A new method to determine a well-dispersed subsets of non-dominated vectors for MOMILP problem

Multi-objective optimization is the simultaneous consideration of two or more objective functions that are completely or partially in conflict with each other. The optimality of such optimizations is largely defined through the Pareto optimality. Multiple objective integer linear programs (MOILP) are special cases of multiple criteria decision making problems. Numerous algorithms have been designed to solve MOILP and multiple objective mixed integer linear programs. However, MOILP have not received the algorithmic attention that continuous problems have. This paper uses the data envelopment analysis (DEA) technique to find a well-dispersed non-dominated vectors of multiple objective mixed integer linear programming (MOMILP) problem with bounded or unbounded feasible region, while the previous methods consider only problems with bounded feasible region. To this end, it uses the L1−norm and the modified slack-based measure (MSBM) model. The proposed method does not need the filtering procedures and it ranks the elements of well-dispersed non-dominated vectors of MOMILP problem. The proposed algorithm is illustrated by using two numerical examples.