2016/08/02 by Juste Raimbault, Raimbault, Juste
Decision Sciences · Environmental Science · #Multi-Criteria Decision Making #Environmental Impact and Sustainability #Sustainability and Ecological Systems Analysis
paper · pdf · doi:10.48550/arxiv.1608.00840
Multi-objective evaluation is a necessary aspect when managing complex\nsystems, as the intrinsic complexity of a system is generally closely linked to\nthe potential number of optimization objectives. However, an evaluation makes\nno sense without its robustness being given (in the sense of its reliability).\nStatistical robustness computation methods are highly dependent of underlying\nstatistical models. We propose a formulation of a model-independent framework\nin the case of integrated aggregated indicators (multi-attribute evaluation),\nthat allows to define a relative measure of robustness taking into account data\nstructure and indicator values. We implement and apply it to a synthetic case\nof urban systems based on Paris districts geography, and to real data for\nevaluation of income segregation for Greater Paris metropolitan area. First\nnumerical results show the potentialities of this new method. Furthermore, its\nrelative independence to system type and model may position it as an\nalternative to classical statistical robustness methods.\n