Interactive multiobjective optimization design strategy for decision baseddesign

This research focuses on multi-objective system design and optimization. Th e primary goal is to develop and rest a mathematically rigorous and efficie nt interactive multi objective optimization algorithm that takes into accou nt the Decision Maker's (DM's) preferences during the design process, An in teractive MultiObjective Optimization Design Strategy (iMOODS) has been dev eloped in this research to include the Pareto sensitivity anal? sis, Pareto surface approximation and local preference functions to capture the DM's p references in an Iterative Decision Making Strategy (IDMS), This new multio bjective optimization procedure provides the DM with a formal means for eff icient design exploration around a given Pareto point. The use of local pre ference functions allows the iMOODS to construct the second order Pareto su rface approximation more accurately in the preferred region of the Pareto s urface. The iMOODS has been successfully applied to two test problems. The first problem consists of a set of simple analytical expressions for the ob jective and constraints. The second problem is the design and sizing of a h igh performance and low-cost ten bar structure that has multiple objectives . The results indicate that the class functions are effective in capturing the local preferences of the DM. The Pareto designs that reflect the DM's p references can be efficiently generated within IDMS.