In many practical situations, decisions are multi-objective in nature. Furthermore, costs and profits are time-dependent, i.e. depending upon the time a decision is taken, different costs and profits are incurred. In this paper, we propose a generic approach to deal with multi-objective time-dependent optimization problems (MOTDP). The aim is to determine the set of Pareto solutions that capture the interactions between the different objectives. Due, to the complexity of MOTDP, an efficient approximation based on dynamic programming is developed. The approximation has a provable worst case performance guarantee. Even though the approximate Pareto set consists of less solutions, it represents a good coverage of the true set of Pareto solutions. Numerical results are presented showing the value of the approximation.
|Name||BETA publicatie : working papers|