We consider complex systems that are composed of two subsystems with single or multiple objectives. Their complexity is reflected, among others, in the fact that direct solution approaches are available only at the subproblem level and not at the level of the entire system. In practical applications, these subsystems model, for example, different parts or aspects of an overall system. Such subsystems often have interactions with each other and they are usually not sequentially ordered or obviously decomposable. Thus, the individual solutions of subsystems do not generally induce a solution for the overall system. The talk focuses on general modeling aspects related to multiobjective optimization of complex systems. Among others, we discuss subsystem and system feasibility and suggest different notions of optimality for complex systems.
Solution concepts are discussed in the light of recent developments in this field.