Optimisation par essaim particulaire de la planification des actions de maintenance d’un parc immobilier

Abstract

Maintenance is a central activity in real estate property management. It requires the development of action plans which is the operational support of the maintenance strategy. But these action plans can be complex to build and justify. To address this problem, we propose a method to model the real estate property and to build optimized action plans. The modelling approach uses a tree description of the real estate property and a multicriteria simulation logic. From this model, a Particle Swarm Optimization allows to establish an action plan with optimal relevance in relation to the requirements of the decision maker. An example related to the maintenance of a major French company real estate property illustrates the method process and to highlight its interest.