Lugar: Sala de Grados CITE III Fecha: 13/01/17 Hora: 10:00

Multiobjective optimization problems arise in many fields, such as engineering, economics, logistics, when optimal decisions must be take in the presence of trade-offs between several conflicting objectives. As a rule, there does not exist a single solution that simultaneously optimizes each objective. Instead, there exists a (usually infinite) set of Pareto optimal solutions. A solution is called non-dominated or Pareto optimal if none of the objective functions can be improved in value without degrading of the other objective values. Without additional subjective preference information provided by a decision maker, all Pareto optimal solutions are considered equally good. Therefore, solving a multiobjective problem is usually understood as finding a Pareto optimal solution which is the most preferred for a decision maker. The most popular approaches for solving multiobjective optimization problems are Multiple Criteria Decision Making (MCDM) and Evolutionary Multiobjective Optimization (EMO). Although they address similar problems as emphasized, they have different research goals. On the one hand, in MCDM, typically the aim is to support a decision maker in identifying the most preferred solution. EMO based on an evolutionary algorithm works with a population of individuals and attempts to approximate the entire Pareto front. Recently, methods that incorporate decision maker’s preferences into evolutionary approaches has become a new trend. So-called preference-based EMO algorithms are being actively developed that aim to find an approximation of the Pareto front which elements are scattered regarding the preference information provided by the DM. Moreover, when solving multiobjective optimization problems with multiple conflicting criteria, decision makers must compare several different alternatives and select the most preferred one. The task of comparing multidimensional data is very demanding for the decision maker without any support. Different graphical visualization methods and tools can be used to support and help the decision maker in understanding similarities and differences between the alternative solutions.

Some of the attendees together Dr. Ernestas Filatovas after his talk