Search

We consider the use of the Choquet integral for evaluating projects or actions in a real-world application starting from the case of the re-qualification of an abandoned quarry. Despite the Choquet integral being a very well-known preference model for which there is a rich and well developed theory, its application in a multiple criteria decision aiding perspective requires some specific methodological developments. This led us to work out and implement, in practice, two new procedures: A first procedure to build interval scales with the objective of assigning utility values on a common scale to the criteria performances, and a second one to construct a ratio scale for assigning numerical values to the capacities of the Choquet integral. This article discusses the strengths and weaknesses of the Choquet integral as appearing in the case study, proposing as well insights related to the interaction of the experts within a focus group

Centrally planned Beveridge healthcare systems typically rely heavily on local or regional "health authorities" as responsible organisations for the care of geographically defined populations. The frequency of reorganisations in the English NHS suggests that there is no compelling unitary definition of what constitutes a good healthcare geography. In this paper we propose a set of desirable objectives for an administrative healthcare geography, specifically: geographical compactness, co-extensiveness with current local authorities and size and population homogeneity, and we show how these might be operationally measured. Based on these objectives, we represent the problem of how to partition a territory into health authorities as a multi-objective optimisation problem. We use a state-of-the-art multi-objective genetic algorithm customised for the needs of our study to partition the territory of the East England into 14 Primary Care Trusts and 50 GP consortia and study the tradeoffs between objectives which this reveals.