] Aquilano N., Chase R.: “ Production and Operations Management “,7th edition,

Irwin, Chicago, 1996, 821 pages

[2] Art B.: ‘World Class Maintenance”, TAPPI and PIMA Solutions, September 2001, Premier issue, pp56-59

[3] Berkeley B., Gupta A.: “Improving Service Quality with Information Technology”, Working Paper 9-93-9, University of Wisconsin, Madison, 1993

[4] Burt D., Doyle M.: “The American Keiretsu: A Strategic Weapon for Global Competitiveness”, Homewood IL, Business One Irwin, 1993

[5] Davenport T.: “Process Innovation: Reengineering Business Processes through Information Technology”, Harvard Business School Press, Cambridge MA, 1993

[6] Flaig S.: “Integrative Manufacturing: Transforming the Organization through People, Process, and Technology”, Homewood IL, Business One Irwin, 1993

[7] Gunn T.: ”Manufacturing for Competitive Advantage”, Cambridge MA, Ballinger, 1987

[8] Hall G., Rosenthal J., Wade J.: “How to Make Reengineering Really Work”, Harvard Business Review 93, n°6 (November-December 1993), pp 119-131

[9] Hammer M., Champy J.: “Reengineering the Corporation : A Manifesto for Business Revolution”, New York: Harper Business, 1993

[10] Hammer M.: “Reengineer Work: Don’t Automate, Obliterate”, Harvard Business Review 90, n°4, Jul-Aug 1990, pp 104-12

41

[11] Kaplinski R.: “The Implication of New Organizational Techniques for Developing Countries”, UNIDO, October 1995, 36 pages

[12] Kauppi K., Ylinen P.: “Make Maintenance Meaningful”, Pulp and Paper International, February 2005, vol 47, n°2, pp 17-18

[13] Kélada J.: “Qualité Totale: Amélioration Continue et Réingénierie”, Edition Quafec, Québec, 2000, 480 pages

[14] Koepenick M. : « The Promise of Paper in China », TAPPI and PIMA Solutions, September 2001, Premier issue, pp 48-51

[15] Miles L.: “Techniques of Value Analysis and Engineering”, MacGraw-Hill Inc., 1972

[16] Nunnenkamp P., Gundlach E.: “Globalization of Manufacturing Activity: Evidence and Implications for Industrialization in Developing Country”, UNIDO, Global Forum on Industry, October 1995

[17] Oakland D.: ”Steam Accumulator”, Pulp and Paper International, December 2004, Volume 46, n°12, p14-17

[18] Ravalison F. : « Contribution à l’Amélioration du Système de Management par la Gestion de Production-Le cas des Papeteries de Madagascar Océan Indien », Mémoire de DEA, Université d’Antananarivo-Ecole Supérieure Polytechnique, 2006, n°009, 63 pages

[19] Ravalison F., Rajaonary P., Raveloson E., Rakotomaria E.:”Are Converging Technologies Tools of Competitiveness?”, PICMET, Aug 6-14, 2007, Portland-USA, Portland State University, 22 pages

[20] Ravalison F. : « Redynamisation de la Compétitivité, dans le Contexte Mondialisé, par la Démarche Intégrée d’Ingénierie Industrielle-Le cas d’une Industrie Papetière Malgache », Thèse de Doctorat Nouveau Régime, Université d’Antananarivo-Ecole Supérieure Polytechnique, 2008, 267 pages

[21] Skinner W.: “ Manufacturing-The Missing Link in Corporate Strategy”, Harvard Business review 47 , n°3 (May-June 1969), pp.120-24

[22] Steinmueller E., Bastos M.: “ Information and Communication Technologies : Growth, Competitiveness and Policy for Developing Nations”, UNIDO, October 2005, 34 pages

[23] Suzaki K.: “The New Manufacturing Challenge: Techniques for Continuous Improvement”, New York, Free Press, pp 7-25, 1987

[24] Wantuck K.: “The Japanese Approach to Productivity”, Southfield MI, Bendix Corporation, 1983

[25] White R.: “An Empirical Assessment of JIT in US Manufacturers”, Production and Inventory Management Journal 34, n°2 (second quarte r 1993), pp 38-42

42

Developing a mathematical concept and process technologies to

accompany firms from rut to change

Ravalison, F.; Rahoelison, T.; Raveloson, E.; Rakotomaria, E.; Ravalison, R.;

Electr. Eng. Dept., Univ. of Antananarivo, Antananarivo, Madagascar

This paper appears in: Management of Engineering & Technology, 2009. PICMET 2009. Portland

International Conference on Issue Date: 2-6 Aug. 2009

On page(s): 786 - 792 Location: Portland, OR Print ISBN: 978-1-890843-20-5


Digital Object Identifier: 10.1109/PICMET.2009.5262072 Date of Current Version: 25 September 2009

ABSTRACT

Many companies, getting into a rut, think that change is crucial and inevitable to improve

competitiveness in the actual globalization era. This paper proposes a model from developing a mathematical concept and process technologies to accompany companies from rut to change. For that, an integrated process has been adopted. By using function analysis system

technique (FAST), daily activities have been mapped. And by utilizing equivalence relation, each function of FAST diagram has been associated with its equivalent function in a

technology. At the end of that process, a technology transform of daily activities has been

obtained. Then the same process is implemented in a case of an international cell phone enterprise based in Madagascar. Analyzing the framework of its daily activities, a process of engineering has been identified. It sequences competitive intelligence (CI) and theory of

inventive problem solving (TIPS). The paper discusses that (CI) and (TIPS) are innovation

technologies whose targets are the improvement and upgrading of competitiveness. The results of this research find that rut can be changed. The conclusion is that the above model

integrates characteristics of knowledge management and innovation process.

INDEX TERMS




43

DEVELOPING A MATHEMATICAL CONCEPT AND PROCESS TECHNOLOGIES TO ACCOMPANY FIRMS FROM RUT TO

CHANGE François Ravalison1, Toky Rahoelison1, Elisé Raveloson1, Etienne Rakotomaria1, Rijamalala

Ravalison2 1University of Antananarivo, Ecole Supérieure Polytechnique, Electrical Engineering Department,

Antananarivo-Madagascar 2Institut Supérieur de Technologie d’Antananarivo-Madagascar

ABSTRACT

Many companies, getting into a rut, think that chan ge is crucial and inevitable to improve competitiveness in the actual globalization era. This paper proposes a model from developing a mathematical concept and pr ocess technologies to accompany companies from rut to change. For that, a n integrated process has been adopted. By using Function Analysis System Tec hnique (FAST), daily activities have been mapped. And by utilizing Equiv alence Relation, each function of FAST diagram has been associated with its equiva lent function in a technology. At the end of that process, a technology transform of daily activities has been obtained. Then the same process is implemented in a case of an international cell phone enterprise based in Madagascar. Analyzing the framework of its daily activities, a process of engineering has been ident ified. It sequences Competitive Intelligence (CI) and Theory of Inventive Problem S olving (TIPS). The paper discusses that (CI) and (TIPS) are innovation techn ologies whose targets are the improvement and upgrading of competitiveness. The r esults of this research find that rut can be changed. The conclusion is that the above model integrates characteristics of knowledge management and innovat ion process. I- INTRODUCTION

Like many developing countries, Madagascar has to face globalization. Unfortunately, no serious preparation has been done. For example, enterprises were not accompanied to that new era of world movement. And they are becoming vulnerable to competitiveness. Mobile telephony enterprises do not make an exception. They have shaped a concept at the beginning and have not updated nor upgraded it anymore, in spite of the coming competition. They are getting into a rut.

In the rapidly changing global competition, how could technology go together with enterprise to convert rut to change? II- LITERATURE REVIEW

a. Definition of “rut” Rut is a settled and monotonous process that is hard to escape. In general, at the beginning of any activities, enterprises design and implement a process. While implementing such process, it unconsciously becomes a habit. Consequently, enterprises get into rut.

b. Function Analysis Systems Technique

Prior Function Analysis Systems Technique (FAST) explanation, Value Analysis (VA) or Value Engineering (VE), developed by Lawrence Miles [7], needs to be enlightened. It

44

is a concept of function analysis. The purpose of VA/VE is to simplify products and processes [1]. Its objective is to achieve equivalent or better performance at a lower cost while maintaining all functional requirements defined by the customer [1]. That function analysis was further developed and extended by Charles Bytheway [12]. That survey had introduced the methodology called Function Analysis Systems Technique [12].

FAST is used to decompose a basic function and organize it into a logic diagram called a FAST model [12]. The FAST modeling process leads to a basic structure which is somewhat similar to a process flow chart. This means each element of decomposition is an independent entity. And the sum of entities is a system. The following figure shows the FAST method.

Figure 1: Basic FAST Model c. Equivalence Relation [11] The concept of Equivalence Relation is an abstraction of the idea of two

mathematic objects being like each other in some respect. If an object “a” is like an object “b” in some specific way, then “b” is like “a” in that respect. a b

Figure 2: Equivalence Relation between two object There is a fundamental fact connecting “Equivalence Relation” and “Partitions”. A

Partition is a set of blocks. Every Equivalence Relation induces Partition and vice versa. If an Equivalence Relation R is given on a set S, the elements of S can be collected together into subsets or “blocks”.

S blocks

Figure 3: Blocks Those blocks will be identified as group technology in the following survey.

HOW WHY

SCOPE OF THE PROBLEM UNDER STUDY

Basic function

Independent function

Acceptable interface function

Higher order function

Lower order function

entity

SYSTEM

45

III- METHODOLOGY a. Sample size In the survey, the case of Besalampy, a District so a second level town, is considered.

Its case is the same as the other second level towns. Then, the sum of all second level towns equals to the case of Madagascar. So Besalampy represents a significant case.

The field of application is the market of mobile telephony in that District in Sofia Region in the northwest part of Madagascar. An enterprise offers three main services: communication, short message service or SMS and Internet. The period of study goes from week 7 to week 32, so twenty six weeks during 2008. The customers comprise the civil servants, the small companies’ agents and the residents.

b. Data collection All data concerning mobile telephony network are weekly and automatically collected

by a computer-integrated machine. Such method gives global data. In the present study, Besalampy’s data are extracted from global data during 2008. The following process shows such mechanism.

Figure 4: Data collection process

c. Research method The engineering methodology FAST is used to decompose a basic function related to

basic activity. That permits to get the FAST model. By a mathematical concept “Equivalence Relation”, technology function equivalent to each function is identified. Such process allows getting a sequence of technology functions. And then, technology is identified from grouping and completing technology functions. The following chart shows the brief process of that research.

Figure 5: Main Process of Research Method

That methodology was conducted in a case of a mobile telephony enterprise. That case illustrated a sample in mobile telephony activities in Madagascar.

Global data 2008

Besalampy’s data 2008

To transform weekly activities into FAST functions

To research equivalence of each function to technology

To group technology function in order to get a partition

To complete each partition in order to get a technology

IDEAL MODEL

46

IV- RESULTS a. Sector and Skinner’s metrics

Sectors, which the mobile phone operator keeps under surveillance, are defined by International Telecommunication Union (ITU) as Statistical Quality Indicator (SQI). They are: Bottleneck, Call Drop, Call Set Up Success, Number of Customers and Rate of Charge. Bottleneck means that all circuits are not available. Call Drop permits to measure the ratio of numbers of interrupted calls to numbers of call set up success. Call Set Up Success rate is the ratio of numbers of call set up success to numbers of call set up. And the Rate of Charge is the ratio of observed traffics to available traffics.

Two sets are considered: sectors and Skinner’s metrics. Then the equivalence of each sector to the four basic operations strategies of Skinner [10] has been established. The following diagram is obtained.

Figure 6: Equivalence Relation Diagram of Sectors u nder Surveillance The equivalence of “bottleneck” and “call set up success” is “speed of delivery”. “Call

drop” and “rate of charge” are respectively liken to “quality” and “flexibility”. Now all sectors under surveillance have their equivalent to Skinner’s metrics. Then, critical zone corresponding to those sectors will be defined. b. Critical Success Factor (CSF)

For CSF, enterprise has benchmarked its experience in Africa. So, CSFs utilized in Africa have been transposed into the Malagasy case. For example in Africa, acceptable bottleneck has to be fewer than 2%. And it is adopted for Madagascar. The following table shows those CSFs.

Table 1: Critical Success Factor (CSF)

SECTORS EQUIVALENCE TO SKINNER’S METRICS CRITICAL SUCCESS FACTOR

Bottleneck Speed of Delivery Acceptable Bottleneck <2%

Call Drop Quality Acceptable Call Drop <0.7%

Call Set Up Success Speed of Delivery Acceptable Call Set Up Success �97% Rate of Charge Flexibility Acceptable Rate of Charge <110%

On the first hand, there are three upper limits: 2% for bottleneck, 0.7% for call drop and 110% for rate of charge. And on the second hand, there is one lower control limit of 97% for call set up success.

Bottleneck

Call Drop

Call Set Up Success

Number of Customers

Cost

Quality

Speed of Delivery

Flexibility

SKINNER Sectors

Rate of Charge

47

c. Sectors profiles A zone, Besalampy, has been under surveillance. The four sectors above and the

number of customers are observed every week. A group of persons has assured collecting information, detecting problems and solving problems. Numerical data are obtained weekly and transformed into control charts. Five figures of life cycle product are obtained.

The figure 7 below shows bottleneck variations with 2% for upper control limit. Ten cases of bottleneck are over the upper control limit. Two partitions could be considered, the first is from week 7 to week 13 and the second one is from week 17 to week 27. The first partition contains more points over upper control limit than the second one does. Especially in the second partition, a significant decrease is noticed, even over the second partition itself.

Figure 7: Bottleneck variations in Besalampy

The figure 8 beneath illustrates call drop variations. All cases of Call Drop are over the upper control limit of 0.7%. The trendline of call drop from week 7 to week 17 is relatively flat. But from week 19 to week 28 point values are far from the upper control limit.

Figure 8: Call Drop variations in Besalampy

The figure 9 gives an idea about the call set up success rate variations. There are 13 cases under the lower control limit at the beginning of the enterprise’s activities. Call set up success rate becomes flat from week 20 while closing to 97%.

0

5

10

15

20

25

30

35

40

week 7

week 9

week 1

1

week 1

3

week 1

5

week 1

7

week 1

9

week 2

1

week 2

3

week 2

5

week 2

7

week 2

9

week 3

1

Bottleneck Maximum Limit

0

1

2

3

4

5

6

week 7

week 9

week 1

1

week 1

3

week 1

5

week 1

7

week 1

9

week 2

1

week 2

3

week 2

5

week 2

7

week 2

9

week 3

1

Call Drop Maximum Limit

48

Figure 9: Call Set Up Success Rate variations in Be salampy

The figure 10 is about rate of charge variations. 13 rates of charge variations are over the upper control limit. At the beginning, from week 9 to week 18, all rates of charge are over 110%. Then after they continuously decrease.

Figure 10: Rate of Charge variations in Besalampy

And the last figure represents evolution of the number of customers in Besalampy. That number is increasing regularly from week 7 to week 26. From week 26, it decreases and becomes flat.

Figure 11: Evolution of the number of customers in Besalampy Only the rate of charge complies with CSF from week 7 to week 9. Then, it deviates

from that norm the following weeks, while all other Skinner’s metrics remain nonstandard.

0

20

40

60

80

100

120

week 7

week 9

week 1

1

week 1

3

week 1

5

week 1

7

week 1

9

week 2

1

week 2

3

week 2

5

week 2

7

week 2

9

week 3

1

Call Set Up Success Rate Minimum Limit

0

50

100

150

200

250

300

week 7

week 9

week 1

1

week 1

3

week 1

5

week 1

7

week 1

9

week 2

1

week 2

3

week 2

5

week 2

7

week 2

9

week 3

1

Rate of Charge Maximum Limit

0

200

400

600

800

1000

1200

1400

1600

week 7

week 9

week 1

1

week 1

3

week 1

5

week 1

7

week 1

9

week 2

1

week 2

3

week 2

5

week 2

7

week 2

9

week 3

1

Customers

49

From week 13 to week 18, the number of customers increases although three metrics, as call drop, call set up success rate and rate of charge are nonstandard. Only bottleneck complies with standard.

From week 18 to week 22, the number of customers does not change. The Call set up success rate and the rate of charge nearly comply with standard. Both call drop and bottleneck remain nonstandard. For the next couple of weeks, the call set up success rate and the rate of charge remain nearly comply with standard. In that period, bottleneck fulfill standard.

From week 25 to week 27, the number of customers does not change; the call set up success rate and the rate of charge continue to be nearly with standard. Then after week 27, the number of customers decreases. V- DISCUSSION

Despite the above sectors’ profiles, some findings need to be mentioned: 1. The mobile telephony enterprise does not comply with CSFs settled by itself.

So many points are not in the critical zones corresponding to CSF. Skinner’s metrics as quality (call drop), speed of delivery (bottleneck, call set up success) and flexibility (rate of charge) do not comply with the standards defined by the CSF. Thus customers’ needs are not understood and unsatisfied.

2. Internal technical problems are solved through a matrix of specific technical problems and corresponding specific technical solutions. But there are some exogenous problems that are not solved in time and have an impact on activities. For example, accessibility constraints due to cyclone or slight disruption to domestic flights are problems not included in the above matrix. They prevent managers and staff from solving internal technical problems.

3. The above practices become routine and do not involve success as strongly supported by the fig.8, particularly from week 27. The weight of evidence suggests that accompanying enterprise from rut to change needs technology adoption. Briefly, aligning activities to a known technology is suggested.

Proposed model In order to become competitive, the mobile telephony enterprise’s activities are

collecting data, detecting problems and solving problems. We consider “solving problem” as a basic function and we decompose it by asking a question as “how does the enterprise solve the problem?”. The answer is placed at the right side of the basic function and is “by detecting problem”, and so on. The corresponding FAST model is presented.

“To solve problem” is the basic function and “to collect information” is an acceptable interface function. In order to transform each function of that model into technology function, let’s define an equivalence relation R as “…is equivalent to…”. And let’s define a function f so that xR y implies f(x)=y, x is element of FAST model and y of technology function one. Then let’s apply f for each element of the FAST model:

x1 =(to solve problems) �(FAST model)

To detect problems To collect information To solve problems

50

x2 =(to detect problems) �(FAST model) x3 =(to collect information) �(FAST model) because R is reflexive so f(x)=x, we apply that definition in the case of xi. f(to solve problems) =(to solve problems). f(to detect problems) =(to detect problems). f(to collect information) =(to collect information)

The question is “what is a technology containing (to collect information), (to detect problems) and (to solve problems)?”. (to collect information) is equivalent to “to collect information”. (to detect problems) (to solve problems)

The equivalence of “to collect information” through “to treat problems” is near the

Competitive Intelligence (CI) but need to be completed. Upstream from “to collect information”, “to research information” is needed. After “to collect information”, “to disseminate information” is required so that information could be treated. And the equivalence of “to treat problems” is near the Theory of Inventive Problem Solving (TIPS).

Figure 12: Technology outlines

Definitely, the two outlined technologies are sequenced and overlapped at “detect problems” level. Each technology has been completed relating to technologies defined by scholars and writers. That step permits to obtain the following model.

Figure 13: Proposed model to accompany enterprise f rom rut to change

To detect problems To collect information To solve problems

To collect information To search information

Detect problems Solve problems

To treat problems

EQUIVALENCE RELATION

THEORY OF INVENTIVE PROBLEM SOLVING

COMPETITIVE INTELLIGENCE

To collect information To search information To disseminate information

Analyze information Decide

COMPETITIVE INTELLIGENCE TIPS

is equivalent to “to treat problems that are from information”

51

First, that model, based on CI, is fundamental for the success of business [2] [3] [4] [6] [9], an absolute imperative for business [5], and important for profitable and sustained growth [8]. So the utility of that model is the competitive advantage.

Second, TIPS permits, to get an ideal solution above all for problems over the matrix. The usefulness of that model for problem solvers is to analyze information by detecting and solving problems. VI- CONCLUSION

The objective of this study is to provide evidence that competitiveness depends on bringing back to technology. And that needs to follow a process. The present survey shows that daily activities can be brought back to system technologies through FAST process and Equivalence Relation. A case study of a mobile telephony enterprise has been undertaken. After implementing FAST process and Equivalence Relation, Theory of Inventive Problem Solving included into Competitive Intelligence appears as a model to face global competition. Such model permits the enterprise to implement its activities according to a system engineering corresponding to prescribed activities and to update or upgrade its standards through “to research information” activities. This avoids into rut.

The results of this study could be generalized to other mobile phone enterprises as far as they are in a same group of enterprises. Meanwhile their implementation to other enterprises needs to be handled carefully.

REFERENCES [1] Aquilano N., Chase R.; Production and Operations Management, United States

of America, Irwin, 1995

[2] Flynn R.; “NutraSweet faces competition: the critical role of competitive intelligence”, Competitive Intelligence Review, vol. 7, n°1, pp.25-28, 1996

[3] Hart S., Tzokas N., Saren M.; “The effectiveness of market information in enhancing new product success rate”, European Journal of Innovation Management, vol. 2, n°1, pp.20-35, 1999

[4] Herring J.P.; “Key intelligence topics: a process to identify and define intelligence needs”, Competitive Intelligence Review, vol. 10, n°2, pp.4-14, 1999

[5] Kahaner L.; Competitive Intelligence, SandS, NY, 1997

[6] Lackman .L., Saban K., Lanasa J.M.; “Organizing the competitive intelligence function: a benchmarking study”, Competitive Intelligence Review, vol. 11, n°1, pp.17-27, 2000

[7] Miles L.; Techniques of Value Analysis and Engineering”, Canada, MacGraw-Hill, 1972

[8] Prescott J.E., Miller S.H.; Proven Strategies in Competitive Intelligence: Lessons from Trenches”, SCIP/Wiley, NY, 2001

[9] Shaker S.M., Gembiki M.P.; The WarRoom Guide to Competitive Intelligence, McGraw-Hill, NY, 1998

[10] Skinner W. C.; “Manufacturing-The Missing Link in Corporate Strategy,” Harvard Business Review, vol. 47, n°3, pp.136-45, 1969

[11] Wells C., “Abstract Mathematics”, retrieved from http://www.abstractmath.org on December 31, 2008

52

[12] Wixson J., “Function Analysis and Decomposition using Function Analysis Systems Technique”, retrieved from http://www.inl.gov/technicalpublications/Documents/3314433.pdf, on July 21st, 2008

53

Towards “Fair Globalization”: Critical Success Factors for

partnering Project of Development and Enterprise Project

Ravalison, F.A.; Raveloson, E.A.; Rakotomaria, E.;

Ecole Super. Polytech., Univ. of Antananarivo, Antananarivo, Madagascar

This paper appears in: Technology Management for Global Economic Growth (PICMET), 2010 Proceedings of PICMET '10:

Issue Date: 18-22 July 2010

On page(s): 1 - 7

Location: Phuket Print ISBN: 978-1-4244-8203-0


Date of Current Version: 14 October 2010

ABSTRACT

The purpose of this paper is to identify and evaluate Critical Success Factors (CSF) that

impact positively on partnering Project of Development (PoD) and Enterprise Project (EP).

The data are collected from an International Labor Office (ILO)'s pilot project implemented in Madagascar which had utilized the Training for Rural Economic Empowerment (TREE)

methodology. Then other data are gathered from a questionnaire survey of 34 enterprises, five months after completion of the project. And Quality Assurance coupled with Correlation is conducted to analyze data. Results reveal that nine Critical Success Factors shape and

impact the partnership of (PoD) and (EP). Then a “Golden Triangle” is identified to model

partnering success. This paper provides a valuable guideline for policy makers to align their economic growth approach with pro-poor growth. This paper shows the importance of

Technology Management in contributing to implement a Fair Globalization. It will help many Projects of Development to integrate their beneficiaries into such Fair Globalization.

Index Terms


References Available to subscribers and IEEE members.

Citing Documents Available to subscribers and IEEE members.

54

Towards “Fair Globalization”: Critical Success Fact ors for Partnering Project of Development and Enterprise Pr oject

François A. Ravalison, Elisé A. Raveloson, Etienne Rakotomaria University of Antananarivo-Ecole Supérieure Polytechnique, Madagascar

ABSTRACT

The purpose of this paper is to identify and evalua te Critical Success Factors (CSF) that impact positively on partnering Project of Development (PoD) and Enterprise Project (EP).

The data are collected from an International Labor Office (ILO)’s pilot project implemented in Madagascar which had utilized the Tr aining for Rural Economic Empowerment (TREE) methodology. Then other data are gathered from a questionnaire survey of 34 enterprises, five months after completion of the project. And Quality Assurance coupled with Correlation is c onducted to analyze data.

Results reveal that nine Critical Success Factors s hape and impact the partnership of (PoD) and (EP). Then a “Golden Trian gle” is identified to model partnering success. This paper provides a valuable guideline for policy makers to align their economic growth approach with pro-poor growth.

This paper shows the importance of Technology Manag ement in contributing to implement a Fair Globalization. It will help many P rojects of Development to integrate their beneficiaries into such Fair Globalization.

I. INTRODUCTION

Africa in general and Madagascar in particular call for Foreigner Direct Investment (FDI) to boost their economy. All their development schemes are FDI driven. Once operating in field, those investors are embarking on advertisement. Their implication in local economic development is limited to donation. So, vulnerable families and poor women are rarely targeted in FDI process.

Partnering with FDI is mentioned, by policy makers, to calm rural population down. Because they think that there are not positive effects on employment. Partnering, analyzed in this study, is an informal process between a (FDI) and a (PoD) to cooperate to reach own goals. Partnering exists in the construction industry [2] [3] [4] [16]. Our study refers to the findings of that sector.

A. Problems In this study, we call (EP) a business project conducted by either a FDI or a national

enterprise. And (PoD) is a local economic or social development projects conducted by Government or Non Government Organization or Association. The first project has a high economic interest since the second one generally targets social attention. So their objectives are different. Sometimes the first donates to the second just for visibility or for social action. FDI’s impact, powered by (EP), rarely influences local population daily life. And many are (PoD) which do not attain the expected results. Some petroleum enterprise tries to partner with. The results were ambivalent.

B. Research questions

55

What are the maintained benefits from partnering a (PoD) and an (EP)? And how do entrepreneurs rank those maintained benefits? Then how do those maintained benefits correlate to previous research results in construction industry?

What are the Critical Success Factors (CSFs) that impact positively partnering of (PoD) and (EP)? And how do entrepreneurs rank those CSFs? Then how do those CSFs correlate to previous research results in construction industry?

C. Objectives

The objectives of this survey are to assess the results of partnering a Project of Development and an Enterprise Project. Then a model is proposed to assure a better partnering of those two projects.

II. LITTERATURE REVIEW

A. From Project Management Success to Project Success

Through a logic frame, project management deals with “activities” and “results”. It is essentially intended to accomplish the related “goal”. All those three levels of logic frame should accomplish three objectives: completing the project within budget, on time and meeting specifications [1] [5]. Project management success treats with “indicators” established in the logic frame. It can influence project’s success [11] [17] [18].

Critical Success Factors are agreed metrics on which the project’s success will be measured [7]. They are initially elaborated through a concurrent engineering roundtable and then selected and validated by the relevant project stakeholders. B. From “Iron-Triangle” to “Iron-Square”

Projects are often considered successful if delivered on time, within the budget and in accordance with user requirements [12]. They are metrics to measure project’s success. These three CSFs are known as the “Iron Triangle”, and shown in Figure 1.

Figure 1: Iron Triangle of project success The same format is valid in enterprise project or government project. This confirms that

the above model is relatively flexible. Besides, there are four dimensions of project success: Benefits to the customer,

Meeting design goals, Commercial success and Future potential [7] [14]. Those dimensions map the “Iron Square”. They may be liken to Skinner‘s [15] metrics:

PROJECT SUCCESS

DELIVERED ON TIME

MEET USER REQUIREMENTS

COMPLETED WITHIN THE BUDGET

56

- flexibility: meeting design goals, - quality and cost: benefits to the customers, commercial success, - speed of delivery: future potential.

The following Figure 2 represents the above four dimensions.

Figure 2: Iron Square for project success

III. METHODOLOGY The research follows the framework of PDCA Cycle. PDCA stands for Plan-Do-Check-

Act, often called Deming Wheel [6]. It conveys the sequential and continual nature of the continuous improvement process.

The “Plan” phase of the cycle is where a global problem and specific problems are identified. It is also where the analysis is done and the countermeasures are devised [9]. The output of that first phase is a Logic Frame or Project Document. At this stage, the “Skills and Employability Department of the International Labor Office” has formulated a project document entitled: “Operationalizing Pro Poor Growth Project”[8]. That Project has three components: Component (i) “To build a common policy understanding at country level on operationalizing pro-poor growth”, Component (ii) “Poverty Reduction through Skills and Micro and Small Enterprises (MSE) Development”, and Component (iii) “Creating decent and productive employment opportunities through local resource-based investment policies and practices for infrastructure” [8].

Our study concerns Component (ii). The “Do” phase of the PDCA Cycle deals with implementing it. It had been implemented on a small scale. It was in two Communes Alatsinainy Ibity and Mangarano of Vakinankaratra Region-Madagascar. This component is geared to addressing skills gaps among rural poor to improve their employability and to support the development of MSEs and their integration into higher national and global value chains, thus promoting more equitable access of the poor to market opportunities and their integration into the mainstream of development [8]. This is being done through developing and adapting ILO TREE and Value Chain Upgrading methodologies to country-level situation, and developing complementarities between both approaches in the context of local development initiatives [8].

The study focuses on “Check” and “Act” phases. The “Check” phase deals with evaluating data collected six months after completion of the Project. And the “Act” phase concerns the modeling so that replication or scaling up is easy to implement. The following Deming Wheel synthesizes what are explained.

PROJECT SUCCESS

BENEFITS TO THE CUSTOMERS

MEETING DESIGN GOALS

CO

MM

ER

CIA

L S

UC

CE

SS

FU

TU

RE

P

OT

EN

TIA

L

57

Figure 3: Deming Wheel

A. Research steps at “Check” phase The following process represents the action plan taken at this level.

Figure 4: Research steps at "Check" phase

B. Research steps at “Act” phase We consider TREE four component stages and enterprise life cycle project:

- design enterprise life cycle project, - develop a Systems engineering.

C. Questionnaire design There are two parts in the questionnaire. The “four basic operations strategies” of

Skinner [15], cost, quality, speed of delivery and flexibility are utilized to elaborate the format of the first part. The second part is shaped by the TREE’s mainstreamed elements and component stages. Each part is designed for collecting the respondent’s perceptive views on employees or sub contractors, from vulnerable families, co-trained by the above Operationalizing Pro Poor Growth Project and enterprises. Each variable of the questionnaire is rated on standardized ordinal scale of 1 to 5. Ranging from “strongly disagree=1” and to “strongly agree=5” was adopted to measure the response of the first part., and from “unusual=1” to “very important=5” to measure the second part.

Elaboration of a Project Document articulated around three Components.

PLAN

Pilot Project:: implemen-tation of Component (ii) in a small scale.

DO

Systems engineering to manage replication or scaling up.

ACT

Evaluation after six months.

CHECK

Design the questionnair

e

Collect data Analyze data

Skinner’s metrics

TREE mainstreamed elements and

component stages

Correlate to previous research results in

construction industry

58

D. Sample and data collection 34 enterprises, which have worked with the above Project and the Project’s trainees, are

contacted. 33 are all based in the Vakinankaratra Region and 1 in Antananarivo). These enterprises operate in a variety of activities such as cement, garment, textile, embroidery, and agriculture. The questionnaire is used to collect data. E. Data analysis

- rank data, - correlate data to previous research results in construction industry

[2][3][4][10][16] Black et al. rated 13 benefits of project partnering [2]. Lu and Yan [10] rated some

parameters which are likable to 8 of those 13 benefits. Concerning the Critical Success Factors that impact partnering, Chan et al. [3], Chen [4] and Tang et al. [16] finding results were utilized for comparison.

IV. FINDINGS

Results, in Table 1, sort in descending order the maintained benefits from partnering (PoD) and (ED). Respondents in TREE project agree that Cost, Quality and a part of Speed of Delivery and a part of Flexibility are maintained benefits during partnering.

Table 1: Maintained Benefits from Partnering (PoD) and (EP)

N° MAINTAINED BENEFITS FROM PARTNERING (PoD) AND (EP) TREE Black

[2] Lu

[10] COST

1 (ES) contribute to the reduction in project cost 4.75 3.81 3.36 QUALITY

2 (ES) respect the quality of deliverables 4.85 3.69 3.45 3 (ES)'s deliverables fit customer satisfaction 4.53 4.19 3.45

SPEED OF DELIVERY 4 (ES) contribute to increase market share 4.25 3.22 4.01 5 (ES) respect productivity 3.12 3.92 3.47

FLEXIBILITY 6 (ES) adapt to new design 3.50 3.38 3.59 7 (ES) adapt to new technology 3.07 3.38 3.59

(ES): employees or sub contractors, from vulnerable families, co-trained by Operationalizing Pro Poor Growth Project and Enterprises.

Pearson Correlation Coefficient is applied on data in Table 1. Similarity and columns comparison were checked. There is similarity within the interval [-1, +1]. A table 2 shows that there is similarity between the results from TREE, the present study, and the results from Black [2], similarity is 0.302. There is also a similarity -0.155 between TREE and Lu [10].

Table 2: Matrix (Pearson Correlation Coefficient) a bout Maintained Benefits

TREE Black Lu TREE 1.000 Black 0.302 1.000 Lu -0.155 -0.738 1.000

In the following Table 3, Critical Success Factors are sorted in descending order. In the mainstreamed elements of TREE: (ES) organized and empowered on the first hand, and

59

Local Economic Development on the second hand are important for the respondents. In the component stages of TREE, the factors arranged in descending order below are important:

- Training are designed in accordance with enterprises' or market requirements,

- Co-training of (ES) by the Project and enterprises, - Enterprise acceptance of partnership, - Understanding the partnership and definition of the responsibilities.

CSFs corresponding to mainstreamed elements of TREE are not included in Chan [3], Chen [4] and Tang [16] surveys.

Table 3: Critical Success Factors that Impact Posit ively Partnering (PoD) and (EP)

N° CRITICAL SUCCESS FACTORS THAT IMPACT POSITIVELY PARTNERING

(PoD) AND (EP) TREE Chan

[3] Chen

[4] Tang [16]

MAINSTREAMED ELEMENTS OF TREE 1 (ES) organized and empowered 4.75

2 Local Economic Development 4.75 3 Gender approach 3.50

COMPONENTS STAGES OF TREE 1 Training are designed in accordance with enterprises' or market requirements 4.85 4.16 4.12 4.50 2 Co-training of (ES) by the Project and enterprises 4.75 4.24 4.32 3.50 3 Enterprise acceptance of partnership 4.55 4.00 4.08 4.00 4 Understanding the partnership and definition of the responsibilities 4.55 4.10 4.24 4.00 5 Formation of partnership at the beginning 3.50 4.38 4.05 4.17 6 Stakeholders involvement at the beginning 3.50 4.00 3.16 3.50


There is a similarity between TREE and Chan [3], TREE and Chen [4], and TREE and Tang [16]. The following Table 4 shows the above similarities.

Table 4: Matrix (Pearson Correlation) about Compone nts Stages of TREE

TREE Chan Chen Tang TREE 1.000 Chan -0.150 1.000 Chen 0.708 0.462 1.000 Tang 0.256 0.254 0.398 1.000

Table 5 shows that three measures are identified. They are all very important according to the respondents. The first and the third are related to the enterprise. Since the second concerns the TREE.

Table 5: Measure for Partnering Success

N° MEASURE FOR PARTNERING SUCCESS TREE 1 Partnering is considered a success if it contributes to the enterprise's competitiveness 5.00 2 Partnering is considered a success if it generates benefit to the (ES) and women 5.00 3 Partnering is considered a success if there is political stability 5.00


60

V. DISCUSSION We have identified that seven benefits have been maintained from partnering a project

of development powered by TREE and projects conducted by 34 enterprises which are employing or working with vulnerable families co-trained by those two groups of project. Our study has shown that in such partnership, the (PoD) can attain its economic and social objectives on the first hand and the (EP) can stay competitive, viewed from Skinner metrics, on the second hand. Vulnerable families can participate in economic activities: in a wage employment or in a self employment. Enterprises continue to export quality products. They can increase their market share and decrease their cost. So in such partnership, decent work for vulnerable families is compatible with competitiveness for enterprises.

Then, nine Critical Success Factors (CSF) affect positively partnering (PoD) and (EP). Three of those identified CSFs are some of the mainstreamed elements of TREE such as Gender equality, local development and organizing and empowering community groups. And six of them are likened to some of the components stages of TREE: Institutional organization and planning, Community assessment and planning-Identification of Economic Opportunity-Training needs assessment, Plan-Design-Deliver training, Provide post training support to help securing sustainability of economic activities. TREE methodology can create synergy between (PoD) and (EP).

And three factors are identified to measure their partnering success. Such partnering contributes to the enterprise’s competitiveness, generates benefit to the community groups and assures political stability. That permits to get the following golden triangle concerning partnering success.

Figure 5: Golden Triangle about Partnering Success

Similarities with previous research results, Black [2], Chan [3], Chen [4], Lu [10] and Tang [16], confirm that there is not any risk in partnering (PoD) and (EP). The first and the project’s beneficiaries do not create any problems, related to quality-speed of delivery-flexibility-cost, to the second. The second assures decent work to those beneficiaries.

Besides, both (PoD) and (EP) could attain their respective objectives. Vulnerable families and women have been inserted in a process of development conducted by private enterprises. It is a TREE objective. Enterprises have maintained or improved their profit.

In such scheme, each part is committed and strives for its goal. Vulnerable families and women could access to a sustainable decent work if there is enterprises’ development. And enterprises could develop in a globalized competition if there is a political stability. That tri dimensional synergy contributes to map and implement a “Fair Globalization”. Fair is mentioned, because each part gets its benefit from the process.

PARTNERING SUCCESS

BENEFITS TO VULNERABLE FAMILIES AND WOMEN

POLITICAL STABILITY

CONTRIBUTION TO ENTERPRISE’S COMPETITIVENESS

61

VI. CONCLUSION

So far, studies are focused on either Project of Development or Enterprise Project. The present study is important as far as it analyzes partnering of those two kinds of project. Some Critical Success Factors have been identified and sorted for partnering Project of Development and Enterprise Project. Study has shown that if the second is powered by the first, its socio economic impact would be significant. It gives a new face of globalization called “Fair Globalization” by the International Labor Office.

In that specific case, TREE methodology may avoid marginalization, of vulnerable families and women, due to hard investment led by FDI. To facilitate replication or scaling up, we suggest two systems engineering. A new Project Life Cycle [13] is powered by TREE. The first two stages of TREE will feed the first stage of the new Project Life Cycle. And the last two stages will supply the “Development” phase.

Project of development can benchmark TREE methodology and enterprise project can

scale the above project life cycle powered by TREE.

ACKNOWLEDGEMENTS We would like to thank: the International Labor Offices in Geneva (Skills and

Employability Department) and in Antananarivo, the Vakinankaratra Region, the Chamber of Commerce and Industry of Vakinankaratra, the Chamber of Agriculture of Vakinankaratra and the Chamber of Craft Industry of Vakinankaratra. REFERENCES

[1] Baccarini, D.; “The Logical Framework Method for Defining Project Success,” Project Management Journal, Vol. 30, No. 4, pp. 25-32, 1999.

[2] Black C., Akintoye A., Fitzgerald E.: “An Analysis of Success Factors and Benefits of Partnering in Construction,” International Journal of Project Management, Vol. 18, pp. 423-434, 2000.

[3] Chan A.P.C., Chan D.W.M., Fan L.C.N., Lam P.T.I., Yeung J.F.Y.: “Partnering for Construction Excellence – A Reality or Myth?” Building and Environment, Vol. 41, pp. 1924-1933, 2006.

[4] Chen W.T., Chen T.T.: “Critical Success Factors for Construction Partnering in Taiwan,” International Journal of Project Management, Vol. 25, pp. 475-484, 2007.

[5] Cookies-Davies, T.; “The ‘Real’ Success Factors on Projects,” International Journal of Project Management, Vol. 20, pp. 185-190, 2002.

[6] Deming W.: “Quality, Productivity, and Competitive Position”, (Cambridge, MA: MIT Center for Advanced Engineering Study), 1982

Institutional organization and planning

Plan, design and deliver training

Provide post-training support to help securing sustainability of economic activities

Community assessment and planning, identification of economic opportunities, training needs assessment

INITIATION DEVELOPMENT VALIDATION DEPLOYMENT CLOSURE

PROJECT LIFE CYCLE

FOUR COMPONENT STAGES OF TREE

62

[7] Dvir D., Lipovetsky S., Shenhar A.J., Tshler A.: “In Search of Project Classification: a Non-Universal Approach to Project Success Factors”, Research Policy, Vol. 27, pp. 915-935, 1998

[8] International Labor Office; “CBT-TREE”, retrieved O6/13/09 World Wide Web, www.ilo.org/english/publications

[9] Lillrank P., Kano N.: “Continuous Improvement: Quality Control Circles in Japanese Industry”, (Ann Arbor: University of Michigan, Center of Japanese Studies), p. 27, 1989

[10] Lu S., Yan H.: “An Empirical Study on Incentives of Strategic Partnering in China: Views from Construction Companies,” International Journal of Project Management, Vol. 25, pp. 241-249, 2007.

[11] Mahaney R. C., Lederer A.L.: “The Effect of Intrinsic and Extrinsic Rewards for Developers on Information Systems Project Success,” Project Management Journal, Vol. 37, no. 4, pp. 42-54, 2006

[12] Peppard J., Daniel E.: “Managing the Realization of Business Benefits from IT Investments”, MIS Quarterly Executive, Vol. 6, Is. 1, pp. 1-11, 2007

[13] Sargent R.G., Nance R., Overstreet M.C., Robinson S., Talbot J.: ”The Simulation Project Life: Models ans Realities”, Proceedings of Simulation Conference, 2006

[14] Shenhar A.J., Dvir D., Levy O., Maltz A.C.: “Project Success: A Multidimensional Strategic Concept”, Long Range Planning, Vol. 54, pp. 699-725, 2001

[15] Skinner W.: “Manufacturing-The Missing Link in Corporate Strategy”, Harvard Business Review, Vol. 47, N° 3, pp. 136-145, 1969

[16] Tang W., Duffield C.F., Young D.M.: “Partnering Mechanism in Construction: An Empirical Study on the Chinese Construction Industry,” Journal of Construction Engineering and Management, pp. 217-229, 2006.

[17] Yu A.G., Flett P.D., Bowers J.A.: “Developing a Value-Centered Proposal for Assessing Project Success,” International Journal of Project Management, Vol. 23, pp. 428-436, 2005

[18] Zwikael O., Globerson S.: “From Critical Success Factors to Critical Success Process,” International Journal of Production Research, Vol. 44, No. 17, pp. 3433-3449, September 2006.

APPENDIX

TREE or Training for Rural Economic Empowerment is an International Labor Office’s Methodology for Local Economic Development.

What is it?

It is an approach to skills development that: • is focused on identifying potential income generating activities (employment and

self employment) before determining training needs, • involves the local community and social partners at all stages of definition,

design and delivery process, • ensures follow up support beyond training, • it is particularly adapted to Rural and otherwise isolated communities where

there are few formal jobs, • it focuses on Training and is intended to increase Economic Empowerment for

the people and communities involved, • it builds on the ILO’s “Community Based Training” approach which has been

successful for over 20 years in rural, isolated and very poor communities in Asia, East Africa and the Caribbean.

Its process is presented as following:

63

At each stage of process, four mainstreamed elements are considered:

- Gender equality - Local Development - Organizing and empowering community groups - Monitoring and evaluation

Where is it used?

• TREE projects are currently operating in Madagascar, Pakistan, the Philippines, Sri Lanka and East Timor. (2006)

• The Government of Pakistan has adopted the methodology for its own employment programme.

• Past projects in Bangladesh, Cambodia and Jamaica have left legacies of tools, institutions and capacity that are still used

Institutional organization and planning Institutional organization and planning

Plan, design and deliver training

Provide post-training support to help securing sustainability of economic activities

Community assessment and planning, identification of economic opportunities, training needs assessment

64

International Journal of Industrial Engineering and Management (IJIEM), Vol.2 No 2, 2011, pp. 45-50 Available online at http://iim.ftn.uns.ac.rs/ijiem

ISSN 2217-2661

Using Patent Statistics and Principal Component Ana lysis to Predict Global Competition

François A. Ravalison 1, Narisoa Rabenja 2

1University of Antananarivo-Ecole Supérieure Polytechnique, Antananarivo, Madagascar 2Malagasy Intellectual Property Office, Antananarivo, Madagascar

Received (14 February 2011); Revised (13 Jun 2011); Accepted (29 Jun 2011)

Abstract

Globalization and acute competition require continuously searching new approach in identifying ways for better economic growth. The purpose of this paper is to identify a process to predict global competition. The data are collected from the Malagasy Intellectual Property Office. Frequencies of foreign patents, registered at that Office, from 1994 to 2009 have been collected. Then, data mining is conducted to bring out an idea. Findings reveal that Competition indicators based on patent statistics are confirmed as appropriate measure of competition. Then Principal Component Analysis (PCA), applied on Competition indicators, permits to predict sector of global competition in a country level. This paper serves as a valuable analytical framework for the management of patent data for continuous innovation for economic growth.

Keywords : Competition Indicator, Global Competition, Patent, Principal Component Analysis

1. INTRODUCTION

Global competition begins to be present on the developing countries’ market. It affects all sectors of economy. Africa in general and Madagascar in particular cannot avoid that economical trend. Lack of appropriate strategy has thrown some national or private enterprises in difficulties. Others have closed their doors. And then, unemployment rate increases. There is no reliable indicator which permits to measure and to predict global competition. Sometimes, data exist but they are insufficiently analyzed and outputs are not complete. Even some foreign enterprises, for example French banks in Madagascar, utilize competitive intelligence but they do not have reliable Critical Success Factors to gauge and to anticipate even local competition. Two research questions appear: How can global competition be measured? How to determine the global competition trends? The objectives of this study are:

- to identify measurement of global competition, - to identify direction of global competition.

2. LITTERATURE REVIEW

2.1 Patent concerns knowledge creation Patents are an objective measure of R&D activities [2]. If that the case, we call it “patent activity”. The ability of patent information to measure R&D activities is attributed to some characteristics of patents [2]. A large amount of technological information is contained in patents [3] since R&D is concerned. For that, they are often the only source for the timely recognition of technological change [1] [2].

2.2 Patent concerns performance and competition R&D indicates technological activities which lead to subsequent market [3]. And when a firm changes its R&D expenditures, parallel changes occur in its level of patenting [4]. Patent quality measures the impact of those activities [3]. Besides, a patent present a patentee’s non-negligible expectation as to the ultimate utility and marketability of the invention [3]. And Ernst [2] has establishes a positive relationship between patent applications and subsequent sales increase. And then, he [3] has established empirical evidence that suggests a positive relationship between patenting strategies and company performance.

65

2.3 Previous research about patent and competition- a Competition Indicator Banerjee et al. have conducted significant research about patent and competition. They say that a patent is an input to the processes of production or services. And patents statistics permit understanding the degree of competition and the competition-driven research strategy [1]. Besides, those patent statistics can be used to understand the nature of competition through game theory modeling of patent participants and the competition in which they are involved [1]. Many scholars do in fact treat these statistics as indicator of “competitiveness” [8], [9], [12]. Banerjee et al. have proposed a set of indicators based on patent statistics [1]. Those indicators make known scale of competition between foreigner and national researchers. We suggest utilizing the following [10]:

Let (tij) represent the number of patents of type i in year j. We calculate: T� � � ��

as the total numbers of patents during year j.

And:

T� � � ��

the total numbers of patents of type i over all years.

So:

T � T� � T�

as total of all types of patents over all years.

The competition indicator Iij for each patent of type i for each year j [10]:

I�� T�T�T

The numerator shows the share of patent of type i for a year j. And the denominator shows the share of patent of type i of all types over all years.

2.4 Principal Component Analysis Principal Component Analysis (PCA) is a multivariate technique that analyzes a data table in which observations are described by several inter-correlated quantitative dependent variables. Its goal is to extract the important information from the table, to represent it as set of new orthogonal variables called principal components. (PCA) also represents the pattern of similarity of the observations and the variables by displaying them as points in maps [6], [11].

Let use the following notations: A: matrix a : vectors x : elements of matrix I : identity matrix superscript T: transpose operation

There are I observations and J variables in the data table represented by IJ matrix X whose element is xi,j. and the level or the rank of matrix X is L which is computed as:

66

rank of matrix X � min��, ��

In general, the data table will be processed before the analysis. The columns of X will be centered so that the mean of each column is equal to zero. It has the following singular decomposition [14]:

X � PΔQ� P: matrix IxL of left singular vectors. Q: matrix JxL of right singular vectors. ∆: diagonal matrix of singular values Δ� � Λ Λ: diagonal matrix of eigenvalues of XTX and XXT.

The inertia of a column is:

γ�� "�,��#�

And the inertia ℐ of the data table or the total inertia is computed as:

ℐ � � γ�,��

The Center of Gravity g of the rows is the vector of the means of each column of X. and the Euclidean distance of the i-th observation to g is equal to:

%�,�� &"�,� ' (�)�*�

In PCA, the components are obtained from the singular decomposition of the data table X. From Takana decomposition above, the IxL matrix of factor scores is calculated by:

F � PΔ

The matrix Q gives the coefficients of the linear combinations used to compute the factors scores: F � PΔ � PΔQQ� � XQ

Matrix Q is a projection matrix which transforms the original data into factor scores. This matrix can also be used to compute factor scores for observations. The latter are called supplementary (sup) or illustrative observations.

The component can also be represented geometrically by the rotation of the original axes. X � FQ�

with: F�F � ∆�

Q�Q � I

This decomposition is called bilinear decompositionThe observations to compute PCA are called observations are obtained by first positioning these observations into PCA space and then projecting them onto the principal components:

The goals of PCA are: extract the moset by keeping only this important information, simplify the description of the data set, analyze the structure of the observations and the variables.

3. METHODOLOGY 3.1 Research process

The research process contained three steps. It is presented in the Figure 1 below.

The inputs of the first step are data from Malagasy Intellectual Property Office. Data, concerning patent deposited by Non Residents (NR), have been captured in Excel. A table with three columns such as Year; International Patent Classification (IPC); Country of Origin, has been obtained. So per year, frequency of patents deposited is obtained. The output of that first step is Patent Statistics. In the present case, Patent Statistics is a two dimensional array. On the basis the second step. Then from the matrix of Competition Indicators the Principal Component Analysis is processed at the third step.

3.2 Software tool Data have been captured in Excel 2007. And XLSAnalysis.

4. FINDINGS The output of the second step of the research process has been charted and has permitted to obtain the following line of Competition indicators.

Figure Source: Authors

Collect frequency of patents deposited by Non Residents in Madagascar

67

bilinear decomposition of X [7]. The observations to compute PCA are called active observations. The factor scores for supplementary observations are obtained by first positioning these observations into PCA space and then projecting them onto

The goals of PCA are: extract the most important information from the data table, compress the size of the data set by keeping only this important information, simplify the description of the data set, analyze the structure of


Figure 1: Research process

The inputs of the first step are data from Malagasy Intellectual Property Office. Data, concerning patent deposited by Non Residents (NR), have been captured in Excel. A table with three columns such as Year;

Patent Classification (IPC); Country of Origin, has been obtained. So per year, frequency of patents deposited is obtained. The output of that first step is Patent Statistics. In the present case, Patent Statistics is a two dimensional array. On the basis of those statistics, Competition Indicators [1] are calculated at the second step. Then from the matrix of Competition Indicators the Principal Component Analysis is processed

Data have been captured in Excel 2007. And XLStat 6.0 has been utilized to process the Principal Component

The output of the second step of the research process has been charted and has permitted to obtain the following line of Competition indicators.

Figure 2: Trend Over Time of Competition Indicators

Calculate matrix of Competition Indicators

Process the Principal Component Analysis

The factor scores for supplementary observations are obtained by first positioning these observations into PCA space and then projecting them onto

st important information from the data table, compress the size of the data set by keeping only this important information, simplify the description of the data set, analyze the structure of


The inputs of the first step are data from Malagasy Intellectual Property Office. Data, concerning patent deposited by Non Residents (NR), have been captured in Excel. A table with three columns such as Year;

Patent Classification (IPC); Country of Origin, has been obtained. So per year, frequency of patents deposited is obtained. The output of that first step is Patent Statistics. In the present case, Patent

of those statistics, Competition Indicators [1] are calculated at the second step. Then from the matrix of Competition Indicators the Principal Component Analysis is processed

tat 6.0 has been utilized to process the Principal Component

The output of the second step of the research process has been charted and has permitted to obtain the

Process the Principal Component Analysis

68

There are eight lines corresponding to eight patent sectors: Human necessities, Performing operations and Transportation, Chemistry and Metallurgy, Textile and Paper, Fixed construction, Mechanical engineering, Physics, Electricity. The majority of CI is superior to 0 and inferior than 2. Five sectors have their CI>2: Electricity, Textile and Paper, Performing operations and Transportation, Fixed construction, and Mechanical engineering. Textile and Paper is a cut above the rest. Its CI was at 5 in 2002. A first cluster of sectors has trendline with gentle slope<0: Human necessities, Performing operations and Transportation, Chemistry and Metallurgy, and Physics. A second cluster has trendline with gentle slope>0: Fixed construction, and Electricity. And a third one has trendline with steep slope>0: Textile and Paper, Mechanical engineering. We notice that the third cluster is competitive followed by the second cluster. The first one is not competitive.

At the third step, we have processed with covariance as type of matrix and variances with 1/n. The number of factors associated with non-trivial eigenvalues is 7, and 1 is removed. The following Principal Component Analysis is obtained.

Figure 3: Principal Component Analysis of Competiti on Indicators Source: Authors

In first quadrant, there is only one IPC relating to Textile and Paper. In the second, there are three IPCs: Electricity, Mechanical engineering and Fixed construction. There are two IPCs in the third and fourth quadrants: Chemistry-Metallurgy and Physics for the first one and Human necessities and Operations-Transportation for the second one. 5. DISCUSSION

Global competition can be measured by Banerjee’s Competition indicators. And Principal Component Analysis permits to determine the global competition trends. Banerjee survey concerns the sector of biotechnology because it is an emerging area with acute competition [1]. Besides, it is a very high expectation on the future economic returns [1]. In our case, we have considered all sectors define by International Patent Classification. Our area of global competition is focused in Madagascar. That global competition is measured by Competition indicators. The corresponding result is acceptable but confused. The values of indicators are alike. Maping of Banerjee is clear than our through trend over time. To determine the global competition trends, we have used Principal Component Analysis. This one is an acceptable and clear result and confirms that Banerjee’s theory on Competition indicators can measure global competition at a local level. Besides, application of Principal Component Analysis permits to map those trends.

0

A: Human necessities

0

B: Performing operations;

Transportation

0

C: Chemistry; Metallurgy

0

D: Textile; Paper

0

E: Fixed construction

0

F: Mechanical engineering

0

G: Physics

0

H: Electricity

-1

-0,5

0

0,5

1

-1 -0,5 0 0,5 1

--ax

is F

2 (2

3.00

%)

-->

-- axis F1 (43.00 %) -->

Variables (axes F1 and F2: 66.00 %)

69

That application contributes to some explanation about some problematic relating to Competition indicators [10]. It is said that the above theory does not show real competition. Our research shows that it can be processed to predict competition trends. The Principal Component Analysis shows that it is an appropriate tool to map factors’ profile [13]. In our case, it shows that “Textile; Paper” is a domain with acute competition. 6. CONCLUSION

In this paper, Competition indicators have been used to map patent competition in an African developing country. Such mapping may be clear or confused according to indicators’ values. The Principal Component Analysis has been used to improve the mapping and to get a new idea concerning competition trends. Practical implications:

- The research process can be used as a system engineering, - The outputs of such system can be utilized to predict what domains of competition are important to

focus on. Originality/value:

- Contribution of an “added value”, through a statistic tool, to a past research. Future research:

- The domain of competition is known. And survey of enterprises created, related to that domain, needs to be conducted.

7. ACKNOWLEDGEMENTS

We would like to thank the Malagasy Intellectual Property Office. Its Director General has authorized us to access their precious data. He has accepted one of his staff to collaborate with us to perform the present paper. 8. REFERENCES

[1] Banerjee P., Gupta B. M., Garg K.C.; “Patent Statistics as Indicators of Competition: an Analysis of Patenting in Biotechnology”, Scientometrics, Vol. 47, pp. 95-116, 2000.

[2] Campbell, R.S.;”Patent Trends as a Technological Forecasting Tool”, World Patent Inf., N°5, Vol.3, pp.137–143, 1983.

[3] Ernst, H.; “Patentinformationen für die Strategische Planung von Forschung und Entwicklung”, DUVVerlag-Wiesbaden, 1996.

[4] Ernst, H.; “Patent Portfolios for Strategic R&D Planning”, Journal of Engineering and Technology Management, Vol.15, pp. 279-308, 1998.

[5] Griliches, Z., Pakes A., Hall B.H.; “The Value of Patents as Indicators of Inventive Activity, Discussion Paper No. 1285, Harvard Institute of Economic Research, Cambridge, MA., 1986.

[6] Jolliffe I.T.; “Principal Component Analysis”, New York, Springer, 2002. [7] Kruskal J.B.; “Factor Analysis of Principal Component Analysis: Bilinear Methods”, International Encyclopedia of

Statistics, pp. 307-330, New York, The Free Press, 1978. [8] Pavitt K., Patel P.; “The International Distribution of Determinants of Technological Activities”, Oxford Review of

Economic Policy, Vol. 4, N°4, pp. 35-55, 1988. [9] Pavitt K.; “Patents Statistics as Indicators of Innovative Activities: Possibilities and Problems”, Scientometrics, Vol.

7, pp. 77-99, 1985. [10] Ramani S. V., De Looze M. A.; “A Note on Using Patent Statistics to Obtain Competition Indicators”,

Scientometrics, Vol. 49, N°3, pp. 511-515, 2000. [11] Saporta G., Niang N.; “Principal Component Analysis: Application to Statistical Process Control”, Data

Analysis,pp. 1-23, 2009. [12] Soete L.; “The Impact of Technological Innovation International Trade Patterns: the Evidence Reconsidered”,

Research Policy, Vol. 16, N° 2-4, pp. 10, 1987. [13] Suhr D.D.; “Principal Component Analysis vs. Exploratory Factor Analysis”, Statistics and Data Analysis, Paper

203-30, 2007. [14] Takana Y.; “Relationship Among Various Kinds of Eigenvalue and Singular Decompositions”, New Development

Psychometrics, pp. 45-56, 2002.

70

Volume X, Number Y, ZZZZ ISSN 1995-6665 Pages XX-XX

Jordan Journal of Mechanical and Industrial Engineering

Ms. No. Ref. JJMIE-201-11” Critical Success Factors are the Essence to Improve Productivity: A Technical Note on the Article “ Jig Design, Assembly Line Design and Work Station Design and their Effect to Productivity” BY François A. Ravalison

Critical Success Factors are the Essence to Improve Productivity A Technical Note on the Article “Jig Design, Assembly Line Design and Work Station Design

and their Effect to Productivity” François A. Ravalison

Ecole Supérieure Polytechnique-University of Antananarivo-Madagascar

An article, “Jig Design, Assembly Line Design and Work Station Design and their Effect to Productivity” from Adi Saptari, Wong Soon Laï, Mohd Rizal Salleh [1], was published in the Jordan Journal of Mechanical and Industrial Engineering on February 2011. It is an important and interesting article: necessary improvements are detected and knowledge is identified. Concerning the design of assembly line, the authors said in the case of design of two operators that the first operator begins from task 1 through task 5 and the second one from 6 to 8. They also said that the second operator did not require waiting for the first operator and could start the experiment at the same time. As a result, the assembly line with cumulative assembly time is:

Figure 1: The assembly line based in the case of tw o operators The above process shows that the second operator did require waiting for the first operator; at the beginning of the experiment, he has to wait for 9.66 sec. In addition, during experiment, he has to wait for 0.32 sec per plug. Therefore, if 32 plugs were produced, there was waste of waiting time of 19.9 sec. It is the equivalent of the processing time of one plug. Moreover, it is significant. Concerning the design of experiment, I suggest the following mapping instead of the summary given in table 2.

Figure 2: Mapping of the design of experiments

The advantage is that it is easy to read. Moreover, that will help us to understand the assessment of the group’s performance under results and discussion.

ONE OPERATOR

VERTICAL JIG

STANDING [X1]

SITTING [X3]

RECTANGULAR JIG

STANDING [X2]

SITTING [X4]

TWO OPERATORS

VERTICAL JIG

STANDING [X5]

SITTING [X6]

RECTANGULAR JIG

STANDING [X7]

SITTING [X8]

9.66 sec First Operator

9.34 sec Second Operator

1 2 3 5 4 8 7 6

JJMIE

71

Concerning the groups’ performance, the authors said, “The first test is to verify whether two groups of workers have different quality of work or performance”. They could not say “quality of work” because according to Fouad and al., they have not studied conformance to valid customer requirements [2]. In addition, they have not developed any methods or tools relating to quality process measurement [2]. Moreover, no corresponding data were obtained. The use of the term “performance” is not appropriate. The work of Skinner and the others identified four basic metrics to measure performance: cost, quality, flexibility and speed of delivery [3][4][5][6][7]. Salleh and al. study the assembly time needed to assembly electric plug. For that reason, I think, “the test is to verify speed of delivery” instead of “the test is to verify performance”. Therefore, it is only one metric out of four according to Skinner. When the authors have assessed the setting of experiments in each group of workers through ANOVA, I have detected an improvement:

- In the tables 4 and 5, it is better to use “criterion” instead of “group” written in the second row and first column. “Group” is already used to indicate group of workers. Besides, I notice that Xi and Yi are criteria not group. For example I observe criterion X1= {one operator, vertical jig, standing} according to the above mapping.

For the “first assessment of jig either vertical or rectangular by assuming other factors are in the same set”:

- “If column X1 and X3 for one operator is considered”. I see that criteria X1 and X3 are both for vertical jig. Therefore, this case does not comply with the assessment frame above. I suggest considering X1 and X2, X1 and X4, X3 and X2, X3 and X4. The assembly times are respectively 19.96 sec and 19.22 sec, 19.96 sec and 18.36 sec, 18.71 sec and 19.22, 18.71 sec and 18.36 sec. Only the second case (one out of four cases) provides the most significant gap of assembly time.

- The case of two operators with X5 and X7 (20.96 sec and 19.75 sec) is good as far as the first is for vertical jig and the second for rectangular. They have significant different assembly time like X5 and X8 (20.96 sec and 19.19 sec), X6 and X8 (20.36 sec and 19.19 sec). X6 and X7 with 20.36 sec and 19.75 sec do not have significant different assembly line. For that reason, three cases out of four have significant different of assembly time.

- Four out of eight cases have significant different to assembly time. Thus, I think that the first assessment is wrong. In other words, design of jig either vertical or rectangular does not provide the most significant different to assembly time by assuming other factors in the same set.

The authors said “the second greater significant different to assembly time is the number of operator either 1 or 2 operators with assuming other factors in the same set”:

- Concerning the “the sitting position (X2 and X6)”, I see that X2 is for standing position and X6 for sitting.

- The above assessment permits to summarize cases in the following table.

72

Table 1: Assessment if the number of operators eith er 1 or 2 is considered with assuming other factors in the same set [group (X)]

N° CASES

ASSEMBLY TIME ABSOLUTE VALUE (1)-(2)

(1) OPERATOR

(2) OPERATORS

1 X1 and X5 19.96 20.96 1.00

2 X1 and X6 19.96 20.36 0.40

3 X1 and X7 19.96 19.75 0.21

4 X1 and X8 19.96 19.19 0.77

5 X3 and X5 18.71 20.96 2.25

6 X3 and X6 18.71 20.36 1.65

7 X3 and X7 18.71 19.75 1.04

8 X3 and X8 18.71 19.19 0.48

9 X2 and X5 19.22 20.96 1.74

10 X2 and X6 19.22 20.36 1.14

11 X2 and X7 19.22 19.75 0.53

12 X2 and X8 19.22 19.19 0.03

13 X4 and X5 18.36 20.96 2.60

14 X4 and X6 18.36 20.36 2.00

15 X4 and X7 18.36 19.75 1.39

16 X4 and X8 18.36 19.19 0.83

- Nine (in blue color) out of sixteen cases have significant different to assembly time. So what

the authors said “the second greater significant different to assembly time is the number of operator either 1 or 2 operators with assuming other factors in the same set” is not entirely right.

The authors said “workstation’s design either standing or sitting provides the smallest significant different to assembly time by assuming other factors in the same set”.

- For the case X1 and X2, they said “with an operator and vertical orientation of jig”. According to the mapping above, they correspond to one operator but X1 is with vertical jig and X2 for rectangular one. Both X1 and X2 are standing position.

For group 2 (Y), the results are not similar with group 1 (X). For example, let us consider the case of the number of operator either 1 or 2 operators with assuming other factors in the same set. I obtain the following table.

73

Table 2: Assessment if the number of operators eith er 1 or 2 is considered with assuming other factors in the same set [group 2 (Y)]

N° CASES

ASSEMBLY TIME ABSOLUTE VALUE (1)-(2)

1 OPERATOR

2 OPERATORS

1 Y1 and Y5 20.09 20.81 0.72

2 Y1 and Y6 20.09 20.21 0.12

3 Y1 and Y7 20.09 19.71 0.38

4 Y1 and Y8 20.09 19.37 0.72

5 Y3 and Y5 19.28 20.81 1.53

6 Y3 and Y6 19.28 20.21 0.93

7 Y3 and Y7 19.28 19.71 0.43

8 Y3 and Y8 19.28 19.37 0.09

9 Y2 and Y5 19.58 20.81 1.23

10 Y2 and Y6 19.58 20.21 0.63

11 Y2 and Y7 19.58 19.71 0.13

12 Y2 and Y8 19.58 19.37 0.21

13 Y4 and Y5 18.87 20.81 1.94

14 Y4 and Y6 18.87 20.21 1.34

15 Y4 and Y7 18.87 19.71 0.84

16 Y4 and Y8 18.87 19.37 0.50

The above table shows that for group 2 (Y) only four (in blue color) out of sixteen cases provide significant different to assembly time. For group 1 (X), there were nine.

What has the Salleh and al.’s research report contributed? - The authors have introduced two designs of jig: vertical and rectangular. In the introduction,

they said it is a special tool designed to increase productivity. Surely, that special tool has eliminated any waste of waiting time when setup is concerned. Hence, it has minimized the setup times. Salleh and al.’s research report has contributed to confirm the theory of Just-In-Time.

- They have introduced three critical success factors (jig design, assembly line design, and workstation design) to improve productivity. With cause and effect diagram developed by Fouad and al. for Jordanian Industrial Organizations, the negative characteristics of those critical success factors are factors that could be negatively impact the productivity. Salleh and al.’s research have specified important elements of Ishikawa diagram.

What are other further researches? - Comparative studies based on the improvement of the design of jig, circular rotating and

circular fixed. - Salleh and al.’s research report has introduced a study of speed of delivery. To complete

performance studies based on Skinner’ metrics, studies on quality, cost and flexibility are interesting through an exploratory research.

74

In conclusion, I would like to emphasize that Salleh and al. have raised and answered an important question: what critical success factors are important to improve the productivity? They have identified three. References

[1] Saptari A., Lai W.S., Salleh M.R., “Jig Design, Assembly Line Design and Work Station Design and their Effect to Productivity”, Jordan Journal of Mechanical and Industrial Engineering, Vol.5, N°1, 2011, 9-16

[2] Fouad R.H., Mukattash A.,”Statistical Process Control Tools: A Practical guide for Jordanian Industrial Organizations”, Jordan Journal of Mechanical and Industrial Engineering, Vol.4, N°6, 2010, 693-700

[3] Skinner C., “Manufacturing-The Missing Link in Corporate Strategy”, Harvard Business Review, Vol.47, N°3, 1969, 136-145

[4] Ravalison F., Rahoelison Toky, Raveloson E., Rakotomaria E.: “Developing a Mathematical Concept and Process Technologies to Accompany Firms from Rut to Change”, IEEE/PICMET Transactions on Engineering and Technology Management, 2009, 786-792

[5] Ravalison F., Rajaonary P., Raveloson E., Rakotomaria E, Gazérian J., Loubet C., Ruiz J.M.: “How does Reengineering sustain Economy?-The case of a paper mill industry in Developing Country”, IEEE/PICMET Transactions on Engineering and Technology Management, 2008, 210-219

[6] Ravalison F., Rajaonary P., Raveloson E., Rakotomaria E., “Are Converging Technologies Tools of Competitiveness?-The case of a paper mill industry in Madagascar, IEEE/PICMET Transactions on Engineering and Technology Management, 2007, 241-245

[7] Ravalison F., Raveloson E., Rakotomaria E. : “Towards “Fair Globalization”: Critical Success Factors for Partnering Project of Development and Enterprise Project”, IEEE/PICMET Transactions on Engineering and Technology Management, 2010, 1-7

75

International Journal of Industrial Engineering and Management (IJIEM), Vol.x No x, Month 20xx, pp. xx-xx Available online at http://www.xx

ISSN xxxx-xxxx

Hadamard Matrix to Improve Enterprise’s Activities: an Exploratory Research

François A. Ravalison University of Antananarivo, Ecole Supérieure Polytechnique, 101 Antananarivo-Madagascar, [email protected]

Jens A. Randriampenohaja University of Antananarivo, Ecole Supérieure Polytechnique, 101 Antananarivo-Madagascar

Received (Day Month Year); Revised (Day Month Year); Accepted (Day Month Year) (Editor only)

Abstract The main purpose of this study is to evaluate the extent to which there is a relationship between Hadamard matrix utilization and the enterprise success in terms of improvement. An exploratory research was conducted to measure the influence of factors identified by Hadamard matrix utilization in enterprises’ results. A broad sample of Malagasy enterprises operating in renewable energy was selected. Data was collected through questionnaire with closed questions. Then it was analyzed using XLSTAT software. The findings suggest that enterprise improvement success is correlated to Hadamard matrix utilization. A case study of a bakery enterprise is given to show that the application is not limited to enterprises operating in renewable energy. The results corroborate that Hadamard matrix is a system approach to pilot enterprise improvement.

Keywords: Hadamard Matrix, Improvement, Innovation, Renewable Energy

1. INTRODUCTION Nowadays, enterprises research many paths to improve or to innovate their activities. Many processes are tried to solve the corresponding problems. It is important to say that when problem related to cost is solved, it negatively affects quality or speed of delivery or flexibility. Additionally, it is the same when one of those four critical factors is unilaterally solved. Is there a methodology that could deal with systemic solving process when improvement or innovation is concerned? Thus, we studied 30 companies operating on renewable energy in Madagascar. It deals with the impact of the use of Hadamard matrix on the improvement of the quality, the flexibility, the speed of delivery, the cost and the output for these enterprises. A case study led in a bakery allows highlighting the positive correlation between the application of this matrix and the improvement of the quality. More precisely, we will talk about the possible improvement of time of conservation of the bread by the means of the Hadamard matrix. In addition, we will examine the possibility of the main factors identification in the case of bread staling.

2. LITERATURE REVIEW 2.1 Definition

Referring to works of some authors [4] [7] [9] [14] [16] having more or less the same idea, a Hadamard matrix, named after Jacques Hadamard (1865-1963), is a matrix [H] of order n with entries ±1 and whose rows are all orthogonal to each other. It is determined by the following equation:

[H]t.[H]=n.[In] (1)

The order of a Hadamard matrix is 1, 2 or n≡(0 mod 4). A Hadamard matrix is one type of the method of experiments design.

2.2 Presentation of the experiments design method The term « experiments design method » indicates a complete method to characterize a system. It is based on the modification and the measure of appropriate variables to the considered system. This mainly includes the examined parameters as well as the origin of their variation. Therefore, this method attempts to build up a link between two types of variable:

76

• The results: studied physical variables • The factors: physical variables, adjustable by experimentalist, supposed to have an influence on the

variation of the result.

More precisely, that method aims at the well understanding of the relation connecting the result with the factors as well as the factors between themselves. As for this, the proposed solution consists, in any case, in establishing a model, expressing the result according to factors [17]. The following fig.1 presents the experiment system.

Figure 1: Experiment system Source: The authors

There are two possible uses of experiments design or experiments matrix: • the screening technique • the result surface method

We are here interested in the screening technique, which enables us to classify the factors between themselves according to their particular influence. This technique requires the use of factorial design.

2.3 Construction techniques • Complete factorial design

A complete factorial design [6] [17] is achieved when at least a trial for each combination of factor is fulfilled.

• Factorial designs at two levels Those factorial designs are the easiest. They enable to understand the principle of the method, and to have many applications. The number of experience (n) to fulfil is calculated by:

(2) The number 2 is due to the two levels; k indicates the number of the factors.

• Notation of Yate [5] Owing to this modeling, the different variables are transformed into reduced centered variables. That allows comparing variables of completely different measures. Hereafter is the formula to change the current variables into reduced centered variables: (3) a : the reduced centered variable; A : the current variable; Ā : the average value of current variable; Step : is the variation between the average position of the variable and the extremity of the domain.

Because of this representation of levels, the field of study is intrinsically unobtrusive. That simultaneously allows the use of quantitative variables as well as the qualitative variables or differential variables. The notation of Yate, in fig. 2, authorizes a simple presentation of combinations of levels and allows an easy calculation of effects and interactions [6]. The levels are showed with + and – symbolizing respectively the high and low values of the factor.

Factors Results Yi

Reasoned variation of factors

Calculation of the effects on the results

MATHEMATICAL MODEL Xi

77

Figure 2: Notation of Yate on variable A

• Plackett – Burman matrix [6] [17] The Plackett – Burman matrix is built from Hadamard matrix [6]. The associated Plackett – Burman matrix constitutes a general and simple mean to perform the Hadamard matrix [17]. Thus, for a k factors, let us choose the matrix such as n ≥ k+1, n ≡ (0 mod 4). For the found value n, we should use the series of +1 and -1 indicated by the table of reference below:

Table 1: Table of reference for the construction of the Hadamard matrix [17]

n Series of levels

4 ++-

8 +++ - + - -

12 ++ - +++ - - - + -

16 ++++ - + - ++ - - + - - -

20 ++ - - ++++ - + - + - - - - ++ -

24 ++++ - + - ++ - - ++ - - + - + - - - -

32 - - - - + - + - +++ - ++ - - - +++++ - - ++ - + - - +

Source: Vivier [17]

Then, the rows of the experiments matrix are deduced from all the cyclic permutation of the sequence of +1 and –1. Thus, for example, with k=2, we must choose n=4. By permutation on the right, we have the matrix:

For the select value of n, we add an additional row of -1:

To obtain a matrix of order n, it is necessary to add a column of +1 on the left (average column). Finally, when the average column is added, we obtain the Plakett – Burman matrix that can serve for the creation of experiments design:

Let’s note [H] this matrix, we have the formula below:

+−+++−−++

−−−+−+++−−++

−−−++−++++−+−+++

Result

A1 Ā step A

Y

⊕ Θ Notation of Yate

A2

78

(4)

[H]t is the transposed matrix of [H], [ln] is the identity matrix and n the number of experiences. The calculation of the model coefficients comes from the relation below:

(5) {Y}: vector of results {E}: vector of effects Thereby, for a complete factorial design, {E} is given by:

(6)

Alternatively, because of the property of the matrix [H]:

(7)

2.4 Hadamard matrix and improvement Generally, literature shows that improvement and innovation are possible with Hadamard matrix. The following paragraphs will show this positive sight. On the first hand, Beaufort A. and al. [3] have determined the factors on the level of Listeria monocytogenes reached during storage of smoked salmon at positive temperature. The results confirmed that the degradation of the quality of this food product could be controlled by using Hadamard matrix as experiments method. On the second hand, Matthew H. and al. [13] have reported that beside the success of Hadamard matrix in the theory of signal processing; many kinds of cyclic permutations of genetic elements unexpectedly lead to reconstruction of initial Hadamard matrices into new Hadamard matrices. Thus, they proved the existence of connection of Hadamard matrices with a structural phenomenology of the genetic code that is a path of improvement and innovation. As previously, the work of Horadam K. J. [8] relates the different applications of Hadamard matrices. He said that Hadamard matrices are marvelously useful when improvement and innovation are concerned. He mentions three principal applications of Hadamard matrices: Hadamard transform spectroscopy, object recognition, and coding of digital signals. Applications of experiments are not included in his work but this current survey will deal with it. Finally, Kazuki H. and al. [9] have confirmed that the ratio of the signal in the electropherogram can be improved by a factor of 5 using Hadamard matrix in comparison with that obtained by a conventional single-injection method. Other authors stipulate a new speech-scrambling concept based on Hadamard matrices [15]. They were also employed by Asaff A. and al. [1] in order to increase submerged conidia yields and brown pigmentation of fungal propagules. The list is likely to be long concerning the improvement of the output and cost reduction carried out by certain industries due to the application of Hadamard matrix parameters [6] [17]. Hadamard matrix and parameters are considered equivalent to customization of a product happens before its production [11].

3. METHODOLOGY

3.1 Enterprises operating in renewable energy This research was conducted in 30 enterprises operating in renewable energy in Madagascar. These enterprises had utilized Hadamard matrix to improve their activities. In each enterprise, we have studied the hypothesis: “there is a relationship between Hadamard matrix and the improvement of quality, flexibility, speed of delivery, cost and output. Pearson, Kendall, and Spearman correlation are used to analyze the collected data.

3.2 Case study: a bakery enterprise

79

A case of a bakery, which plans to improve its activities, has been studied. The use of the logical framework allowed to detect that this closure is due, in particular, to the bread staling. The Ishikawa diagram applied in this case shows six main factors causing the bread staling: moulding, fermentation, heating, yeast, water and salt. The levels of the three less important factors, molding, yeast and salt were fixed. In order to improve the quality of the bread to improve the activity of the bakery, the influences of the other three factors were studied using Hadamard matrix. It is obvious that other experimental fields, because of the existence of these 6 initial factors, are also important to explore. Therefore, because of the cost of experimentation for a complete factorial design 26, our study is focused on these 3 factors to show the advantage got by the use of Hadamard matrix in the resolution of our problem.

3.3 Fixation of the levels of the 3 factors: We have fixed a lower and a higher levels of the 3 factors in order to know the effect of those factors in bread staling. These levels are reported in table 2.

Table 2: Level of factors

FACTORS Notation Lower level Higher level

Hydratation rate A 60% (bastard dough) 64% (soft dough)

Mode of fermentation B Polish Direct

Heating (Time-Temperature) C 25 min - 250˚C 15 min - 300˚C

Source : The authors

3.4 Experiments recipes The table 3 shows a matrix of experiences for 3 factors at 2 levels or matrix of complete experiences. It allows determining the trials to carry out.

Table 3: Experiments matrix of a factor planning at 3 factors

Factor 1 Factor 2 Factor 3

n˚ A B C Result

1 - - - Y1

2 + - - Y2

3 - + - Y3

4 + + - Y4

5 - - + Y5

6 + - + Y6

7 - + + Y7

8 + + + Y8 Source Gillon [6]

The detailed recipes of the 8 experiences arise from the previous experiments design 23. The calculation base of hydration rate has been done with regard to the farina. The weight of the farina used for each experience is of 400g. The method of direct fermentation corresponds to 0g of polish and 11g of baking powder. 4g of baking powder have been used for the polish. The levels of the 3 factors are shown in the following table:

80

Table 3: Recipes of experiments

Kneaded dough Polish Salt (g) Time-Temperature n˚ Farina

(g) Water

(g) Yeast

(g) Farina

(g) Water

(g) Yeast

(g)

1 280 120 4 120 120 4 9 25’-250˚C

2 272 128 4 128 128 4 9 25’-250˚C

3 400 240 11 0 0 0 9 25’-250˚C

4 400 256 11 0 0 0 9 25’-250˚C

5 280 120 4 120 120 4 9 15’-300˚C

6 272 128 4 128 128 4 9 15’-300˚C

7 400 240 11 0 0 0 9 15’-300˚C

8 400 256 11 0 0 0 9 15’-300˚C

Source: The authors

After each experiment, we have measured the duration of conservation and the development of the bread.

4. RESULTS

30 enterprises operating in renewable energy

The table below resumes Pearson, Kendall, and Spearman correlation analysis. It shows the relationship between Hadamard matrix and the improvement of quality, flexibility, speed of delivery, cost and output carried out in the 30 enterprises.

Table 4: Relationship between Hadamard matrix and t he improvement of some domains in 30 enterprises

HADAMARD MATRIX ENTREPRISE PEARSON KENDALL SPEARMAN

RELATIONSHIP r p τ p ρ p

Quality 30 0.694 0.000 0.510 0.001 0.692 0.000 Yes (p<0.0005)

Flexibility 30 0.585 0.001 0.433 0.001 0.589 0.001 Yes (p<0.0005)

Speed of delivery 30 0.396 0.0017 0.164 0.197 0.144 0.123 Yes (p<0.02)

Cost 30 0.557 0.001 0.415 0.001 0.584 0.001 Yes (p<0.0015)

Output 30 0.585 0.00 0.433 0.00 0.589 0.00 Yes (p<0.0005)

Source: The authors

Case study: a bakery enterprise

The results of the experiments are shown in the table 6 below. “A” is the hydration rate, “B” the fermentation, “C” the parameter of cooking. The unit of results is “day” for the duration of conservation and “cm3” for the development. After sweating, the breads are left under an ambient temperature under shelter of the sun in order to evaluate the duration of their conservation. Its appreciation has been done through the visual scrutinizing of the change of the texture of the crumb. The results are shown in table 6.

81

Table 5: Effects of factors on the duration of conservation

n˚ Average A B C AB AC BC ABC Result

1 + - - - + + + - Y1 = 3

2 + + - - - - + + Y2 = 3

3 + - + - - + - + Y3 = 2

4 + + + - + - - - Y4 = 2

5 + - - + + - - + Y5 = 4

6 + + - + - + - - Y6 = 4

7 + - + + - - + - Y7 = 2

8 + + + + + + + + Y8 = 2

Effect I = 2,75 E A = 0 EB = -0,75 EC = 0,25 IAB = 0 IAC = 0 IBC =-0,25 IABC = 0

The effects of the factors on the development are shown in table 7.

Table 6: Effects of the factors on the development

n˚ Average A B C AB AC BC ABC Result

1 + - - - + + + - Y1 = 2,51

2 + + - - - - + + Y2 = 2,60

3 + - + - - + - + Y3 = 2,08

4 + + + - + - - - Y4 = 2,49

5 + - - + + - - + Y5 = 2,39

6 + + - + - + - - Y6 = 2,66

7 + - + + - - + - Y7 = 2,26

8 + + + + + + + + Y8 = 2,49

Effect I = 2,43 EA=0,125 EB=-0,10 EC=0,01 IAB=0,03 IAC= 0 IBC =0,03 IABC=-0,04

5. DISCUSSION

Using Hadamard matrix permits to improve enterprise’s activities. Two cases have been studied. The case of 30 enterprises operating in renewable energy has revealed positive and significant relationship between Hadamard matrix and the 5 metrics of performance. First we have obtained relationship between Hadamard matrix and quality (r = 0.694, τ = 0.510, ρ = 0.692 (p < 0.0005)). Second, relationship between Hadamard matrix and flexibility is (r = 0.585, τ = 0.433, ρ = 0.589 (p < 0.0005)). Thirty, Hadamard matrix and speed of delivery (r = 0.396, τ = 0.164, ρ = 0.144 (p < 0.02)). Fourth, Hadamard matrix and cost (r = 0.557, τ = 0.415, ρ = 0.584 (p < 0.0005)). Fifth Hadamard matrix and output (r = 0.585, τ = 0.433, ρ = 0.589 (p < 0.0005)). Statistically, the results above showed that Hadamard matrix has permitted to improve and to innovate the performance of these enterprises operating in renewable energy. Such innovation generates financial results as verified by Fatur and al. [5]. That is because organizational innovations are conducive of cost reduction [11]. Therefore, our idea is to benchmark Hadamard process in baking industry. According to the table 6, we can deduce that the change of the hydration rate from 60% to 64% does not practically have any effect on the conservation (EA = 0). Change in the method of direct fermentation into polish (high level to low level, hence the sign minus for EB = -0,75) brings an effect of 75% for the improvement of conservation duration. In the same way, the level of high temperature (300°C) followed up with the short time of cooking (15’) contributes at 25% of the conservation of the bread (EC = 0,25). Moreover, we have noticed an effect of interactions of 25% (IBC = -0,25) between the cooking method and the fermentation method. The other interactions are null. The best results (4 days) agree to experience 5 and 6; that is to say, a fermentation method on polish and a couple time-

82

temperature of 15’-300°C. The result concerning pol ish confirms the previous researches while the one in relation to the cooking method is inconsistent with the affirmation saying that the lowering of temperature improves the conservation [12]. In fact, the prolonged cooking (25’-250°C) removes a great quantity of water from the dough, and more than that, the crust of the bread becomes thicker. Experimentally, the bread staling is explained as a migration of water from the crumb to the crust [2]. The bread cooked at 250° is getting stale quickly because of th is loss of water and because the importance of the thickness of the crust which amplifies the trapping of the water in this external part of the bread. To sum up, we can deduce that the most influential factor for a better conservation of the bread is the pre fermentation, in our case the polish. However, we have also to use a higher temperature and a shorter time of cooking. As for the development, the calculation of effects of factors cited in the table 6 proves that the change of the hydration rate from 60 to 64% contributes to 12.5% (EA = 0,125). The fermentation method on polish improves also the development with a weight of 10.5% (EB=-0,105). The effects of the method of cooking and the other interactions are insignificant. To reach the aim of our researches, we have to keep an eye on the good conservation of the bread as well as its development at the same time. Among the six levels of factors studied in the experiments matrix, we have to retain a high rate of hydration, pre fermentation instead of direct fermentation, a higher temperature followed by a shorter time of cooking. This corresponds to the experiment n°6. In other respects, let us note that we can reduce the quantity of yeast until 50% by using the fermentation on polish. Hadamard matrix has not only brought an improvement to the bread quality but it also allows a low operating cost. Fatur and al. demonstrated the last when they say that there is a correlation between the revenues arising from innovation and the financial results of a company [5].

6. CONCLUSION

An investigation was carried out in 30 enterprises operating in renewable energy in Madagascar to evaluate the relationship between Hadamard matrix and the success of the enterprises and projects, in term of improvement. The results of Pearson and Kendall and Spearman correlation analysis were positive and significant. Hadamard matrix has permitted to improve the performance of these enterprises operating in renewable energy. Moreover, a case study in bakery highlights the effectiveness of the improvement of bread conservation and its development at the same time. The previous results open another field of experiences in the optimization of the duration of bread conservation. Other matrices as composite matrices, Tagutsi design and simplex can be used in order to find the optimum of this duration. This will contribute generally to the development of the field of researches and particularly the baking and biscuits industry. Practical implications:

- The research output permits to guide enterprise managers when improvement and innovation are concerned.

- The research process allows benchmarking to other areas.

Originality/value: - Hadamard Matrix may be included in the list of classical strategic auditing models.

Future research: - A research about the comparison between Hadamard Matrix and other strategic auditing models to find

precision of each method can be conducted. It permits to appreciate the Hadamard Matrix strengths.

7. REFERENCES

[1]. Asaff A., Escobar F., Mayra de la Torre (2009 ), “Culture Medium Improvement for Isaria Fumosorosea Submerged Conidia Production”, Biochemical Engineering Journal, Vol. 47, pp. 87-92.

[2]. Baik M., Chinachoti P. (2000),”Moisture Redistribution and Phase Transitions During Bread Staling”, Cereal Chem., Vol. 77, pp. 484-488.

[3]. Beaufort A., Bourdin G., Lebail A., Cardinal M. (2007), “Effects of a Storage at -2°C/-3°C Before Retail Dis playing on the Qualities of Smoked Salmon”, 22nd International Congress of Refrigeration, Beijing, China.

[4]. Cameron P. J. (2006), “Hadamard Matrices”, The Encyclopaedia of Design Theory, pp. 1-7. [5]. Fatur P., Likar B., Ropret M. (2010), “Going More Open in Innovation: Does It Pay?”, International Journal of Industrial Engineering and

Management, Vol. 1, N°3, pp. 77-83. [6]. Gillon F. (2004), “Modeling and Optimization by Experimental Design of an Engine with Electronic Commutations”, Sciences and

Technologies University, Lille, France, pp. 40-60.

83

[7]. Holzmann W. H., Kharaghani H., Orrick W. (2010), “On the Real Unbiased Hadamard Matrices”, University of Lethbridge, Lethbridge, Alberta, T1K 3M4, Canada, Department of Mathematics, Indiana University, Bloomington, Indiana 47405, USA.

[8]. Horadam K. J. (2006), “Hadamard Matrices and Their Applications”, Princeton University Press. [9]. Jennifer S., Beata J., Wysocki and Tadeusz A. (2003), “Williamson–Hadamard Spreading Sequences for DS-CDMA Applications”,

Wireless Communication Mobile Computer, University of Wollongong, NSW2522, Australia, pp. 597-607. [10]. Kazuki H., Yasuhiko K., Takashi K., Totaro I. (2003), “Hadamard Transform Microchip Electrophoresis Combined with Diode Laser

Fluorometry”, Anal. Chem., Vol. 75, N° 7, Kyushu University, Hakozaki, Fukuo ka, Japan, pp. 1765-1768. [11]. Kohl H., Depner H. (2010), “The Implementation of an Organizational Innovation: Examples of Mass Customizing Firms of the Capital

Goods Industry”, International Journal of Industrial Engineering and Management, Vol. 1, N°3, pp.85-95. [12]. Le-Bail A., K. Boumali, Jury V., Ben-Aissa F. (2009), “Impact of the Kinetic of Baking on Staling Rate and on the Mechanical Properties

of Crumb and degassed Crumb”, Journal of Cereal Sciences, Vol. 50, pp.235-240. [13]. Matthew. H., Petoukhov S. (2010), “The Genetic Code, Hadamard Matrices and Algebraic Biology”, Journal of Biological Systems, Vol.

18, Special Issue, pp. 159-175. [14]. Onoghojobi B. (2010), “Subsample Goal Model for Multivalve on Outliers”, Journal of Mathematics and Statistics, Vol. 6, pp. 347-349. [15]. Senk V., Delic V.D., Milosevic V.S. (1997), “A New Speech Scrambling Concept Based on Hadamard Matrices”, Signal Processing

Letters-IEEE, Vol. 4, pp. 161-163. [16]. Singh S., Singh M. K., Singh D. K., Sindri B. (2010), “Generalized Hadamard Matrices from Generalized Orthogonal Matrix”, Global

Journal of Computer Science and Technology, Vol. 10, pp. 22-30. [17]. Vivier S. (2002), “Optimization Strategies Using the Experimental Design Method and Application to Electrotechnical Devices Modeled

by Finite Elements”, Sciences and Technologies University, Lille, France, pp. 290-296.

84



Ms N° Ref. JJMIE-231-11 “Using Principal Component Analysis to Determine Key Factor of Rural Electrification Development Investment” by François A. Ravalison, Tokary A. Rajaoarisoa

Using Principal Component Analysis to Determine Key Factor of Rural Electrification Development Investment François A. Ravalison∗, Tokary A. Rajaoarisoa

University of Antananarivo, Ecole Supérieure Polytechnique, 101 Antananarivo-Madagascar Abstract The main objective of this study is to determine the key factors when decision on investment is required in rural electrification development. This study develops a process based on Principal Component Analysis to identify key factors. Principal Component Analysis is a data reduction method, that is, a method for reducing the number of variables. We have identified three variables related to investment in such context: State contribution, Bank contribution, Enterprise contribution and Commune contribution. We have extracted three factors: State contribution, Enterprise contribution and Commune contribution. In developing country and especially in the case of Madagascar, we have identified that State contribution plays a principal position in rural electrification development. The majority of rural electrification projects are Government contribution driven.

Keywords: Knowledge Management, Principal Component Analysis, Quality, Rural Electrification Development Investment,

1. Introduction

The majority of the Malagasy households live in inaccessible and scattered villages making the rural electrification through connection with the national network distribution complex. We know distributed energy systems, nowadays, are very hot subject all over the world [1]. Policy makers prefer decentralized energy solutions when connection with the national network is expensive. However, such preference requires hard technological solutions like the utilization of renewable energy. The primary advantage of the latter is its lack of greenhouse gas and other emissions in comparison with fossil fuel combustion [2]. The State policy is determinant and plays a great place as far as institutional and financial structure adoptions are suitable. Generally, the installation costs and production coupled with the faint need do not make profitable the offer of electric power. Related to this subject, we propose the following research question: what is the key factor when investment of a decentralized rural electrification project is concerned?

This study is structured as follows: a literature review on the general policy in the rural electrification program is presented. Then Principal Component Analysis is utilized to identify and visualize key factors. After this analysis, we can draw that State contribution in the investment holds a great part in the project. Then principal results will be listed. Furthermore, discussion will be conducted to answer the research question based on the results.

2. Literature review 2.1 Rural electrification development The decentralized rural electrification project is very expensive. Its operation cost is high but the rural people income is

low. That makes the rural electrification difficult and the private investors discouraged. In this case, the project must be seriously subsidized to be viable [3] and sustainable. The emergence of energy sector reform such as structural changes, privatization, and relative success of the public-private partnership have given rise to solve those problems. Some countries stimulate the adoption of public-private partnership [4]. According to the analysis in industrialized countries, they mainly carried out reforms to facilitate the competition between suppliers, to improve the quality of service and to lower the prices of electricity. Moreover, in developing countries, they conducted reforms based on needs [5]. The target is to improve the technical and the financial execution on the first hand, and to support the investment in the system development and to reduce the policy interference in the tariff arrangement and the service management on the second hand [5]. In other terms, these reforms are based on market theories by which electricity is treated as product in opposition to old vision as an integrated service [6]. In these reforms, it is advisable to consider the difference between the entity offering the energy service and the owner of infrastructure. The ownership of infrastructure can has four possible cases:

- A public company ensured the electrical service. It is the most current form of the property, where the State must give an account of its management within the framework of contract-plan or contract-management.

∗ Corresponding author: [email protected]

JJMIE

85

- The State is owner, but the running operation is sub-contracted to the private sector. - The private service is owner of the infrastructure but its activity is regulated. - The Community ensures the service. This solution is privileged when the public service does not allow the

satisfaction of such needs.

Our study is focused on the second case. The initial idea is to find a solution that encourages the investors to create a true appropriate business model related to profitable investments through security by institutions [7]. Then the State contribution concerns equipments. The private owner assures installation of these equipments. However, it is obvious that the public funding will continue to play an important part in rural electrification as far as the aim is to increase the accessibility rate of the rural people to electricity [8].

Besides, subscribers and technologies also play an important part in rural electrification. The financial assistance allocated for a project and the estimated numbers of subscribers are decisive factors the owners consider during invitation to tender. Choice of technology is important to determine the costs of investments and running operations [9]. At present, the evolution of renewable energy technology makes possible supporting the promotion of decentralized rural electrification policy [10]. More the environmental advantage brought by this technology is even more competitive, profitable and sustainable compared to other energy technology such as fossil energy [11]. Utilizable in these technologies are the photovoltaic solar system, the wind turbine, the hydroelectric and the gasification of biomass. Knowing that certain technology is expensive than the others for the same produced power, but it could be reconcilable with the processing site [12].

Indeed, rural electrification is complex without government intervening more or less energetically, because rural electrification is an expensive and unprofitable operation [13]. The developed countries knew a variety of institutional trajectories influenced by the institutional environment of each country. Moreover, electrification has found different translation according to countries. It clarifies the question of the complexity of electrification in the developing countries [14]. Among the biggest challenges is the regulatory design for achieving political independence and introducing rules to ensure accountability. There is a long way to follow in the best practices evolution for contracting, operating, and regulating rural electrification service concessions and in understanding the most appropriate standard of concessions in different contexts [15]. It seems whereas the speed with which electricity progressed in the villages the more the government intervenes, the more the rural people profit early from the electric service [16].

2.2 Principal Component Analysis (PCA) The PCA is a method for multiple variables analysis, which seeks to identify the principal axes explaining the best

correlation between descriptive variables [17] [18] [19]. These variables are characterized in a data matrix. The data processed by this method are presented in the statistical table. Let us note that a data table can be compared to a rectangular matrix. This matrix describes two different spaces as a vector space. A reading on line allows seeing the vectors where each element gives information on each variable for a given individual. A reading in column describes the variable vectors where each element gives the variation of the variable according to individuals of the sample.

In this table, the first column defines the individuals’ identity. In order to obtain relevant results, it is necessary to analyze these individuals by eliminating the exceptional cases. The other columns of the table give the various variables that characterize the individuals. The following matrix -. is a table of data with N observations and / variables:

01 � 23334"55 "5� … "57"�5 "�� … "�78 8 8 88 8 8 8"95 "9� … "97:;;

;< &1)

"�� : indicate the variable >?@ line and the A?@column. For an initial data matrix with a dimension (N;k), the aims in this method are:

• To minimize the cloud points deformations, in the individuals space BCwith a dimension C &C D E). • To explain the best initial connections between these variables in the space FC. • To reduce the dimension of the initial matrix by a matrix of row C &C D E).

The PCA interests the variables study in the table of information to confront the various distributions and to discover

irregularities in these distributions. Besides, it proposes to analyze interrelationships between the variables, and to highlight the combinations between the variables.

86

The first stage of the method is the data matrix transformation into square matrix variance-covariance or correlation matrix. Since this method objective is to reduce the variables in order to determine the factorial axes or factors followed by a connection analysis or correlation between these variables. The statistical indicators that make this analysis possible are the variance covariance matrix GHI&J)and the correlation matrix GHK&J). For this matrix, the statistical concepts allow to write:

The variable average described in LMN column of the matrix -..

"O.� � 1Q � ".�9

�RS &2)

The variables covariance U.V et U.W: XYZ[".� ; ".]^ � 1Q �&"�� ' "O.�)&"�] ' "O.])9

�RS &3) The variance of U .V:

Z`a[".�^ � 1Q �["�� ' "O.�^�9�RS &4)

The standard deviation is the root of the variance:

c.� � d1Q �["�� ' "O.�^�9�RS &5)

Let us transpose these formulas in the matrix case. We notice that the difference fgL ' fh.L explains the variation between the observation and the average value. The first transformation then consists in forming a square matrix ij such as:

Mlm �23334"55 ' "O .5 "5� ' "O.� … "57 ' "O.7"�5 ' "O.5 "�� ' "O.� … "�7 ' "O.78 8 8 88 8 8 8"95 ' "O.5 "9� ' "O.� … "97 ' "O.7:;;

;< &6)

As well as we have: cov&X) � 1N Mr ls Mlm &7)

cor&X) � u XYZ&".� ; ".])c.�c.] v &8)

Then, the second stage consists in determining the principal results of the method. Total inertia is one of them. Inertia measures the total cloud point dispersion. The principle is to obtain an under-space with a dimension on which projection resembles more with the initial cloud. In distance terms, in a vector space / dimension, the vector projection inertia in this under space x dimension lower than / is:

�&yz) � 1Q �&%&"�; yz))�9�R5 &9)

However, in algebraic term, the factorial plan is built starting from the resolution of an optimization problem. This plan results from the inertia projected cloud maximization. While passing by the theorem of the spectral decomposition, the under space B| is the under vector space generated by the vectors }~; … ; }x associated with the eigenvalues �~; … �x of the matrix GHI&J) or the matrix GHK&J).

The next result is the information-selected share. We carry out a change of reference in B| in order to place a new representation system where the first axis as much as possible brings total inertia cloud. The second axis, as much as possible, centers inertia, which is not taken into account by the first axis, and so on. Therefore, as soon as we manage to determine the under space B| we have all the factors. It may be that these factors are more than two. Then we consider only the plan generated by the eigenvectors associated by two greater Eigen values. When we consider more than two factors, we can couple these factors. The criterion to choose the factors is the following, which represents the information-selected share in the matrix / column after under space projection of dimension x lower than / if factors x are considered:

87

∑ ��R5∑ ��7�R5 &10)

To obtain a plane representation of the individuals, it is necessary to build an orthogonal reference mark equipped with a standard value whose unit starts from the eigenvectors associated by two greater Eigen values. The cloud associated with the matrix is the orthogonal projection of each point. The coordinates of these projected points are called principal

component noted �gL. We can characterize those principal components. Let us consider the individual g located by his square cosine

compared to the factorial axis. While noting � the angle between the reference mark axis and the origin and the coordinate’s individual distance, we have: XY�� ∑ &X��)��R5∑ &X��)�7�R5 &11)

Besides, each individual has his contribution to the factorial axis formation. This indicator allows locating the aberrant value in the observations series. Indeed, the g individual contribution to the L axis formation is: 1Q X�� &12)

The orthogonal vectors define the factorial axes. In addition, orthogonal vectors are associated with the eigenvectors corresponding to the matrices GHI&J) or GHK&J) Eigen values. The variables projection on the factorial axis allows identifying the factorial axis significance. In other term, when we project the centered reduced variables on the factorial axis F/, the projections belong to a circle of ray 1. The best representation is when the point is close to the circle. Therefore, the factorial charts that we have obtained previously concern only the individuals; that is to say variables. By using isometric transformation plan, we can simultaneously represent the individuals and the variables in only one factorial chart.

3. Methodology

To conduct the research, we have implemented the process below:

Figure 1: Research Process

We have collected data in 28 rural communes. The questions concern rural electrification. Then we have extracted data related to investment. We have applied Principal Component Analysis to those data.

4. Results

Collect data related to rural electrification

Extract data related to investment

Apply Principal Component Analysis

Elaborate an investment model

88

When PCA is applied to the data table with Pearson Correlation Coefficient (variance with 1/n). We have obtained the correlation matrix in table 1.

Table 1: Correlation matrix

Government Enterprise Community Bank

Government 1 0.716 -0.082 -0.114

Enterprise 0.716 1 -0.100 -0.072

Community -0.082 -0.100 1 -0.057

Bank -0.114 -0.072 -0.057 1

In bold, significant values (0.716) at the level of significance alpha=0.050 (two-tailed test) are obtained. Government and enterprise are correlated.

Then we consider the plan generated by the eigenvectors associated by two greater Eigen values. The table 2 below shows those Eigen vectors.

Table 2: Eigen vectors

F1 F2 F3 F4

Government 0.691 0.025 0.142 0.708

Enterprise 0.688 -0.029 0.170 -0.705

Community -0.154 0.707 0.690 -0.013

Bank -0.158 -0.706 0.689 0.041

In axis F1, Government and enterprise Eigen vectors are respectively 0.691 and 0.688. We notice that Government Eigen value is slightly superior to enterprise one.

By using isometric transformation plan, the following figure 2 represents the individuals and the variables in only one factorial chart.

Figure 2: Principal component analysis

In axis F1, we observe that Government contribution and enterprise contribution are correlated. The following table 3 shows the contributions of the variables.

0 Government0 Enterprise0

Community

0

Bank

-1

-0,5

0

0,5

1

-1 -0,5 0 0,5 1

--ax

is F

2 (

26.0

0 %

) --

>

-- axis F1 (44.00 %) -->


89

Table 3: Contributions of the variables (%)

F1 F2 F3 F4

Government 47.759 0.062 2.012 50.167

Enterprise 47.379 0.086 2.888 49.647

Community 2.364 49.992 47.627 0.017

Bank 2.498 49.860 47.473 0.169

At F1 axis level, contributions of Government and Enterprise are respectively 47.75% and 47.37%.

5. Discussion

We have four variables or four types of financial contributions in this study: Government, Enterprise, Community and Bank. In the correlation matrix, the linear relation between the Government and Enterprise variables are significant. The negative sign of the linear correlation coefficient between these variables results in the variation, which is done in the opposite direction. In other words, if the Government financial contribution increases, this involves a reduction in the Enterprise financial contribution and conversely. In the same way when the Commune financial contribution decreases then the Government financial contribution increases.

Concerning the variables, we notice the three following financial contributions: Government, Enterprise and Community that represent the best in the factorial chart formed by the F1 and F2 axes because they are in the quadrants [0,1]. Then, the position of each variable justifies the complementarities between these values: the variables Community and Government are in quadrant [0,1] of F2 axis. In addition, variables Government and Enterprise are in quadrant [0,1] of F1 axis. Financial contributions of Government influence financial contributions of Community and Enterprise. We also view this connection between the Community and Enterprise variables with the Government variable such as the coefficient correlation preceded by a negative sign. Lastly, the variables square cosine compared to the factorial chart axes shows that the Government variable dominates the F1 axis while the Community variable orders the F2 axis. Thus, the F2 positive part shows the strong Community contribution and the negative part proves the importance of Enterprise contribution. In short, the Government share in a rural electrification project investment is dominant.

According to the literature review, Government should bring more to carry out a rural electrification project. The results above confirm that Government part occupies an important place in the corresponding investment. Despite the initiative takes by Government, access to rural electrification is still in a low level. That is why the vision «energy for all» has become all developing countries’ vision. The cost of the project is the principal problem. For the case that we are dealing here, private enterprise holds the operation contract with the Government in order to manage the electrical energy supply of a Community. Of course, after analyzing the business plan of those enterprises we can deduce that their part on investment associated with their operation cost influence the bills the subscribers have to pay. Thus, the rural electrification achievement is shaped by a triangle enterprise profit/populations’ ability to pay/capacity of grantor authority to subsidize investment. However, given that enterprise is looking for an acceptable profit, considering also that electrical service costs highly exceed population abilities to pay, grantor authority generally agree to allocate financial support to investment so that it can merge users and enterprise’s interests. A new form of partnership needs innovation process so that project will be viable for all stakeholders.

Nevertheless, prediction of electricity consumption plays an important role in future energy related research and planning strategies [20]. What is more, in an adequate model of partnership, Enterprise will always stay directing the work and running the equipment during the contract period. This idea is far being satisfactory to stop the profit of private operators but interesting in the competitiveness sector. However, a unit should be set up to analyze activities and local opportunities for the eventual program using at the first time an electrical energy. It will be in charge of the project launching and of different missions, as well as the selling of the energy and the promotion of the program linked to a social, economical, or environmental project. That is why it will have the idea and the instruments of production modernization in rural field. Then it will play the interface between users and Government. In this case, it can be the new partner in the investment process. This idea will reduce fixed cost effect on price. Because price is an important factor although capacity utilization is more important in electricity prediction [20]. Moreover, technological choice is an influential factor in investment.

6. Conclusion

Rural electrification project faces many problems because fossil resources are distinctly expensive [21]. Besides, distributed energy systems are very hot subject all over the world [1] and above all in Madagascar. Government has to

90

intervene so that rural population accessibility to electricity is feasible and effective. Such intervention should be coupled with enterprise’s intervention as far as running operation is concerned.

Bank contribution is relatively impossible. Bank process is not suitable for rural context, rate of return is low and investment return takes a long time. So rarely, rural electrification projects are bank financed. In the factorial chart, Bank contribution is in negative quadrant for both F1 and F2.

Considering the above problems, the movement from small, pilot, and experimental renewable energy projects is a must and inevitable [2]. Alsaqoor and al. confirm such concept when they say the challenge is on extensive use of renewable sources [1].

There are some domains for further research when rural electrification development is concerned: a. How could Community become a key factor in electrification development? b. What process are we going to implement to popularize the use of renewable sources?

References [1] S. Alsaqoor, M. AlAjlouni, K. AlQdah, H. Kruczek, E. Pelinska-Olko, “Development Chances of Distributed

Energy Production on Small Scale”, Jordan Journal of Mechanical and Industrial Engineering, Vol.4, N°1, 2010, 135-142

[2] M. Al zou’bi, “Renewable Energy Potential and Characteristics in Jordan”, Jordan Journal of Mechanical and Industrial Engineering, Vol.4, N°1, 2010, 45-48

[3] Davis M., "Institutional Frameworks for Electricity Supply to Rural Communities", EDRC, 1995 [4] N. Wamukonya, "Power Sector Reforms in Developing Countries: Mismatched Agendas", in Wamukonya Njeri,

eds. Electricity Reforms: Social and Environmental Challenges, UNEP Collaborating Centre on Energy and Development, Roskilde, Denmark, 2003

[5] Bhagavan M.R. Reforming the Power Sector in Africa, African Energy Policy Research Network (AFREPREN), Zed Books Limited, London, United Kingdom & New York, USA, 1999

[6] J. Byrne, Y. Mun, “Rethinking Reform in the Electricity Sector: Power Liberalization or Energy Transformation”, in Njeri Wamukonya, eds. Electric Reform: Social and Environmental Challenges, UNEP Riso Centre, Roskilde, Denmark, 2003

[7] A. Jadresic, “Promoting Investment in Rural Electrification, the Case of Chile”, the World Bank Viewpoint, N° 214, 2000

[8] M. Nangammbi, L. Mosomane, “Rural Electrification In The Energy Policies Of The SADC Region”, Sustainable Energy Futures Group-Natural Resources And The Environment, 2011

[9] E. Martinot, R. Kilian, “Regulatory Approaches to Rural Electrification and Renewable Energy: Case Studies from Six Developing Countries”, the World Bank-Washington DC, Working Paper, June 2000

[10] Haanyika M.C.; "Rural Electrification Policy And Institutions In A Reforming Power Sector"; In Energy Policy Journal, Elsevier Ltd; October 2005

[11] S. Kaufman, “Rural Electrification with Solar Energy as a Climate Protection Strategy”, Renewable Energy Policy Project Research Report No. 9, 2000

[12] ESMAP, Energy and Development Report 2000, Energy for the World’s Poor, the World Bank, Washington DC, 2000

[13] Brown D.C. Electricity for Rural America: the Fight for REA, Westport, Greenwood Press, 1980 [14] Finon D. La Variété des Trajectoires Institutionnelles d’Electrification, Working paper CIRED (Centre

international de recherche sur l’environnement et le développement), 2006 [15] A. Estache, “Designing Regulatory Institutions for Infrastructure Lessons from Argentina”, the World Bank

Public Policy Note N°114, 1997 [16] Y. Tremblay, “Les Etapes de l’Electrification Rurale dans l’Ouest Canadien, 1920-1960”, Cahiers franco-

canadiens de l’ouest, Vol.7, N°1, 1995, 37-67 [17] Jaeon K., Mueller C.W, “Introduction to Factor analysis: What it is and how to do it”, Beverly Hills, Sage

Publications, 1978, 80-160 [18] J. Shlens, “A Tutorial on Principal Component Analysis”, Systems Neurobiology Laboratory, Salk Institute for

Biological Studies and Institute for Nonlinear Science, University of California, San Diego, Version 2, 2005 [19] Steven M. H. Principal Components Analysis, University of Georgia, Athens, 2008 [20] A. Al-Ghandoor, M. Samhouri, “Electricity Consumption in the Industrial Sector of Jordan: Application of

Multivariate Linear Regression and Adaptive Neuro-Fuzzy Techniques”, Jordan Journal of Mechanical and Industrial Engineering, Vol.3, N°1, 2009, 69-76

[21] O. Badran, E. Abdulhadi, R. Mamlook, “Evaluation of Solar Electric Power”, Jordan Journal of Mechanical and Industrial Engineering, Vol.4, N°1, 2010, 121-128

91



Ms. No. Ref. JJMIE-24-12 “Quality Control Mapping to Assess Quality Assurance Performance in Textile and Garment Factories BY François A. Ravalison, Zo Rakotomalala

Quality Control Mapping to Assess Quality Assurance Performance in Textile and Garment Factories

François A. Ravalison∗, Zo Rakotomalala Ecole Supérieure Polytechnique-University of Antananarivo-Madagascar

Abstract In the Jordan Journal of Mechanical and Industrial Engineering, Volume 4, N°6, 2010, Fouad and Mukatta sh have worked on “Statistical Process Control Tools: A Practical Guide for Jordanian Industrial Organizations”. The purpose of this paper is to contribute to such Guide by identifying another measure of quality assurance or quality performance. We have collected data from two exporting factories, textile and garment. The data mining is conducted to identify a measure or a mapping. Result reveals and proposes a trapezoid, which permits to map quality assurance performance. There are a few limitations for this paper. First, only cases with work and rework operations are concerned. Second, this paper is confined only to answer if quality assurance is correct or not. This paper serves as an alert for factories to review their quality assurance programs and to conduct cause and effect assessments on a regular basis in order to correct the process. It shows a new important tool which is not studied in many past researches. This paper also looks at quality assurance practices from the textile and garment industries where quality process is usually and rigorously conducted. Keywords : Garment, Jordan, Madagascar, Quality Assurance, Quality Performance, Textile,

1. Introduction

In the Jordan Journal of Mechanical and Industrial Engineering, Volume 4, N°6, 2010, Fouad and Mukatta sh have worked on “Statistical Process Control Tools: A Practical Guide for Jordanian Industrial Organizations” [1]. The authors have cited and utilized seven tools in a case study on Jordanian Steel Manufacturing Company. These tools are Pareto Diagram, Cause and Effect Diagram, Check Sheets, Process Flow Diagram, Scatter Diagram, Histogram and Control Chart. In other domain of research, Al-Shobaki, Fouad and Al-Bashir have studied implementation of Total Quality Management for the Banking Sector in Jordan [2]. However, Jordan is leader among Arab countries in textile and garment, is there a tool to well map and to evaluate quality assurance performance? In this research, we are going to propose another tool, especially for Jordan textile and garment industries.

2. Why Jordan garment and textile sector?

Jordan garment industry plays a principal part in Jordan industry sector. It contributes in direct foreign and local investment for about US$ 700 million [3]. Moreover, export to the United States or US market is US$ 770 million according to Al Khoja [3]. Importing country is not only the US but also the European Union countries or EU. Because the Jordan export target market is the US and the EU countries [4]. Their imports from Jordan are increasing because free of duty and quota [4].

Niche is not only garment but also textile. A US International Trade Commission study predicted that Jordanian exports would increase measurably in only one sector, textile and apparel [5]. Besides, many agreements make Jordan a strong competitor in the world garment and textile market [6]. These agreements are: General Agreement on Tariffs and Trade or GATT, Jordan-European Union Association Agreement and the General Agreement or Services or GATS, and Jordan-United States Free Trade Agreement.

As a result, textile and garment industries are important. Over 55,000 people are employed in the sector, which is about 1/3 of Jordan’s industrial labor force [6]. In addition, over 30% of Jordan’s total exports are from textile and garment manufacturing sector.

∗ Corresponding author: [email protected]

JJMIE

92

Al Khoja has noted that the future of garment and textile sector needs “Research and Development Capabilities” [3]. He has even specified that improving the enterprises’ productivity pass through quality management [3]. Practically, some training needs have been identified [4]:

- 900 middle level staff in quality control, - Create a Jordanian capacity in quality management - Train senior level staff in quality assurance management

Moreover, concerning program upgrading, “Quality Enhancement Program” is a principal axis [4].

3. Methodology

We have conducted research in two factories: textile and garment. We have collected data during 46 and 32 weeks respectively for textile and garment factories. And the research processes are presented by the following.

Figure1: Research processes

Each week, for the textile factory, we have collected global production (Pg), production that conforms to quality requirements (Pq) and non-quality (NQ1). Then after re-work of non-quality (NQ1), we have collected data that conforms to quality requirements (Pq2) and non-quality (NQ2). The defect rate after first work is (NQ1)/(Pg). And, the defect rate after re-work is (NQ2)/(NQ1).

At the same time, for garment factory at first re-work level, we have collected production (P), quality production (Pq) and non-quality production (NQ1). Then after re-work of non-quality production (NQ1), we have collected quality production (Pq’) and non-quality (NQ2). The defect rate after first re-work is (NQ1)/(P). And the defect rate after second re-work is (NQ2)/(NQ1).

4. Results

On the first hand, the textile case is observed. And the fig.1 shows the corresponding mapping.

CALCULATE DEFECT RATE AFTER FIRST WORK

COLLECT SECOND QUALI-TY PRODUCTION AND SE-COND NON-QUALITY PRO-DUCTION AFTER RE-WORK

CALCULATE DEFECT RATE AFTER RE-WORK

COLLECT GLOBAL PRODUCTION, QUALITY PRODUCTION AND NON-QUALITY PRODUCTION

IN EXCEL INSERT A CORRESPONDING LINE

CHART OF THE TWO DEFECT RATES

IN EXCEL TOOLBAR “DESIGN”, SWAP THE DATA

OVER THE AXIS

TEXTILE FACTORY

CALCULATE DEFECT RATE AFTER FIRST RE-WORK

COLLECT SECOND QUALI-TY PRODUCTION AND SE-COND NON-QUALITY PRO-DUCTION

AFTER SECOND RE-WORK

CALCULATE DEFECT RATE AFTER SECOND RE-WORK

COLLECT PRODUCTION, QUALITY PRODUCTION AND NON-QUALITY PRODUCTION

AFTER FIRST RE-WORK

INSERT A CORRESPONDING LINE CHART OF THE TWO

DEFECT RATES

IN EXCEL TOOLBAR “DESIGN”, SWAP THE DATA

OVER THE AXIS

GARMENT FACTORY

Figure

We notice a trapezoid ABCD. AB depicts the maximum defect rate of 37%. CD defect rate of 3%. The latter is the target defines in the quality assurance policy. The computer automatically determines the distance AD. In the present case, all defect rates after rewith quality assurance target of 3%. 50% of defect rates after readdition, 50% are merely the same as those after first work.

On the second hand, the garment case is studied. The followin

Figure

In the garment case, we also perceive the trapezoid ABCD. AB gives a picture of defect rates after first rework. And CD represents 6%, the maximum defect ratecomputer automatically determines the distance AD.

After first re-work, the majority of defect rates are included in [6%,23%]. And after second repart of defect rates are included in [0%,6%].

0

5

10

15

20

25

30

35

40

Defect Rates after 1st re

De

fect

Rat

es

[%]

Defect Rates: after first re

93

Figure 1: Defect Rates Mapping of the Textile Case

We notice a trapezoid ABCD. AB depicts the maximum defect rate of 37%. CD defect rate of 3%. The latter is the target defines in the quality assurance policy. The computer automatically

In the present case, all defect rates after re-work are over CD. Practice in the textile factwith quality assurance target of 3%. 50% of defect rates after re-work are poorer than those after first work. In addition, 50% are merely the same as those after first work.

On the second hand, the garment case is studied. The following fig.2 shows the corresponding mapping.

Figure 2: Defect Rate Mapping of the Garment Case

In the garment case, we also perceive the trapezoid ABCD. AB gives a picture of defect rates after first rework. And CD represents 6%, the maximum defect rate acceptable in the assurance quality policy. The computer automatically determines the distance AD.

work, the majority of defect rates are included in [6%,23%]. And after second repart of defect rates are included in [0%,6%].

Defect Rates after 1st re-work Defect Rates after 2nd re

Defect Rates: after first re-work vs. after second re-work

A

B

C

D

We notice a trapezoid ABCD. AB depicts the maximum defect rate of 37%. CD represents the minimum defect rate of 3%. The latter is the target defines in the quality assurance policy. The computer automatically

work are over CD. Practice in the textile factory does not comply work are poorer than those after first work. In

g fig.2 shows the corresponding mapping.

In the garment case, we also perceive the trapezoid ABCD. AB gives a picture of defect rates after first re-acceptable in the assurance quality policy. The

work, the majority of defect rates are included in [6%,23%]. And after second re-work, the best

Defect Rates after 2nd re-work

work

94

5. Discussion

There is a tool to map and to evaluate quality assurance performance. It is the quality trapezoid. The results show that it is a new tool. It has some strengths and limits in comparison with control chart developed by Fouad and Mukattash [1].

Quality trapezoid and control chart have two things in common. Both have measurable variables. For quality trapezoid, they are time in abscissa and defect rate in ordinate. In addition, both are line chart. The origin of fig. 2 and 3 above is line chart. So why control chart and quality trapezoid are line chart. This constitutes strength for the quality trapezoid as far as it shows that the new tool belongs to the group of the Statistical Process Control Tools [1].

If control chart needs to be used in pairs [1], quality trapezoid is fully significant with one chart. It is quality trapezoid’s strength. Two positions, defect rates after first re-work and defect rates after second re-work in the garment case, may be observed in one chart and so make easy assessment. In the textile case, a same chart depicts two positions; defect rates after first work and defect rates after re-work. With quality trapezoid, we can report on quality assurance. Quantitative assessment is practically possible. It is again quality trapezoid’s strength.

However, quality trapezoid cannot mention the causes of quality assurance fails. This is a limit. In quality management “causes” are important not only for lessons learned but also for solving the problems. In quality trapezoid, we just see the problem but we cannot solve it because we do not know its causes. Control chart can be used as dashboard and our quality trapezoid can be used as performance evaluation.

Related to the Fuzzy Approach to Quality Evaluation Processes [7], we use continuous scale in ordinate in terms of defect rates. Besides, the above figures express positions that are not extremes but intermediate, revealing organizations present evolution in terms of quality assurance. Therefore, our findings fit the Fuzzy Approach.

6. Conclusion and further researches

A new control chart named by “Quality Trapezoid” is a new finding. It permits to map quality control and to assess quality assurance performance. It has some advantages related to control chart: easy in mapping and easy in assessing quality assurance performance. However, it cannot point out the causes of non-quality. Those causes are necessary for improvement. That limit of Quality Trapezoid generates some future researches:

- Coupling quality trapezoid and cause and effect diagram to better assess quality assurance, - Implementation of quality trapezoid and Intelligent Quality Function Deployment into computerized

environment as developed by Daws and al. [8].

References [1]. R.H. Fouad, A. Mukattash, “Statistical Process Control Tools: A Practical Guide for Jordanian Industrial

Organizations”, Jordan Journal of Mechanical and Industrial Engineering”, Vol.4, N°6, 2010, 693-700. [2]. S. D. Al-Shobaki, R.H. Fouad, A. Al-Bashir, “The Implementation of Total Quality Management (TQM) for

the Banking Sector in Jordan”, Jordan Journal of Mechanical and Industrial Engineering”, Vol.4, N°2, 2 010, 304-313.

[3]. Al Khoja R. in “Garment Sector in Jordan: Facts and Figures”, http://www.gsc-jo.com accessed January 4, 2012.

[4]. EJADA, “Strategy and Action Plan to Enhance the Competitiveness of the Garment Industry in Jordan”, 2010.

[5]. “A US-Jordan Free Trade Agreement Would Have No Measurable Impact on US Production or US Employment”, International Trade Commission, News Release 00-122, September 26, 2000, http://www.usitc.gov/er/nl2000/ER0926X1.HTM accessed January 4, 2012.

[6]. “Jordan: Rising Demand For Indian MMF Textile Exports”, Synthetic & Rayon Textiles Export Promotion Council (SRTEPC), Vol.25, N°23, 2010.

[7]. E. P. Paladini, “A Fuzzy Model for Quality Evaluation Processes”, International Journal of Computers, Vol.1, Iss.4, 2007.

[8]. K. Daws, Z. A. Ahmed, A. A. Moosa, “An Intelligent Quality Function Deployment (IQFD) for Manufacturing Process Environment”, Vol.3, N°1, 2009, 23-30.

95

CHAPITRE 2 : SYNTHÈSE DES ACTIVITÉS DE RECHERCHE

INTRODUCTION

Les travaux de recherche constituent un des produits finis d’une université ou d’un laboratoire de

recherche. Dans le contexte globalisé actuellement, ces travaux de recherche sont appelés à

répondre aux besoins nationaux face à une compétition globale qui devient de plus en plus inévitable

et qui menace les entreprises nationales [27]. En plus, cette compétition globale est confortée d’un

côté par notre appel aux Investisseurs Directs Etrangers ou IDE [32]. De l’autre côté, les différents

accords normalisant et facilitant les transactions tels que la Common Market of Eastern and Southern

Africa ou COMESA, la Southern African Development Community ou SADC et la Commission de

l’Océan Indien ou COI ne facilitent pas le pilotage de l’économie Malagasy [27][31]. Ceci est ramifié

sous l’Organisation Mondiale du Commerce ou OMC qui véhicule le « Consensus de Washington »

[27]. Le retour à la prééminence de la technologie a été évoqué [30]. Et il a été explicité que la

technologie constitue un facteur clé pour le redémarrage de l’industrialisation donc pour faire face à la

globalisation.

Dans ce système influencé par la globalisation, les entreprises n’ont pas engagé une préparation

sérieuse [28]. Elles deviennent vulnérables face à la compétitivité [28]. Et elles ne sont plus

compétitives face aux industries régionales [28]. Au rythme actuel, on risque de retrouver l’économie

nationale au même format qu’avant 1972 « exportateur de matières premières et importateur de

produits finis ». En quoi les produits de recherche, cités plus haut, contribuent-ils à l’amélioration du

système actuel ? Existe-t-il un modèle de système d’ingénierie pour synthétiser les théories

développées dans chaque produit de recherche ? Pour ce faire, la démarche globale suivante sera

adoptée pour synthétiser les travaux de recherche et pour proposer un modèle.

Figure 1: Démarche globale

Dans un premier temps, les publications internationales et les mémoires encadrés de Diplôme

d’Etudes Approfondies ont été collectés. Ensuite, ces produits ont été synthétisés en utilisant la

SCIENTOMETRIE CONCEPT SYSTEME

PUBLICATIONS INTERNATIONALES

ENCADREMENT DE MEMOIRES DE DEA

COLLECTER

SYNTHETISER CES PUBLICATIONS

INTERNATIONALES ELABORER UN

MODELE SYNTHETISER CES ENCADREMENTS DE MEMOIRE DEA

96

scientométrie. Et par la suite un modèle a été élaboré à partir de la précédente synthèse et en

s’inspirant du concept système.

MATERIELS

Le matériel d’étude est constitué par les publications internationales et les mémoires encadrés de

DEA. Ils sont présentés dans les tableaux suivants

Tableau 1 : Les publications internationales et les encadrements DEA

CODE PUBLICATIONS INTERNATIONALES

ENCADREMENT DEA CODE

P1

Assessment of Technology Management in a Context of Sustainable Development-The Case of a Paper Mill Industry in a Developing Country

Démarche qualité utilisant la méthode des plans d’expériences pour améliorer la conservation du pain

E1

P2 Are Converging Technologies Tools of Competitiveness?-The Case of a Paper Mill Industry in Madagascar

Adéquation de programme de formation professionnelle et technique aux besoins des entreprises

E2

P3 How does Reengineering Sustain Economy?-The Case of a Paper Mill Industry in Developing Country

Management qualité vers l’innovation de la gestion d’un établissement d’enseignement technique et de la formation professionnelle

E3

P4 Developing a Mathematical Concept and Process Technologies to Accompany Firms from Rut to Change

Veille Technologique et TRIZ pour un processus d’innovation qualité dans la téléphonie mobile

E4

P5 Towards “Fair Globalization”: Critical Success Factors for Partnering Project of Development and Enterprise Project

Révision du système de production par l’adaptation de la démarche Juste-A-Temps aux contraintes industrielles

E5

P6 Using Patent Statistics and Principal Component Analysis to Predict Global Competition

Amélioration de l’accessibilité à l’eau potable par l’Analyse de la Valeur et la TRIZ

E6

P7

Critical Success Factors are the Essence to Improve Productivity: A Technical Note on the Article “Jig Design, Assembly Line Design and Work Station Design and their Effect to Productivity”

Amélioration de l’accessibilité des ruraux à l’énergie électrique par l’Analyse de la Valeur E7

P8 Hadamard Matrix to Improve Enterprise’s Activities: An Exploratory Research

Processus géomatique pour la localisation des zones à risques de feux de brousse

E8

P9 Using Principal Component Analysis to Determine Key Factor of Rural Electrification Development Investment

Méthode de planification via l’aide à la décision multicritères E9

P10 Quality Control Mapping to Assess Quality Assurance Performance in Textile and Garment Factories

La visibilité de ces travaux de recherche ainsi que les technologies y développées seront évaluées.

Les évaluations vont se porter sur les indicateurs scientométriques : degré de collaboration, revue

97

des pairs, occurrence des mots clés [1][5][14][16][40] pour les publications internationales. Et

l’indicateur scientométrique, l’occurrence des mots clés, va être utilisé pour le cas des mémoires de

DEA encadrés.

Scientométrie

La scientométrie est une branche de la science [21]. C’est une science et un instrument fondamental

pour l’évaluation statistique des résultats de recherche scientifiques [33]. Il y a une similarité entre

bibliométrie et scientométrie. En fait, les deux termes ont été introduits, presque simultanément, par

Pritchard, Nalimov et Mulchenko en 1969. La bibliométrie est définie comme l’application des

mathématiques et statistiques sur les ouvrages publiés [24]. Quant à la scientométrie, Nalimov et

Mulchenko l’ont définie comme l’application de ces méthodes quantitatives dans l’analyse de la

science dans un processus d’information. Donc la scientométrie analyse les aspects quantitatifs pour

l’analyse de la science.

L’analyse scientométrique permet, en effet, d’évaluer la performance scientifique d’un travail de

recherche [23]. Elle est basée essentiellement sur les nombres de publications et citations [23]. Et

c’est sur ces nombres de publications et citations que vont porter l’analyse de la science par

l’application des méthodes quantitatives. Et ici, la similarité entre bibliométrie et scientométrie est très

rapprochée.

Degré de Collaboration

Une collaboration scientifique peut être définie comme une interaction de deux ou plusieurs

scientifiques qui facilite le partage de connaissances et l’accomplissement de tâches vers un but

commun [14]. Les raisons de collaboration sont multiples [12] et on peut résumer que telle

collaboration est une démarche qualité intégrée et convergente [30] dans le processus de recherche

scientifique. Et ceci est confirmé par un point de vue scientifique : la disponibilité d’expertise

additionnelle améliore le succès et la convergence de plusieurs personnes et garantie la qualité des

résultats [2].

Le co-auteur est une mesure importante de la collaboration scientifique [10]. Et des fois on parle de

réseau de collaboration d’auteurs [22]. En effet dans un tel réseau, il y a collaboration entre deux

auteurs s’il y a au moins un article en co-auteur [14]. Et pour mesurer cette collaboration, une

approche quantitative de Subramanyam permet de déterminer le degré de collaboration [20] :

C � NMNM � NS

98

Où C est le degré de collaboration, NM est le nombre de publications en co-auteurs et NS est le

nombre de publications en mono-auteur.

Le degré de collaboration peut être aussi observé sous la collaboration internationale. Cette dernière

est calculée par le % d’articles en co-auteurs en provenance de pays étrangers [35]. On peut

proposer la formule suivante pour la calculer :

CI � NENA � 100

Où CI est la collaboration internationale, NE est le nombre d’auteurs étrangers et NA est le nombre

total d’auteurs.

Revue des Pairs

Un pair est une personne experte dans un domaine. Une revue des pairs est nécessaire pour juger la

recherche et pour évaluer le groupe de chercheurs [16]. La revue des pairs est une évaluation

qualitative de performance de recherche [16]. Quand c’est qualitatif, la tendance vers la subjectivité

constitue un problème majeur où les opinions des experts peuvent être influencées par des éléments

négatifs [39].

Occurrence des mots clés

Une structure est un ensemble de composantes ou d’entités liées entre elles. En effet, une structure

est identique à un système [28] qui est lui-même une somme de processus [32]. Donc, une structure

définit un système [38]. Ce système peut être élaboré à partir de la méthodologie FAST [28]. Dans la

modélisation FAST, la structure ou le système est composé de réseau de fonctions de base et de

réseau de fonctions de soutien. Et chaque réseau comprend un réseau d’acteurs et un réseau de

liens correspondant à des composantes et des interrelations entre composantes [38].

Si les réseaux sont bien définis, la structure ou le système de connaissance ou technologie pourrait

être identifiée. Une récente investigation a permis d’extraire une structure de connaissance ou de

technologie à partir d’application de théorie de réseau sur des publications [18]. En effet, la structure

de technologie peut être obtenue à partir d’un réseau de mots clés [38]. Les différentes informations

des publications, tel que mot clé, seront utilisées dans le but de comprendre la structure de

technologie. Finalement il s’agit de faire une analyse bibliométrique de mots clés pour configurer une

technologie.

99

Système d’Ingénierie

Un système est généralement défini comme une organisation complexe de ressources,

d’équipements et d’opérations à travers quoi les fonctions intégrées et besoins opérationnels sont

satisfaits [19]. Et un système d’ingénierie est un état ou un code de pratique [3] pour les points

suivants :

- Un système est un ensemble pour décrire un ensemble de composants, d’attributs et de

relations.

- Le comportement du système peut être observé à travers les activités du système ou à

partir de la transformation d’intrants en extrants.

- L’amélioration des méthodes pour définir le produit ou service doit aboutir aux besoins du

client et à la satisfaction du marché.

Le cadre conceptuel ou modèle conceptuel qui sera utilisé dans la présente synthèse de travaux de

recherche sera le suivant :

Figure 2: Système

Source : Auteur à partir de la définition de l’ISO ou International Standards Organization

Des analyses révèlent que l’ensemble de processus ou système d’ingénierie permet de structurer les

objectifs de l’entreprise [41]. En effet, il permet de structurer les éléments d’un ensemble.

METHODOLOGIE

La scientométrie s’avère le point focal de la méthodologie de cette synthèse de travaux de recherche.

Le processus de synthèse, ci-après, est cartographié selon cette théorie. Il comprend cinq étapes qui

sont présentées comme suit :

ACTIVITÉS INTRANTS EXTRANTS

Figure 3 :Source : Auteur

Dans la première étape, il s’agit de

publications internationales et les encadrements de mémoires de DEA

des indicateurs scientométriques, ces produits de recherche ont été évalués

statistique d’occurrence de mots clés

cartographier les réseaux de mots clés correspondants. Et cinquièmement, on a déduit un modèle de

système d’ingénierie.

RÉSULTATS

1. Contribution par format d’auteurs

Un format d’auteurs est formé par le nombre d’auteurs pour un article. Pour

produits de recherche, la contribution par format d’auteurs est p

colonnes : format d’auteurs, nombre de contributions et fréquence.

Table

Source: Auteur

COLLECTER LES PRODUITS DE RECHERCHE PUBLIÉS OU EN COURS DE PUBLICATION

ÉVALUER CES PRODUITS DE RECHERCHE À PARTIR DES INDICATEURS SCIENTOMÉTRIQUES

FAIRE LA STATISTIQUE D’OCCURRENCE DES MOTS CLÉS SUR LA

CARTOGRAPHIER LE RÉSEAU DE MOTS CLÉS SUR LA

TIRER UN MODÈLE DE SYSTEME D'INGÉNIERIE

100

: Processus de Synthèse de Produits de Recherche

Dans la première étape, il s’agit de collecter les produits de recherche. Ils comprennent

et les encadrements de mémoires de DEA. En deuxième étape et

des indicateurs scientométriques, ces produits de recherche ont été évalués

occurrence de mots clés sur la technologie a été établie. La quatrième


Contribution par format d’auteurs

Un format d’auteurs est formé par le nombre d’auteurs pour un article. Pour

produits de recherche, la contribution par format d’auteurs est présentée par le tableau suivant à trois

: format d’auteurs, nombre de contributions et fréquence.

Tableau 2: Contribution par Format d'Auteurs

FORMAT D’AUTEURS

NOMBRE DE

CONTRIBU-TIONS

FREQUENCE [%]

1 0 0.00

2 5 50.00

3 1 10.00

4 1 10.00

5 2 20.00

6 0 0.00

7 1 10.00

TOTAL 10 100.00

COLLECTER LES PRODUITS DE RECHERCHE PUBLIÉS OU EN COURS DE PUBLICATION

ÉVALUER CES PRODUITS DE RECHERCHE À PARTIR DES INDICATEURS SCIENTOMÉTRIQUES

FAIRE LA STATISTIQUE D’OCCURRENCE DES MOTS CLÉS SUR LA TECHNOLOGIE

CARTOGRAPHIER LE RÉSEAU DE MOTS CLÉS SUR LA TECHNOLOGIE

TIRER UN MODÈLE DE SYSTEME D'INGÉNIERIE

Processus de Synthèse de Produits de Recherche

er les produits de recherche. Ils comprennent les

deuxième étape et à partir

des indicateurs scientométriques, ces produits de recherche ont été évalués. En troisième étape, la

La quatrième étape consiste à


Un format d’auteurs est formé par le nombre d’auteurs pour un article. Pour le cas de synthèse de

résentée par le tableau suivant à trois

101

Les contributions en binôme représentent 50% des contributions. Et ceci représente la contribution

maximum. Il est suivi par le format à 5 auteurs qui a enregistré une fréquence de contribution de 20%.

Trois formats de 3, 4 et 7 auteurs ont chacun une fréquence de contribution de 10%. Et il n’y a pas eu

de contribution en 1 ou 6 auteurs.

2. Contribution dans le sens institutionnel

Quatre institutions ont participé dans la production des produits de recherche : l’Université

d’Antananarivo à travers l’École Supérieure Polytechnique, l’École Centrale de Marseille, l’Institut

Supérieure de Technologie d’Antananarivo, et les entreprises entre autres les Papeteries de

Madagascar-Océan Indien ou PAPMAD-OI, ORANGE-Madagascar, l’Office Malagasy de la Propriété

Intellectuelle ou OMAPI, la Cotonnière d’Antsirabe ou COTONA. Donc, il y a quatre variables sur

lesquelles on a appliqué l’Analyse des Composantes Principales ou ACP. Le diagramme suivant est

obtenu :

Figure 4: Diagramme d'Analyse des Composantes Princ ipales de Contributions Institutionnelles

Source : Auteur

Quatre composantes pricipales ont été mises en évidence : « université », « entreprise », « école », et

« institut ». Trois composantes « université », « entreprise » et « école » se trouvent dans le premier

cadrant, c’est-à-dire F1>0 et F2>0. Et une composante « institut » dans le quatrième quadrant, c’est-

à-dire F1>0 et F2<0. Le diagramme montre une collaboration étroite entre « l’université » et

« l’entreprise » avec un correlation de 1. La collaboration de « l’université » et « l’école » est rare. Il

en est également pour la collaboration de « l’université » et de « l’institut ».

0UNIVERSITE

0

INSTITUT

0

ECOLE

0

ENTREPRISE

-1

-0,5

0

0,5

1

-1 -0,5 0 0,5 1

--a

xis

F2 (

28.0

0 %

) --

>

-- axis F1 (44.00 %) -->


102

Il y a quatre facteurs associés : F1, F2, F3, F4. Ils sont présentés par l’histogramme suivant.

Figure 5: Les Facteurs Associés

Source : Auteur

Les facteurs F1, F2, F3 et F4 sont associés respectivement avec des valeurs Eigen qui sont 1.761,

1.100, 0.645 et 0.494. Donc les valeurs Eigen significatives retenues sont celles ≥1 associées à F1 et

F2. Et l’Analyse de Composants Principaux a donné les facteurs de charge suivants :

Table 3: Facteurs de Charge

F1 F2

UNIVERSITE 0.834 0.000

INSTITUT 0.492 -0.742

ECOLE 0.492 0.742

ENTREPRISE 0.763 0.000 Source : Auteur

Il y a une collaboration très forte entre l’Université et les Entreprises. Ce binôme est positionné

respectivement par (0.834;0.000) et (0.763;0.000). Les valeurs 0.000 au niveau de F2 explique les

positions horizontales de ces deux variables. Le binôme est suivi par la collaboration avec l’Ecole

Centrale de Marseille. Les coordonnées de cette dernière est (0.492;0.742). Et enfin la collaboration

Institut qui est positionnée par le point (0.492;-0.742). Ces deux dernières collaborations sont

symétriques par rapport à la contribution du binôme. Par degré « principale », les contributions sont

classées comme suit : [Université,Entreprise], [Ecole] et [Institut].

3. Contribution internationale

Dans chaque produit de recherche, le degré de collaboration internationale a été évalué. Et un

tableau à deux colonnes suivant est obtenu.

1

F2

F3F4

0,000

1,000

2,000

Eigenvalues

103

Table 4: Contribution Internationale

PRODUITS DE RECHERCHE

DEGRE DE COLLABORATION

INTERNATIONALE [%]

P 1 0

P 2 0

P 3 43

P 4 0

P 5 0

P 6 0

P 7 0

P 8 0

P 9 0

P.10 0

Source: Auteur

Il y a dix produits de recherche. Seul l’article 3 a obtenu un degré de collaboration internationale de

valeur 43%. C’est la collaboration avec l’équipe d’enseignants de l’Ecole Centrale de Marseille. Les

autres ont 0% en matière de collaboration internationale.

4. Revue des Pairs et Indexation

Chaque produit de recherche a fait l’objet d’évaluation en termes d’indicateurs qui sont : revue des

pairs et indexation. Le tableau suivant récapitule l’état de cette évaluation.

Table 5: Revue des Pairs et Indexation

PUBLICATIONS REVUE DES

PAIRS

INDEXATION

INSPEC COMPENDEX SCOPUS

P 1 � �

P 2 � �

P 3 � �

P 4 � � �

P 5 � � �

P 6 � � �

P 7 �

�

P 8 � � �

P 9 �

�

P.10 �

�

Source : Auteur

104

Tous les articles ont fait l’objet d’une revue des pairs. Concernant l’indexation, les articles 1 à 3 sont

indexés sur INSPEC. Les articles 4 à 6 et 8, quant à eux, sont indexés en même temps sur INSPEC

et COMPENDEX. Et les articles 7-9-10 sont indexés sur SCOPUS.

5. Réseau de mots clés

Les publications internationales contiennent au total 91 mots clés, dont les quinze premiers par ordre

décroissante d’occurrence ont été considérés. La figure suivante présente le réseau de mots clés qui

en découle.

GROUPE DE MOTS CLES PUBLICATIONS

INTERNATIONALES

RANG OCCURRENCE MOT CLE [total=91]

1 6 SKINNER

2 5 Processus

3 4 Veille Technologique P 1

4 4 Système P 2

5 4 Compétitivité P 3

6 3 Production P 4

7 3 TRIZ P 5

8 3 Produit P 6

9 3 Innovation P 7

10 2 Reengineering P 8

11 2 Analyse de la Valeur P 9

12 2 Brevet P 10

13 2 Gestion de Connaissances

14 2 Qualité

15 2 Assurance qualité

Figure 6: Réseau de Mots Clés contenus dans les pub lications internationales

Source : Auteur

Le mot clé le plus cité est « SKINNER ». C’est le nom du chercheur qui a inventé que la performance

d’une manufacture peut être mesurée par quatre métriques : coût, qualité, délai de livraison et

flexibilité. Le second mot clé le plus cité est « Processus ». Les mots clés Veille Technologique,

Système et Compétitivité ont été cités quatre fois. Et quant aux mots clés Production, TRIZ, Produit et

Innovation, ils ont été cités trois fois. Les mots clés les moins cités dans ce lot sont : Reengineering,

Analyse de la valeur, Brevet, Gestion de Connaissances, Qualité et Assurance qualité. Ils ont été

cités deux fois.

On remarque une intensité de lien au niveau des publications P1 à P7.

105

GROUPE DE MOTS CLES ENCADREMENT

DEA

RANG OCCURRENCE MOT CLE [total=32]

1 5 SKINNER E 1

2 5 Veille Technologique E 2

3 2 Gestion de Projet E 3

4 2 Système E 4

5 2 TRIZ E 5

6 2 Analyse de la Valeur E 6

7 2 Assurance qualité E 7

E 8

E 9

Figure 7: Réseau de Mots Clés contenus dans les mém oires de DEA encadrés

Dans le cas des mémoires de DEA encadrés, les mots clé le plus cités sont : SKINNER et Veille

Technologique. Ensuite, les mots clés suivants ont été chacun cité deux fois : Gestion de Projet,

Système, TRIZ, Analyse de la Valeur et Assurance Qualité.

6. Synthèse des Travaux de Recherche

Le réseau de mots clés révèle cinq théories d’ingénierie : Veille Technologique, Reengineering, TRIZ

ou Teoria Reshenia Izobretatelskih Zadatch, Analyse de la Valeur et Gestion de Connaissances. Le

cadre ou modèle conceptuel pour chaque théorie est élaboré en se conformant au modèle développé

dans le matériel. Les cadres qui en résultent sont énumérés ci-après.

La Veille Technologique est la théorie la plus citée. La figure 8 présente son cadre conceptuel.

Figure 8: Cadre Conceptuel de Veille Technologique Source : Auteur

Dans ce cadre conceptuel de Veille Technologique, les intrants sont : brevet, thèse, publication et

autres bases de données. Donc, ce sont des informations scientifiques ou techniques ou

technologiques ou technico-économiques. Et les extrants sont : prédiction, innovation, compétitivité et

performance.

CONDUIRE UNE VEILLE

TECHNOLOGIQUE PUBLICATION

THÈSE

BREVET

AUTRES BASES DE DONNÉES

INNOVATION

COMPÉTITIVITÉ

PERFORMANCE

PRÉDICTION

106

La théorie en deuxième position est la Re-Ingénierie. Son modèle conceptuel est présenté dans la

figure 9.

Figure 9: Cadre Conceptuel de Re-Ingénierie Source : Auteur

Dans le modèle conceptuel ci-dessus, il y a deux intrants : organisation et processus. Ce dernier

comprend le processus produit et le processus client. Les activités de Re-Ingénierie engendrent des

extrants : vision stratégique, amélioration stratégique, compétitivité et performance. On aura en

extrant un processus innovant.

La troisième théorie citée est la TRIZ. Son cadre conceptuel est présenté dans la figure 10 suivant :

Figure 10: Cadre Conceptuel de TRIZ Source : Auteur

Le cadre conceptuel de la TRIZ comprend un intrant et un extrant. L’intrant comprend les problèmes

spécifiques et l’extrant les solutions spécifiques. Ces dernières concernent l’innovation, la

compétitivité et la performance. On parle généralement d’idéalité.

La quatrième théorie est l’Analyse de la Valeur dont le cadre conceptuel est donné dans la figure

suivante.

Figure 11: Cadre Conceptuel d’Analyse de la Valeur Source : Auteur

Il y a trois intrants : les fonctionnalités, les coûts et la qualité. En extrants il y a les fonctionnalités qui

sont maintenues, les coûts qui sont réduits et la performance qui est maintenue ou améliorée.

CONDUIRE UNE TRIZ

PROBLÈMES SPÉCIFIQUES SOLUTIONS SPÉCIFIQUES

CONDUIRE UNE ANALYSE DE LA

VALEUR QUALITÉ

COÛTS

FONCTIONNALITÉS

PERFORMANCE

COÛTS RÉDUITS

FONCTIONNALITÉS

CONDUIRE UNE RE-INGÉNIERIE

ORGANISATION

PROCESSUS

VISION STRATÉGIQUE

RECONFIGURATION DE PROCESSUS

107

La dernière théorie est la Gestion de Connaissances. Son cadre conceptuel est montré dans la figure

9 suivant :

Figure 12: Cadre Conceptuel de Gestion de Connaissa nces Source : Auteur

Les intrants du cadre conceptuel de Gestion de Connaissances sont les données. Ces dernières

comprennent les brevets, les documents de projet, les documents de recherche et autres documents.

L’extrant est la banque de connaissances.

La synthèse de ces cadres conceptuels, compte tenu des occurrences des mots clés, est donnée par

la figure suivante.

CONDUIRE UNE GESTION DE

CONNAISSANCES DOCUMENTS DE RECHERCHE

DOCUMENTS DE PROJET

BREVET

BANQUE DE CONNAISSANCES

AUTRES DOCUMENTS

108

Figure 13: Synthèse des Cadres Conceptuels Source : Auteur

CONDUIRE UNE TRIZ

VISION STRATÉGIQUE

RECONFIGURATION DE PROCESSUS

CONDUIRE UNE REENGINEERING

SOLUTIONS SPÉCIFIQUES

CONDUIRE UNE GESTION DE

CONNAISSANCES DOCUMENTS DE

RECHERCHE

DOCUMENTS DE PROJET

BREVET

BANQUE DE CONNAISSANCES

CONDUIRE UNE VEILLE

TECHNOLOGIQUE PUBLICATION

THÈSE

BREVET PRÉDICTION

COMPÉTITIVITÉ

PERFORMANCE AUTRES BASES DE DONNÉES

AUTRES DOCUMENTS

INNOVATION

CONDUIRE UNE ANALYSE DE LA

VALEUR

FONCTION-NALITÉSCOÛTS

QUALITÉ

FONCTIONNALITÉS

COÛTS REDUITS

PERFORMANCE

ORGANISATION

PROBLÈMES SPÉCIFIQUES

PROCESSUS

Figure 14 : Modèle

RECHERCHER

COLLECTER

VALIDER

ANALYSER

DECIDER

PROCES-SUS

d'IDeA

RE

-E

NG

INE

ER

ING

T

RIZ

G

ES

TIO

N D

E

CO

NN

AIS

-S

AN

CE

S

VE

ILLE

T

EC

HN

OLO

GIQ

UE

CLI

EN

T

AN

ALY

SE

DE

LA

VA

LEU

R

PR

OB

LEM

E

PROCESSUS DE JAKOBIAK

QUALITE

COÜT

DELAI

FLEXIBILITE

Enca

dre

men

t

DEA

SKINNER

PARTAGE/CAPTURE DE CONNAISSANCES

109

: Modèle de Système d’Ingénierie pour piloter une innovation

MATRICE D'ALTSHULLER

PROCESSUS DE MILES

6 ÉTAPES DE HAMMER

PR

OB

LEM

E

SOLU

TIO

N

IDEA

LE

PR

OC

ESSU

S À

PR

OB

LÈM

ES

CO

ÛT

ETQ

UA

LITÉ

À

PR

OB

LÈM

ES

CO

ÛT

ETQ

UA

LITÉ

AM

ÉLIO

RÉE

S

ÉTAPES DE HAMMER

QUALITE

DELAI

COÛT

FLEXIBILITE

CLI

EN

T

PR

OC

ESSU

S IN

NO

VA

NT

110

Dans le modèle ci-dessus, on a la synthèse de cinq théories d’ingénierie les plus fréquemment

utilisées dans les recherches effectuées. La Veille Technologique est placé tout au milieu, elle a

la fréquence record en occurrence des mots clé. Il s’agit d’un processus de Jakobiak

schématisé par un flux circulaire continu à cinq étapes. Les quatre flux circulaires espacés

régulièrement signifient que la Veille Technologique est un processus périodique. Et la grande

flèche bleue indique qu’elle est un processus continu et permanent.

Mais avant de procéder à la Veille Technologique, les attentes du client, selon les métriques de

Skinner, ont été extraites. Ces métriques sont : qualité, coût, délai et flexibilité. Tout au long du

processus de Veille Technologique, ces quatres métriques seront périodiquement et

continuellement observées afin de se conformer aux attentes du client.

La TRIZ est la théorie qui vient en deuxième position en termes d’occurrence de mots clés. On

part d’un problème identifié lors de la conduite de la Veille Technologique. Ce problème est

traité par le processus utilisant la Matrice d’Altshuller. Et la solution idéale correspondante sera

introduite dans le processus de la Veille Technologique. Dans ce modèle, le recours à la TRIZ

se fera à la demande de la situation.

Pour la reengineering, le processus de Hammer en six étapes sera adopté. Ce processus sera

déclenché par l’identification d’un ou des processus à problème. En passant par ces six étapes,

un processus innovant est identifié et sera implémenté dans le processus de la Veille

Technologique. Le reengineering se situe en dernière position pour signifier le recours à cette

théorie à la fin de toutes les tentatives.

La théorie de l’Analyse de la Valeur dans ce modèle a sa raison d’être dans le cas où il y a des

coûts et qualité à problèmes dans le processus étudié. Ces coûts et qualité à problèmes vont

être criblés dans le processus d’Analyse de la Valeur. Et lorsque les coûts et qualité améliorées

sont identifiés, ils seront administrés dans le processus de Veille.

Enfin, le processus de Gestion de Connaissances élaboré par IDeA, un bureau de recherche

britannique, est utilisé. Le cercle bleu foncé indique une banque de connaissances et la flèche

horizontale en bleu clair montre l’avancement du processus. Au niveau de la banque de

connaissances se situe un partage et capture de connaissances d’une manière continue et

périodique. Les flèches convergentes schématisent ce partage/capture. Les blocs de banque de

connaissances/flèches qui se répètent à intervalle régulière montrent que le processus de

Gestion de Connaissances est un processus continu, périodique et permanent.

111

DISCUSSION Le modèle pour synthétiser les travaux de recherche réalisés est donné par le modèle de

système de management d’ingénierie supra.

Par rapport au Modèle de Système d’Ingénierie dével oppé par Zarczynski [41]

D’abord selon Zarczynski, le Système d’Ingénierie commence avec l’étude de marché et

l’identification des attentes du client. Par rapport à ces deux points, le modèle développé dans

la figure 14 comprend un processus de Veille Technologique qui est déclenché juste après avoir

identifié les besoins du marché ou les attentes du client à travers les métriques de Skinner.

Ensuite notre modèle poursuit par une observation continue et périodique de ce marché et de

ces attentes du client tout en apportant une amélioration continue. Ce qui constitue son plus par

rapport à celui de Zarczynski.

Le système de Zarczynski est développé et structuré sur la base de facteurs critiques et de

valeurs ajoutées. Et les méthodes et techniques de tel système ont été utilisées pour améliorer

la qualité et pour réduire le coût et le délai. Enfin, ce système améliore la conformité par rapport

aux exigences du client. Quant à notre modèle, les quatre paramètres de Skinner, coût-qualité-

délai-flexibilité, constituent nos facteurs critiques de succès. Ce sont des paramètres qui ont été

démontrés comme source de performance et de compétitivité. Et l’interaction de la Veille

Technologique, de l’Analyse de la Valeur, du TRIZ, du Reengineering et de la Gestion de

Connaissances autour de ces facteurs critiques de succès permet de se conformer

continuellement aux attentes du client.

Par rapport au Modèle de Système d’Ingénierie propo sé par Gonzalez et al. [13]

Les bénéfices du modèle de Gonzalez et al sont : standardisation, concepts d’utilisation de ces

standards dans l’opérationnel basé sur les besoins du client, retour au client de ses attentes

pour montrer comment ces standards ont satisfait ces attentes, solutions standards aux

attentes, mécanisme et test de conformité de produit, produit qualité, réduction de coût,

vérifiabilité, interface et profil d’ingénierie.

Notre modèle fournit des solutions collées à l’aspiration du client donc personnalisée. TRIZ, par

exemple, identifie dans un premier temps une solution standard dite générique. Et ensuite, elle

propose une solution spécifique correspondant au problème spécifique. Dans notre modèle, la

Veille Technologique, la Gestion de Connaissances, l’Analyse de la Valeur, le TRIZ et le

Reengineering comprennent en chacune un mécanisme et test de conformité. Dans le Gestion

de Connaissances, un test de conformité périodique a lieu pour vérifier si la connaissance

112

extraite apportera de la valeur ajoutée. Et quant à l’Analyse de la Valeur, elle permet de réduire

le coût et d’améliorer la qualité.

Par rapport au Modèle de Système d’Ingénierie de Fr iedman et Sage [9]

Un système d’ingénierie, selon Friedman et Sage, est utilisé pour identifier un cadre adéquat

pour une étude de cas de recherche. Notre modèle a été utilisé pour faire des recherches sur

l’innovation et la compétitivité. En plus de celui des deux auteurs ci-dessus, notre modèle a

permis de valoriser et de capitaliser les résultats de ces recherches dans un processus de

Gestion de Connaissances.

Friedman et Sage ont établi neuf concepts. Les six concepts de ces derniers représentent des

phases du cycle de vie d’un système d’ingénierie. Notre système d’ingénierie comprend ces six

concepts à savoir :

1. définition des attentes du client en termes de coût, qualité, délai et flexibilité

2. architecture du système

3. détail du système et configuration des sous-systèmes

4. système et intégration d’interfaces

5. validation et vérification

6. déploiement du système

Par rapport au Modèle de Système d’Ingénierie de Cr ider et DeRosa [4]

Crider et DeRosa ont conçu un système d’ingénierie entreprise. Ce système comprend cinq

processus : la planification technologique, la capabilité basée sur l’analyse de l’ingénierie,

l’architecture d’entreprise, la planification technique stratégique, l’évaluation et l’analyse

d’entreprise. Notre système, en termes de planification technologique, a :

- la Veille Technologique et la Gestion de Connaissances continues, périodiques et

permanentes

- l’Analyse de la Valeur, la Reengineering et le TRIZ planifiées selon le cas

Mais notre système comprend aussi l’évaluation et l’analyse d’entreprise. La Reengineering

comprend dans ses six phases une phase concernant l’évaluation et l’analyse d’entreprise

notamment son processus à problème. Le TRIZ aussi, dans son processus de modélisation des

problèmes, permet de faire une analyse de la situation de l’entreprise. L’Analyse de la Valeur

permet également d’analyser l’entreprise à travers l’analyse de la qualité et de la fonctionnalité

de son produit. Et enfin, la Gestion de Connaissances est une technologie pour évaluer

113

l’entreprise à travers les connaissances qu’on peut extraire tout au long du processus de Veille

Technologique.

Par rapport au Modèle de Système d’Ingénierie d’Est efan [7]

Estefan a présenté le concept de l’International Council of Systems Engineering ou INCOSE qui

est un Système d’Ingénierie Orienté Objet ou SIOO. Les objectifs de SIOO sont :

- capture et analyse des exigences et configuration de l’information pour spécifier le

système

- intégration avec d’autres processus d’ingénierie

L’avantage de notre modèle dans ce cas précis, notamment dans le deuxième point, est

l’intégration de cinq processus d’ingénierie. Ceci constitue une force dans la mesure où ces

cinq processus d’ingénierie convergent vers l’innovation et la compétitivité.

Par rapport au Modèle de Système d’Ingénierie de Sh eard [36]

Sheard a établi un tableau montrant douze rôles d’un système d’ingénierie. Elle a noté qu’aucun

système d’ingénierie ne peut assurer les douze rôles d’un seul coup. Notre modèle se distingue,

d’un rôle parmi les douze, par sa capacité de lier les sous-systèmes entre eux. Cette liaison est

caractérisée par l’occurrence des mots clés, un concept de scientométrie.

Par rapport au Modèle de Système d’Ingénierie de Sc hoening et Miller [34]

Un processus d’innovation a été mis en évidence par Schoening et Miller : Imaginer,

Développer, Evaluer et Vendre. Dans « imaginer », ils ont spécifié qu’il s’agit d’analyser les

besoins et d’analyser les fonctions. La Veille Technologique, le Reengineering et l’Analyse de la

Valeur comprennent cette étape « imaginer ». En plus ces trois sous systèmes apportent de

l’innovation. En particulier la Veille technologique permet de prédire le futur.

L’étape « développer » concerne la synthèse. Notre modèle, notamment au niveau de la Veille

Technologique, comprend également une étape « synthèse » où un réseau d’experts fait

intervenir les mathématiques et les statistiques pour présenter les résultats. L’étape « évaluer »

est commune aux cinq sous-systèmes. Ici l’évaluation de la solution innovante est au niveau de

chaque sous-système.

Par rapport au Modèle de Système d’Ingénierie de Ho nour [15]

Un système d’ingénierie génère un impact sur la qualité et le temps selon Honour. En fait, trois

systèmes d’ingénierie de Boeing [8], qui ont des similarités, ont été élaborés en même temps.

114

Ces trois systèmes sont utilisés pour la manipulation des grands assemblages durant la

construction d’avion. Ils ont une complexité différente et des interfaces et des processus

différents. Ils réduisent le délai et améliorent la qualité. Dans notre modèle, de système

d’ingénierie, la Gestion de Connaissances et l’Analyse de la Valeur permettent d’atteindre ces

mêmes performances. Ceci rapproche notre modèle de celui de Honour.

Le système de a un impact positif sur les systèmes complexes [17] et un impact sigificatif sur le

coût. Notre système comprend l’analyse de la Valeur qui a pour objectif de réduire le coût. Mais

en plus, l’Analyse de la Valeur améliore la qualité et la fonction. Deux points très important qui

n’ont pas n’ont pas été mentioné pour le modèle de Honour.

Par rapport au Modèle de Système d’Ingénierie de Fu gle [6]

DuPreez a développé le modèle de Fugle comme étant un processus dans une compagnie

d’assurance. L’objectif est d’identifier, d’évaluer, de développer, d’implementer et d’exploiter des

nouveaux produits et services pour l’efficience et l’efficacité. Notre modèle avec la Veille

Technologique permet d’atteindre cet objectif. Mais en plus avec la Veille Technologique, notre

modèle peut faire une projection du futur.

La discussion se résume comme suit :

Table 6 : Comparaison entre modèle Ravalison et aut res modèles

CE QUI EST DÉVELOPPÉ MODÈLE

Pla

nific

atio

n st

raté

giqu

e

Pla

nific

atio

n te

chno

logi

que

Axé

clie

nt

Fle

xibi

lité

Qua

lité

Coû

t

Dél

ai

Inno

vatio

n

Am

élio

ratio

n co

ntin

ue

Com

pétit

ivité

Ravalison � � � � � � � � � �

Zarczynski � � � �

Gonzalez et al. � � � �

Friedman et Sage � � � �

Crider et DeRosa � �

Estefan �

Sheard �

Schoening et Miller � �

Honour � �

Fugle � �

115

CONCLUSION La synthèse des travaux de recherche aboutissant à un modèle de système d’ingénierie revêt

plusieurs intérêts. D’abord le modèle à lui seul permet son application dans divers domaines

allant du management de l’ingénierie à la compétitivité globale. Le modèle a été extrait à partir

des cas étudiés donc son application dans d’autres cas sera aisé. Ensuite, un système

d’ingénierie constitue le point focal de toute démarche qualité ou d’innovation. Ici, le système

constitue une configuration de l’ensemble de l’entreprise mais aussi de ses démarches

startégiques. Et enfin, le modèle de système d’ingénierie permet de piloter une entreprise en

entière dans un contexte de compétition internationale.

Il est à noter l’adéquation entre le thème traité et le résultat qui est un modèle de système

d’ingénierie. Si le thème est de synthétiser des travaux de recherche, les résultats obtenus

aboutis à un modèle.

Pour piloter une entreprise ou toute autre entité, la Veille Technologique devra être mise en

place d’une manière permanente, continue et périodique. Parallèlement, à la Veille

Technologique, la Gestion de Connaissances devra être également mise en place d’une

manière permanente, continue et périodique. Et quand c’est nécessaire, on peut procéder à

l’Analyse de la Valeur ou à la TRIZ ou à la Reengineering. Le tout, qui est intégré dans la Veille

Technologique, permet d’avoir un modèle de système d’ingénierie pour piloter compétitivité et

innovation dans un contexte globalisé.

Ce modèle pourrait être utillisé dans l’amélioration du système santé. Un secteur qui demande

une réforme systémique. La réclamation des clients le témoigne. Le modèle est aussi conseillé

pour la bonne gouvernance dans la mesure où les réclamations des citoyens et des

contribuables foisonnent.

Mais ce modèle a ses limites. Sans la participation active et inclusive de tous les acteurs, la

résultante des actions sera à somme nulle. La participation active et inclusive, la concrétisation

du cercle de qualité, fait partie du concept de management qualité qui est la fondation du

modèle. Cette participation est aussi bien verticale que transversale. Le modèle a ses limites en

matière d’implémentation. Sa mise en œuvre demande l’intervention d’un expert en Système.

Ceci veut dire l’intervention d’une équipe de chercheurs. C’est le cas de plusieurs bureaux

116

d’études à démarche recherche américains tels qu’Ernst & Young, ou Boston Consulting Group,

ou MacArthur, ou MacKinsey.

Des questions apparaissent à l’issu de ce Modèle de Système d’Ingénierie :

- Aura-t-on les résultats escomptés, en termes de compétitivité et d’innovation, si on

applique le modèle dans le système santé ou dans le système gouvernance ou dans

le système défense nationale par exemple ?

- Quelle est la durée de vie de ce Modèle ? Peut-on l’appliquer à travers plusieurs cas

d’entreprise pour essayer de simuler une durée de vie ?

- Quelle serait la structure exigée par ce Modèle ?

117

CHAPITRE 3 : CURRICULUM VITAE

Université d’Antananarivo-Madagascar

Ecole Supérieure Polytechnique-Département Génie Electrique

Laboratoire d’Ingénierie de Projets Industriels

Bloc Technique Ankatso-101 Antananarivo-Madagascar

Téléphones : +261344358451

+261331289220

e-mails : [email protected]

[email protected]

Situation familiale : marié (trois enfants)

EDUCATION


Ecole Supérieure Polytechnique

Doctorat Nouveau Régime en Ingénierie de Projets Industriels (2008)



Diplôme d’Etudes Approfondies en Ingénierie de Projets Industriels (2006)

Université Nord de Madagascar-Madagascar


Diplôme d’Ingéniorat en Electromécanique (1984)

AFFILIATIONS

Membre des Anciens Saint Michel (ASM)

Membre de l’Ordre des Ingénieurs de Madagascar (OIM)

Délégué Régional de l’Ordre des Ingénieurs, section Antananarivo (DROI-T)

Membre de l’Union Nationale des Ingénieurs Chimistes et Ingénieurs en Techniques

Industrielles (UNICITI)

Senior member of the Institute of Industrial Engineers (IIE)

Country Representative of Portland International Center for the Management of Engineering

and Technology (PICMET)

DOMAINE D’ENSEIGNEMENT

Gestion de Production

Systèmes d’Engineering et Applications

Supply Chain Management

EXPERIENCE D’ENSEIGNEMENT

Analyse de la Chaine de Valeur

118

Benchmarking

Gestion de Production

Six Sigma

Reengineering

Théorie des Contraintes

Supply Chain Management

TRIZ ou Théorie de Résolution de Problèmes Inventifs

Value Stream Mapping

Veille Technologique

EXPERIENCE PROFESSIONNELLE 2012 : CENTRAFRIQUE

Bureau International du Travail (BIT)

Financement : Duché de Luxembourg

Expert International sur la méthodologie TREE (Training for the Rural Economic

Empowerment)

2012 : BURKINA FASO ET BENIN

Bureau International du Travail (BIT) : Projet de Renforcement des Compétences pour l’Emploi

des Jeunes et le Développement Rural en Afrique de l’Ouest

Financement : Danish International Development Agency (DANIDA)


Empowerment)

2011 : BURKINA FASO et BENIN

Bureau International du Travail (BIT) : Projet de Renforcement des Compétences pour l’Emploi

des Jeunes et le Développement Rural en Afrique de l’Ouest

Financement : Danish International Development Agency (DANIDA)


Empowerment)

2006-2008 : MADAGASCAR

Bureau International du Travail (BIT) : Projet sur la Croissance Economique Pro-Pauvre.

Financements : Swedish International Development Agency (SIDA)

Norwegian Agency for Development (NORAD)

Expert National.


Institut National de l’Infrastructure (ININFRA).

Chef de Service de la Pédagogie, de la Programmation et de la Scolarité.

Chef de Service Scolarité et Logistique de Formation.


Société Malgache d’Equipements Frigorifiques (SMEF).

Responsable Technico-commercial.

119

1988-1990 : BURKINA FASO

Ministère de l’Eau/PNUD-DTCD.

Ingénieur Spécialiste.


Commune Urbaine d’Antananarivo/Banque Mondiale : Projet sur la maintenance.

Chef de Service de la Maintenance.

ACTIVITES EN PUBLICATION INTERNATIONALE

Journal of Enterprise Transformation

Senior Reviewer

RAPPORTS D’ETUDES ET DE CONSULTANCE

Ravalison F. « Formation sur la Méthodologie TREE ou Training for the Rural Economic

Empowerment », Bureau International du Travail, Burkina Faso & Bénin, 2011.

Ravalison F. « Enquête sur le Travail des Enfants dans la Région Vakinankaratra », PACT

Madagascar, 2009. Ravalison F., Razafitsiatsipy R., Razafiarison C., « Analyse de la Chaîne de Valeur du Secteur

Broderie », Bureau International du Travail, Madagascar, 2008.

Ravalison F., Rakotozanany O., « Approfondissement des Groupes Cibles », Bureau

International du Travail, Madagascar, 2007. Ramananarivo R., Ramananarivo S., Ratsimbazafy J., Ravalison F., Rakoto D., Raveloson E. :

« Etat des Lieux de la Corruption dans le Secteur Industrie et Artisanat », Conseil Supérieur de

Lutte Contre la Corruption et le Royaume de Norvège, 2005.

Ravalison F. :« Etude sur les Méthodes Novatrices de Transfert de Connaissances et de

Compétences pour les Centres de Formation Professionnelle et les Entreprises », Fonds

d’Appui au Développement de l’Enseignement et la Banque Mondiale, 2004.

Ravalison F., Shabani J. :« Etude sur l’autonomisation de l’ONBAH et de l’ONPF », Programme

des Nations Unies pour le Développement et le Ministère de l’Eau, Burkina Faso, 1989.

Ravalison F., « Etude de scénarii de privatisation ou de concession de quelques services

municipaux : Unité de Concassage, Machine à Découper le Granit, Usine de Compostage,

Unité de Fabrication de Béton Préfabriqué », Banque Mondiale et la Municipalité

d’Antananarivo, Madagascar, 1987.

COMMUNICATION SCIENTIFIQUE

Ravalison F., Joromanantsoa H., Rarivoson A., Randriaherindrainy S., Raveloson E.,

120

Rakotomaria E. : « Evaluation de la Production Juste-A-Temps dans les Industries Textiles

Malagasy à Vocation Exportation », Université d’Antananarivo, 2005.

PUBLICATIONS INTERNATIONALES

Ravalison F., Rakotomalala Z.: “Quality Control Mapping to Assess Quality Assurance

Performance in Textile and Garment Factories”, Jordan Journal of Mechanical and Industrial

Engineering, 2012 (Ms. No. Fef. JJMIE-24-12: in progress)

Ravalison F., Rajaoarisoa T.: “Using Principal Component Analysis to Determine Key Factor of

Rural Electrification Development Investment”, Jordan Journal of Mechanical and Industrial

Engineering, 2011 (Ms. No. Ref. JJMIE-231-11: in progress) Ravalison F., Randriampenohaja J.: « Hadamard Matrix to Improve Enterprise’s Activities: An

Exploratory Research », International Journal of Industrial Engineering and Management, 2011

(in progress)

Ravalison F.: “Critical Success Factors are the Essence to Improve Productivity: A Technical

Note on the Article “Jig Design, Assembly Line Design and Work Station Design and their Effect

to Productivity””, Jordan Journal of Mechanical and Industrial Engineering, 2011 (Ms. No. Ref.

JJMIE-201-11: in progress).

Ravalison F., Rabenja N.: “Using Patent Statistics and Principal Component Analysis to Predict

Global Competition”, International Journal of Industrial Engineering and Management, Vol.2,

N°2, pp.45-50, 2011 .

Ravalison F., Raveloson E., Rakotomaria E.: “Towards “Fair Globalization”: Critical Success

Factors for Partnering Project of Development and Enterprise Project”, IEEE/PICMET, pp.1-7,

2010

Ravalison F., Rahoelison Toky, Raveloson E., Rakotomaria E.: “Developing a Mathematical

Concept and Process Technologies to Accompany Firms from Rut to Change”, IEEE/PICMET,

pp.786-792, 2009.

Ravalison F., Rajaonary P., Raveloson E., Rakotomaria E, Gazérian J., Loubet C., Ruiz J.M.:

“How does Reengineering sustain Economy?-The case of a paper industry in developing

country”, IEEE/PICMET, pp.210-219, 2008.

Ravalison F., Rajaonary P., Raveloson E., Rakotomaria E.: “Are Converging Technologies

Tools of Competitiveness?-The case of a paper mill industry in Madagascar”, IEEE/PICMET,

121

pp.241-245, 2007.

Ravalison F., Randrianasolo A., Mananjean D., Raveloson E., Rakotomaria E.,: “Assessment of

Technology Management in a Context of Sustainable Development: The Case of a Paper Mill

Industry in a Developing Country”, IEEE/PICMET, pp.2372-2381, 2006.

ORGANISATION DE CONFERENCES INTERNATIONALES

Portland International Conference for Management of Engineering and Technology or PICMET

in Phuket-Thaïland

Session Chair

The Moderating Effect of Employees’ P ersonality on the Relationship b etween Charismatic Leadership and Organizational Citizensh ip Behavior: Analysis of the High-

Tech Firms in Southern Taiwan Yuan-Duen Lee , Chang Jung Christian University, Taiwan

Shu-Hua Chiang, Chang Jung Christian University, Taiwan

Shih-Hao Chen , Chang Jung Christian University, Taiwan

Pi-Ching Chen , Chang Jung Christian University, Taiwan

Session Chair

The Moderating Role of Cognitive Style Congruence in the Relationship b etween Task Conflict and Team Performance

Chia-wu Lin , National Dong-Hwa University, Taiwan

Ya-Jen Cheng , National Dong-Hwa University, Taiwan

Ling-Ling Shen , Taiwan Power Company, Taiwan

Guo-Shu Yuan , National Sun Yat-Sen University, Taiwan

Session Chair

Communication Mechanism between Scientists and Entr epreneurs Haoshu Peng, Chinese Academy of Sciences, China

RECHERCHE EN COURS

Reengineering of the Malagasy Military System

Maintenance Analysis and Production Performance

Value Stream Mapping and Assessment of a Factory Performance

ENCADREMENT DE MEMOIRE POUR L’OBTENTION DU DIPLÔME D’ETUDES APPROFONDIES EN INGENIERIE DE PROJETS INDUSTRIELS

Co-Encadreur de Mémoire ANDRIAMPENOHAJA Andriamanalina Jens Démarche qualité utilisant la méthode des plans d’expériences pour améliorer la conservation du pain

122

Encadreur de Mémoire RASAMISON Jackot Herimanitra Adéquation de programme de formation professionnelle et technique aux besoins des entreprises Co-Encadreur de Mémoire RANDRIAMORA Edmond Management qualité vers l’innovation de la gestion d’un établissement d’enseignement technique et de la formation professionnelle Co-Encadreur de Mémoire RAHOELISON Toky Harivony Veille Technologique et TRIZ pour un processus d’innovation qualité dans la téléphonie mobile Co-Encadreur de Mémoire RAKOTOMALALA Zoho Révision du système de production par l’adaptation de la démarche Juste-A-Temps aux contraintes industrielles Encadreur de Mémoire SAÏD Hamid Abdoul Amélioration de l’accessibilité à l’eau potable par l’Analyse de la Valeur et la TRIZ Co-Encadreur de Mémoire RAJAONARISOA Tokary Amélioration de l’accessibilité des ruraux à l’énergie électrique par l’Analyse de la Valeur Co-Encadreur de Mémoire RAKOTOARISOA Jean Yves Erica Processus géomatique pour la localisation des zones à risques de feux de brousse Co-Encadreur de Mémoire RALAIHOVA Hajaharimalala Méthode de planification via l’aide à la décision multicritères INSTITUTE OF INDUSTRIAL ENGINEERS WEBINARS

“ How to get the most out of the Industrial and Syste ms Engineering Function for Full Potential Impact ” Date : Thursday, September 8, 2011 Time : 2 p.m. Eastern Time Presenter: D. Scott Sink, LeanSigma Certification Program Director, Integrated Systems

Engineering, College of Engineering, The Ohio State University

“ Recognizing and Managing Bottlenecks in Process Pla nts ” Date : Tuesday, October 4, 2011 Time : 2 p.m. Eastern Time Presenter: Peter L. King, President of Lean Dynamics LLC

“Accessible Simulation M ethods to Support Healthcare I mprovements ”

123

Date : Thursday, January 26, 2012 Time : 2 p.m. Eastern Time Presenters: Claire Cordeaux, Lead for Healthcare, SIMUL8 Corporation

Michele Stuart, President, Efficiency Engineers

“ Trends in Healthcare Process Improvement ”

Date : Friday, January 27, 2012 Time : 11 a.m. Eastern Time Presenters: Pam Arlotto, Maestro Strategies

Karl Kraebber, St. John's Hospital in Springfield, Illinois

“ How to Publish Your Work ” Presenter : Ricardo Valerdi, Systems & Industrial Engineering , Arizona University “ Value Stream Costing in Process Industries ” Presenter : Dirk Van Goubergen, Department of Industrial Management, Ghent University “ Fundamentals of Reliability Engineering and Applications ” Presenter : E. A. Elsayed, Rutgers University

“ Getting Your Engineering Management Research Published: What Not to Do?” Presenter : Toni L. Doolen, PhD Editor, Engineering Management Journal SERVICE UNIVERSITAIRE

Recrutement des étudiants pour le Diplôme d’Etudes Approfondies en Ingénierie de Projets

Industriels (Conception d’un processus de recrutement, Elaboration de guide d’entretien,

Sélection, Entretien, Sélection finale)

Membre de l’équipe préparant le LMD au sein du Département Génie Electrique (Conception

des Crédits pour le Master en Système et Ingénierie de Projets Industriels et suivi du processus

d’habilitation)

Chargé du montage et du suivi du dossier d’habilitation d’Ecole Doctorale de Génie Industriel,

de Système d’Ingénierie et d’Innovation Technologique

124

CONCLUSION GÉNÉRALE, PERSPECTIVES ET POSTULATION À L’HABILITATION À DIRIGER DES RECHERCHES

Ce mémoire d’Habilitation à Diriger des Recherches a permis de synthétiser des travaux de

recherche : publications internationales et encadrement de mémoires de Diplôme d’Etudes

Approfondies. Un modèle de Système d’Ingénierie a été obtenu à l’issu de cette synthèse. Ce

modèle a été comparé par rapport à neuf modèles de Système d’Ingénierie reconnu avoir

apporté changement, compétitivité et innovation. Notre modèle a des similarités par rapport ces

neufs modèles. Mais il a aussi ses spécificités et ses forces.

Ses spécificités sont : le fait que chaque composante du système est une théorie d’ingénierie

qui continue jusqu’à ce jour à révolutionner le contexte actuel façonné par une compétition

globalisée, l’intégration de quatre théories autour de la Veille Technologique. Quant à ses

forces, chaque composante du système converge vers innovation et compétitivité, parce-que le

dénominateur commun de ces composantes est la qualité.

Le modèle identifié dans le présent mémoire a ses limites. D’abord sa mise en œuvre nécessite

le pilotage par une personne acquis à la cause du système et processus. Ensuite, cette mise en

œuvre requiert une participation de tous. Et enfin, le modèle nécessite une reconfiguration de

l’entreprise ou de l’entité qui souhaite le mettre en place.

Ce modèle qui sera dénommé « Modèle Ravalison » va configurer les activités de recherche,

du Laboratoire de Systèmes et Ingénierie de Projets Industriels, dans cinq ans. Certes, le

modèle est une agrégation de démarches d’ingénierie, donc celles-ci vont configurer les

perspectives de recherche après l’Habilitation à Diriger des Recherches.

1. Objectif global

« Contribuer à la recherche de solutions innovantes pour l’université et l’entreprise ». D’une

part, cet objectif est aligné à la stratégie nationale de recherche, actuellement en cours

d’élaboration. D’autre part, il s’inscrit dans le cadre des objectifs stratégiques des organismes

nationaux, régionaux et internationaux. Il s’agit de recherche pour un développement global et

durable.

L’objectif global se décline en deux objectifs immédiats, le premier axé sur le milieu université et

le second axé sur le milieu entreprise. Mais ils sont interdépendants dans la mesure où les

résultats dans le premier milieu vont être utilisés pour améliorer le deuxième milieu. Et les

résultats dans le deuxième milieu vont servir d’intrants dans le premier.

125

2. Objectifs immédiats

Objectif immédiat 1 : Au niveau université, conduir e la globalisation, l’innovation, les

applications et les recherches à partir du modèle R avalison

Le modèle en question est une synthèse de plusieurs travaux de recherche : publications

internationales et encadrements de mémoires de Diplôme d’Etudes Approfondies. Son

application dans le cadre de recherche des étudiants master va être encouragée pour détecter

davantage ses forces et ses limites. Mais telle application pourrait stabiliser ou asseoir sa

notoriété. De même pour les doctorants, son application permet à ces derniers de conduire leur

recherche dans le cadre d’une démarche innovante. Le modèle va servir également de piloter le

procesus de globalisation et d’innovation des universités.

Objectif immédiat 2 : Au niveau entreprise, appliqu er le modèle Ravalison pour

redynamiser les entreprises

L’entreprise nationale actuelle subit des effets indésirables de la concurrence internationale.

Des fois, les dirigeants d’entreprise ne savent pas par où commencer pour redynamiser

l’entreprise en question ou pour piloter l’entreprise par rapport à cette concurrence. Le modèle

Ravalison leur sera proposé pour configurer les actions à entreprendre une fois une Value

Stream Mapping achevée. Dans ce cas, la mise en œuvre du lean manufacturing ou service

sera aisée.

3. Stratégie à développer

Pour atteindre ces objectifs, une démarche incluant les autres enseignants chercheurs et

étudiants sera engagée. Dans cette démarche, il sera développée un processus globaliste et

globalisant.

4. Résultats attendus

Résultat 1.1

Recherche des étudiants master orientée publication s internationales

Le modèle Ravalison sera le vecteur directeur de toutes activités de recherche, notamment les

recherches des étudiants master. Et une des finalités de ces recherches est la rédaction d’un

article en vue d’une publication intenationale. Cette démarche donnera une position globale de

l’Ecole doctorale.

Résultat 1.2

Recherche des doctorants orientée solution innovant e et globale

126

Ici également, le modèle supra reste le vecteur directeur de toutes les activités de recherche

doctorale. Dans le cadre de ces dernières, l’investissement se focalisera sur les solutions

innovantes. Des solutions innovantes qui seront identifiées logiquement par la Veille

Technologique, colonne vertébrale du modèle en question, et les autres composantes. Ceci

pourrait aboutir soit à une publication internationale soit à un autre modèle métamorphisé.

Résultat 1.3

Recherches avec co-encadrement global développées

La collaboration internationale sera développée avec plusieurs universités étrangères. Les

réseaux actuels, l’Institute of Industrial Engineers (IIE)-le Southern African Institute of Industrial

Engineers (SAIIE), l’International Conference on Industrial Engineering and Operations

Management (IEOM)-l’International Council on Systems Engineering (INCOSE), seront sollicités

pour des co-autariats ou des co-encadrements. Telle démarche donnera une dimension

internationale aux produits de recherche.

Résultat 2.1

Partenariat avec les entreprises dans le cadre de p ilotage de lean manufacturing ou

service par le modèle Ravalison

Le modèle sera mis à la disposition des entreprises pour faciliter la mise en œuvre du lean

manufacturing ou service. Les entreprises rencontrent une certaine difficulté pour cette mise en

œuvre. Le modèle va facilter le pilotage du lean.

5. Activités

Correspondant au résultat 1.1. :

(1.1.).1. Intégrer le modèle Ravalison dans la méthodologie de recherche des étudiants en

master. Il s’agit de recherche dans le domaine d’ingénierie industrielle. Les étudiants sont

encouragés à rechercher des données secondaires qu’ils vont rechercher par la veille

informationnelle. Cette démarche apportera qualité et vitesse dans la recherche.

(1.1.).2. Intégrer dans l’encadrement des enseignants en provenance d’universités étrangères.

Dans le cadre de globalisation des recherches et des produits de recherche, le co-encadrement

de format international dynamique est l’idéal.

(1.1.).3. Faire, au moins, une publication internationale à partir d’un mémoire de master avec un

format d’auteurs internationaux.


127

(1.2.).1. Intégrer le modèle Ravalison dans la méthodologie de recherche des doctorants.

Généralement, une recherche comporte un ensemble de méthodes. Le modèle en question

sera intégré dans cet ensemble pour démarrer la recherche et en même temps pour tester la

force de la méthode.

(1.2.).2. Explorer le modèle Ravalison pour aboutir à un autre modèle métamorphisé. Selon le

champ de recherche et le contexte, l’applicabilité du modèle sera engagée. Ce processus

pourrait aboutir à un changement du modèle.

(1.2.).3. Faire, au moins, une publication internationale à partir d’une thèse de doctorat avec un

format d’auteurs internationaux.


(1.3.).1. Etablir une convention de collaboration avec le Pensylvania State University pour la

mise en place de « learning factory » à l’Ecole Supérieure Polytechnique d’Antananarivo. Des

échanges de point de vue avec cette Université dans le cadre de cette « learning factory » ont

déjà eu lieu. La mise en place effective avec collaboration de recherche sera la prochaine

étape,

(1.3.).2. Etablir une convention de collaboration avec l’Arizona University pour l’intégration des

enseignants malagasy dans le processus de revue d’articles à publication internationale.

Actuellement, « reviewer » dans une revue internationale de cette université, Journal of

Enterprise Transformation, quelques chercheurs Malagasy seront intégrés dans cette

processus,

(1.3.).3. Dynamiser la convention de collaboration avec l’Ecole Centrale de Marseille par

l’échange d’enseignants chercheurs ou par le co-encadrement ou par le co-autariats de

recherche. Une convention existe déjà. Quelques mises à jour seront à négocier pour pouvoir

développer davantage la collaboration.


(2.1.).1. Partager le modèle Ravalison aux entreprises à travers des séries de conférences. Ces

conférences seront organisées en partenariat avec la Chambre de Commerce et d’Industrie

d’Antananarivo, la Chambre des Mines et l’American Chamber of Commerce. Les groupements

tels que le Syndicat des Industries de Madagascar, le Groupement des Entreprises Malagasy,

le Fivondronan’ny Mpandraharaha Malagasy et le Groupement des Entreprises Franches et

Partenaires seront à intégrer dans le processus,

(2.1.).2. Former les entreprises aux pratiques de lean manufacturing ou lean service. Le plan de

formation correspondant mettra en avance le modèle en question. L’approche pourra se faire

128

par groupe d’entreprises ou par entreprise. Cette série de formations devrait aboutir à la mise

en place effective du lean dans les entreprises bénéficiaires de la formation.

(2.1.).3. Accompagner les entreprises dans le lean. Le processus ne s’arrête pas au niveau

formation. Un accompagnement des entreprises sera programmé pour s’assurer que les

résultats escomptés sont obtenus.

(2.1.).4. Monter un « Observatoire des Industries Nationales ». Un mémoire d’ingéniorat portant

sur cet observatoire vient d’être achevé l’année dernière. L’étude va être approfondie et

l’observatoire va être mis en place. Le modèle Ravalison sera un outil utilisé dans cet

observatoire.

Au vue de ces perspectives de recherche, je me sens pleinement responsable et engagé dans

le processus correspondant. Responsable parce-qu’une recherche est un travail d’équipe et

une équipe nécessite un leadership acquis à la cause de la recherche. Responsable parce-que

des résultats tangibles et rapides sont attendus de la part de cette équipe. Et responsable

parce-qu’un positionnement à atteindre et à préserver, autant que possible, est en jeu.

Je suis aussi engagé par la dimension globale de la perspective de recherche. Engagé parce

que la conduite d’une recherche globale nécessite un engagement total. Engagé puisqu’il y va

de la cote et de l’honneur de l’Université d’Antananarivo. Et engagé car si on ne la fait pas on

ne saura pas qui va la faire.

En effet, responsabilité et engagement, deux dimensions qui repèrent ma forte motivation dans

la recherche et dans le partage des produits de recherche. Mais aussi, deux facteurs clés qui

me propulsent, au plus haut point, à postuler pour l’Habilitation à Diriger des Recherches. Une

étape que je pense déterminante pour l’avenir de la recherche.

129

BIBLIOGRAPHIE [1] Aksnes D. W., Taxt R. E.: “Peer Reviews and Bibliometric Indicators: A comparative Study

at a Norwegian University”, Research Evaluation, N°13, Vol.1, pp.33-41, 2004

[2] Beaver D. D.: “Reflections on Scientific Collaboration (and its study): Past, Present and Future”, Scientometrics, N°52, pp.365–377, 2001

[3] Blanchard B.S., Fabrycky W.J.: ”Systems Engineering and Analysis”, Prentice-Hall International, 3-rd edition, 1998

[4] Crider K.A., DeRosa J.K.: “Findings of Case Studies in Enterprise Systems Engineering”, http://www.mitre.org/tech papers/tech papers07/01 05 61.pdf , consultée le 30 Janvier 2012.

[5] De Bellis N.; “Bibliometrics and Citation Analysis: From the Science Citation Index to Cybermetrics”, Lanham, Maryland: The Scarecrow Press, Inc, 2009

[6] DuPreez N., Louw L.: “A Framework for Managing the Innovation Process”, IEEE-PICMET Transactions, pp.546-558, 2008

[7] Estefan A.J.: “Survey of Model-Based Systems Engineering Methodologies”, www.omgsysml.org/MBSE_Methodology_Survey_RevA.pdf, consulté le 4 Août 2010

[8] Frantz W.F.: “The Impact of Systems Engineering on Quality and Schedule-Empirical Evidence”, INCOSE International Symposium, St Louis, MO, 1995

[9] Friedman G., Sage A.: “Case Studies of Systems Engineering and Management in Systems Acquisition”, Wiley Periodicals, Systems Engineering, Vol.7, N°1, pp.84-97, 2004

[10] Garfield E., Pudovkin A. I., Istomin V. S.: “Algorithmic Citation-Linked Historiography Mapping: The Literature of Science”, In ASIS&T 2002: Information, Connections and Community. 65th Annual Meeting of ASIST in Philadelphia, PA, pp.18-21, 2002

[11] Girvan M., Newman M. E.: “Community Structure in Social and Biological Networks”, Proceedings of the National Academy of Sciences of the United States of America, Vol.99, N°12, pp.7821–7826, 2002

[12] Glänzel W.: “Bibliometrics as a Research Field: A Course on the Theory and Application of Bibliometric Indicators”, http://www.citeseerx.ist.psu.edu/viewdoc/Download?doi=10.1.1.97. 5311&rep=repl&type=pdf, consulté le 22 Octobre 2003

[13] Gonzalez P., White J., Christie B.: “Applying Systems Engineering Principles to the Development of Transportation Communication Standards”, http://www.noblis.org/NewsPublications/NewsReleases/Documents/ApplyingSystemsEngine eringPrinciplestoITSStandards.pdf

[14] Guiliani F., De Petris M.P., Nico G.: « Assessing Scientific Collaboration Through Coauthorship and Content Sharing », Scientometrics, N°85, pp.13-28, 2010

[15] Honour E.: “Understanding the Value of Systems Engineering”, http://www.incose.org/secoe/0103/ValueSE-INCOSE04.pdf, consulté le 13 Décembre 2011

130

[16] Juznic P., Peclin S., Zaucer M., Mandelj T., Pusnik M., Demsar M.: “Scientometric Indicators: Peer Review, Bibliometric Methods and Conflict of Interests”, Scientometrics, N°85, pp.429-441, 2010

[17] Kludze A.K.: “The Impact of Systems Engineering on Complex Systems”, Conference on Systems Engineering Research, University of Sotuhern California, Los Angeles, CA, 2004

[18] Lee P.-C., Su H.-N., Wu F.-S. : “Quantitative Mapping of Patented Technology-The Case of Electrical Conducting Polymer Nanocomposite”, Technological Forecasting and Social Change, N°77, pp.466-478, 2009

[19] M’Pherson P.K.: ”Systems Engineering: An Approach to Whole System Design”, The Radio and Electronic Engineer, Vol. 50, N°11/12, pp.545-558, 1980

[20] Nattar S.: “Indian Journal of Physics: A Scientometric Analysis”, International Journal of Library and Information Science, Vol. 1, N°4, pp.55-61, September, 2009

[21] Nattar S.: “Indian Journal of Physics: A Scientometrics Analysis”, International Journal of Library and Information Science, N°4, Vol.1, pp.55-61, 2009

[22] Newman M. E. J.: “Coauthorship Networks and Patterns of Scientific Collaboration”, Proceedings of the National Academy of Sciences of the United States of America, Vol.101, N°1, pp.5200–5205, 2004

[23] Pouris A.: “A Scientometric Assessment of the Southern Africa Development Community: Science in the Tip of Africa”, Scientometrics, N°85, pp.145-154, 2010

[24] Pritchard A.: ”Statistical Bibliography or Bibliometrics?”, Journal of Documentation, N°24, pp.348-349, 1969

[25] Ravalison F., Rajaoarisoa T.: “Using Principal Component Analysis to Determine Key Factor of Rural Electrification Development Investment”, Jordan Journal of Mechanical and Industrial Engineering, 2011 (Ms. No. Ref. JJMIE-231-11: in progress)

[26] Ravalison F., Randriampenohaja J.: « Hadamard Matrix to Improve Enterprise’s Activities: An Exploratory Research », International Journal of Industrial Engineering and Management, 2011 (in progress)

[27] Ravalison F. : “Redynamisation de la Compétitivité, dans le Contexte Mondialisé, par la Démarche Intégrée d’Ingénierie Industrielle”, Thèse de Doctorat Nouveau Régime n°023/08 , Ecole Supérieure Polytechnique, Université d’Antananarivo, Madagascar, 2008

[28] Ravalison F., Rahoelison Toky, Raveloson E., Rakotomaria E.: “Developing a Mathematical Concept and Process Technologies to Accompany Firms from Rut to Change”, IEEE/PICMET, pp.786-792, 2009.

[29] Ravalison F., Rajaonary P., Raveloson E., Rakotomaria E, Gazérian J., Loubet C., Ruiz J.M.: “How does Reengineering sustain Economy?-The case of a paper industry in developing country”, IEEE/PICMET, pp.210-219, 2008.

[30] Ravalison F., Rajaonary P., Raveloson E., Rakotomaria E.: “Are Converging Technologies

131

Tools of Competitiveness?-The case of a paper mill industry in Madagascar”, IEEE/PICMET, pp.241-245, 2007.

[31] Ravalison F., Randrianasolo A., Mananjean D., Raveloson E., Rakotomaria E.,: “Assessment of Technology Management in a Context of Sustainable Development: The Case of a Paper Mill Industry in a Developing Country”, IEEE/PICMET, pp.2372-2381, 2006.

[32] Ravalison F., Raveloson E., Rakotomaria E.: “Towards “Fair Globalization”: Critical Success Factors for Partnering Project of Development and Enterprise Project”, IEEE/PICMET, pp.1-7, 2010

[33] Repanovici A.: “Measuring the Visibility of the Universities’ Scientific Production Using Scientometric Methods”, Advanced Educational Technologies, 2010

[34] Schoening W.W., Miller A.J.: “Using Systems Engineering to Expand Opportunities for Innovation”, http://www.libsys.uah.edu/library/incose/Contents/Papers/93/93104.pdf consulté le 24 Juillet 2011.

[35] SCImago, SJR-SCImago, Journal & Country Rank, http://www.scimagojr.com, consultée le 14 Août 2009

[36] Sheard S.: “Twelve Systems Engineering Roles”, http://incose.org/educationcareers/PDF/ 12-roles.pdf , consulté le 11 Février 2011

[37] Sonnenwald D. H.: “Scientific Collaboration”, Annual Review of Information Science and Technology, N°41, pp.643–681, 2007

[38] Su H. N., Lee P. C.: “Mapping Knowledge Structure by Keywords Co-Occurrence: A First Look at Journal papers in Technology Foresight”, Scientometrics, N°85, pp.65-79, 2010

[39] Van Raan A. F. J.: “Advanced Bibliometric Methods as Quantitative Core of Peer Review Based: Evaluation and Foresight Exercises”, Scientometrics, N°36, Vol.3, pp.397–420, 1996

[40] Vinkler P.: “Indicators are Essence of Scientometrics and Bibliometrics”, Scientometrics, N°85, pp.861-866, 2010

[41] Zarczynski W.: “Developing a Systems Engineering Management Model for Use by Eskom Enterprises in Order to Gain a Competitive Advantage”, IEEE/PICMET, pp.304-313, 2008

DDDÉÉÉVVVEEELLLOOOPPPPPPEEEMMMEEENNNTTT DDD’’’UUUNNN MMMOOODDDÈÈÈLLLEEE DDDEEE SSSYYYSSSTTTÈÈÈMMMEEE DDD’’’ IIINNNGGGÉÉÉNNNIIIEEERRRIIIEEE ÀÀÀ PPPAAARRRTTTIIIRRR DDDEEE LLLAAA SSSYYYNNNTTTHHHEEESSSEEE DDDEEESSS TTTRRRAAAVVVAAAUUUXXX DDDEEE RRREEECCCHHHEEERRRCCCHHHEEE

RESUME

Le système académique Malagasy a intégré l’Habilitation à Diriger des Recherches comme processus permettant à tout titulaire de doctorat de devenir Professeur. Ce processus consiste à produire des travaux de recherche, à les synthétiser et à soutenir le produit final. Et l’objectif du présent mémoire s’inscrit dans le cadre de ce processus. La scientométrie a été utilisée comme méthodologie pour faire l’évaluation et la synthèse de ces derniers. La scientométrie est une branche de la science. C’est une science et un instrument fondamental pour l’évaluation statistique des résultats de recherche scientifique. On a en amont des travaux de recherche : dix publications internationales et neuf encadrements de niveau Diplôme d’Etudes Approfondies. Le tout dans le domaine de l’ingénierie industrielle. La scientométrie a été utilisée pour évaluer la qualité des travaux, d’un côté. De l’autre côté, à partir de l’occurrence et le réseau de mots clés, on a élaboré un système d’ingénierie. La démarche a montré la qualité des travaux. En matière de contribution par format d’auteurs, en moyenne un produit de recherche a été réalisé par deux à cinq auteurs. Concernant la contribution dans le sens institutionnel, il y a eu une tendance remarquable en matière de collaboration avec l’entreprise. Sur la contribution internationale, une publication a été produite en collaboration avec une institution étrangère. Et six produits de recherche ont été indéxés sur INSPEC. La synthèse des travaux de recherche a été effectuée à partir des réseaux de mots clés. De ces derniers, un système d’ingénierie dénommé « Modèle Ravalison » a été identifié et qui a configuré la synthèse en question.

Mots clés : réseau de mots clés, scientométrie, système d’ingénierie, Modèle Ravalison

DDDEEEVVVEEELLLOOOPPPMMMEEENNNTTT OOOFFF AAANNN EEENNNGGGIIINNNEEEEEERRRIIINNNGGG SSSYYYSSSTTTEEEMMM MMMOOODDDEEELLL FFFRRROOOMMM TTTHHHEEE SSSYYYNNNTTTHHHEEESSSIIISSS OOOFFF RRREEESSSEEEAAARRRCCCHHH WWWOOORRRKKKSSS

ABSTRACT

The Malagasy academic system has integrated the “Habilitation à Diriger des Recherches” as a process allowing all titular of doctorate to become Professor. This process consists in producing research works, synthesizing those latter and presenting the final outputs. Moreover, the objective of the present dissertation is within that process. We have used scientometry to process assessment and synthesis. The scientometry is a branch of the science. It is a science and a fundamental instrument for the statistical assessment of the scientific research results. We have achieved ten international publications and nine master students mentoring. All are in the area of the industrial engineering. We have used scientometry to evaluate the quality of works, on one hand. On the other hand, from the occurrence and the network of key words, we have elaborated an engineering system model. These works have showed the quality of works. Two to five authors have achieved, concerning authorship pattern, on average a product of research. Concerning the institutional contribution, there was a remarkable trend concerning collaboration with the enterprise. On the international contribution, we have published manuscripts in collaboration with a foreign institution. In addition, we have issued six research products on INSPEC. The mapping and study of the keyword networks has allowed synthesizing the research works. From these last, we have identified a system of engineering model named “Ravalison model”. It has permitted to map and visualize the synthesis.

Keywords: keywords network, scientometry, engineering system, Ravalison model