Search

International audience; Dans un marché fortement concurrentiel, les industriels doivent constamment améliorer leurs produits pour satisfaire leurs clients tout en restant compétitifs et en limitant la prise de risques. Or, en phase préliminaire de conception, la définition encore incomplète du produit rend la prévision de ses performances incertaine. C'est donc dans un contexte d'évaluation incertaine que les décideurs doivent faire les choix de conception les plus satisfaisants pour les clients. Plusieurs indices de plus-value ont été proposés dans la littérature dans le contexte de l'analyse multicritère pour estimer l'espérance de gain de satisfaction par rapport aux améliorations concurrentes envisageables. Ces modèles reposent cependant sur des hypothèses qui font abstraction de la faisabilité des améliorations espérées. Sur la base de techniques d'analyse de décision multicritère en environnement incertain, cet article propose une extension possibiliste du concept d'indice de plus-value. Mots-clés : Evaluation de performance ; Décision dans l'incertain ; Analyse multicritère ; Choquet integral ; Théorie des possibilités ; indice de plus-value. Abstract: In a highly competitive environment, manufacturers must constantly improve their products to remain competitive and satisfy their customers while minimizing incurred costs and risk taking. At the early stages of (re-) engineering, performance forecasting of future product is uncertain. It is therefore in a context of uncertainty that decision makers must choose the most satisfactory design for customers. Some theoretical worth indexes have been proposed in the multi-criteria analysis literature to estimate the expectable gains when improving changes are planned. However, they generally rely on non-realistic assumptions on the achievability of the expected improvements. Based on multi-criteria decision analysis techniques and uncertainty theory, this paper proposes an extension of the worth index concept into a possibilistic framework.

International audience; The new information and communication technologies represent useful tools for improving the health system. Decisions on their implementation and use must be based on solid observations to assess their impacts on care and health. The difficult use of classical economic evaluation methods is due to the multidimensionality of the assessment of telemedicine implementations. Indeed, to evaluate the impact of the introduction of telemedicine, it is necessary to take into account various indicators identified in the literature, as cost, accessibility, acceptability and quality. Furthermore, they need to be considered with respect to all the stakeholders involved in the decision process (physicists, patients and hospital director). In this context, decision support theories provide additional insight into the evaluation of telemedicine.The aim of this paper is to present a multicriteria evaluation methodology involving multiple decision makers for the establishment of remotely monitored medical dialysis solutions.To reach a decision, the hospital director wishes to take into account, next to the above mentionned indicators, the preferences of the physicians and the dialysis patients. In order to explicit these preferences, we suggest to consider that the patients and the physicians are decision makers which have to express a ranking of care modalities evaluated on multiple criteria. We thus face a situation involving multiple judges which overlay a multiplicity of viewpoints. To solve this problem, we use techniques both from social choice theory and multicriteria decision analysis. First we propose to agregate the individual rankings of the patients on the potential multicriteria care modalities into a patient-wide order (via the Borda count and Copeland's pairwise aggregation), which is then used in the multicriteria preference elicitation phasis of the physicians. The individual rankings of the latter ones is then aggregated into a physician-wide ranking which is then combined with the patients' one and further economical criteria to elicit the hospital director's preferences on the care modalities. This methodology is applied to the implementation of medical dialysis units in Brittany (France), on basis of the preferences of 16 patients, 8 physicians and 1 hospital director.The preferences of all the stakeholders of this decision process are representable by additive value models. The preferred alternative among the patients and the physicians is the medical dialysis unit with remote monitoring. The criterion which has the highest impact on the patients' (resp. physicians') decision is the distance to the dialysis unit (resp. the patients' satisfaction). Finally, the analysis shows that the most important criteria for the hospital director are the physicians' and the patients' satisfaction (which are generally not considered).This work shows that it is important to consider the heterogeneity and the multiplicity of viewpoints and actors in the evaluation of the implementation of remotely monitored medical dialysis solutions via multicriteria decision support techniques. We believe that this approach is an important stepping stone in the evaluation of telemedecine.

Nous proposons dans cet article une analyse critique du classement mondial des universités publié chaque année par l’Institut de l’Enseignement supérieur de l’Université Jiao Tong de Shangaï et plus connu sous le nom de « classement de Shangaï ». Après avoir rappelé comment le classement est construit, nous discutons de la pertinence des critères utilisés. Nous analysons ensuite la méthode d’agrégation proposée. Notre analyse se fonde sur les outils et concepts de l’« aide multicritère à la décision ». Nos conclusions principales sont que les critères utilisés ne sont pas pertinents, que la méthode d’agrégation présente des problèmes majeurs et que l’exercice souffre d’une réflexion insuffisante sur les questions liées à la structuration du problème. Le classement de Shangaï, malgré la grande couverture médiatique qu’il reçoit chaque année, n’est donc pas un outil pertinent pour juger de la « qualité » des institutions académiques, guider le choix des étudiants ou des familles, ou promouvoir des réformes du système d’enseignement supérieur. This paper proposes a critical analysis of the “Academic Ranking of World Universities”, published every year by the Institute of Higher Education of the Jiao Tong University in Shanghai and more commonly known as the Shanghai ranking. After having recalled how the ranking is built, we first discuss the relevance of the criteria and then analyze the proposed aggregation method. Our analysis uses tools and concepts from Multiple Criteria Decision Making (MCDM). Our main conclusions are that the criteria that are used are not relevant, that the aggregation methodology is plagued by a number of major problems and that the whole exercise suffers from an insufficient attention paid to fundamental structuring issues. Hence, our view is that the Shanghai ranking, in spite of the media coverage it receives, does not qualify as a useful and pertinent tool to discuss the “quality” of academic institutions, let alone to guide the choice of students and family or to promote reforms of higher education systems. We outline the type of work that should be undertaken to offer sound alternatives to the Shanghai ranking. Nos proponemos en este artículo un análisis crítico de la clasificación mundial de las universidades publicado cada año por el Instituto de Enseñanza Superior de la Universidad de Jiao Tong de Shangai y más conocido bajo el nombre de “clasificación de Shangai”. Después de haber recordado como se construyó esta clasificación, nosotros discutimos la pertinencia de los criterios utilizados. Luego analizamos el método de agregación propuesto. Nuestro análisis de basa sobre los instrumentos y los conceptos de la “ayuda multicriterio a la decisión”. Las conclusiones principales son que los criterios utilizados no son pertinentes, que el método de agregación presenta grandes problemas y que el ejercicio padece de una reflexión insuficiente sobre las cuestiones relacionadas con la estructuración del problema. La clasificación de Shangai, a pesar de su gran cobertura mediática que recibe cada año, no es entonces un instrumento pertinente para juzgar de la calidad de las instituciones académicas, para guiar la elección por parte de los estudiantes y de sus familias, o promover las reformas del sistema de enseñanza superior.