Optimization of Stochastic Discrete Event Simulation Models

Abstract

Many systems in logistics can be adequately modeled using stochastic discrete
event simulation models. Often these models are used to find a good or optimal
configuration of the system. This implies that optimization algorithms have to
be coupled with the models. Optimization of stochastic
simulation models is a challenging research topic since the approaches should
be efficient, reliable and should provide some guarantee to find at least in
the limiting case with a runtime going to infinite the optimal solution with a
probability converging to 1.
The talk gives an overview on the state of the art in simulation
optimization. It shows that hybrid algorithms combining global and local
optimization methods are currently the best class of optimization approaches
in the area and it outlines the need for the development of software tools
including available algorithms.