DistriNet Seminar:
Simulation-based Optimisation

by Juergen Branke
Professor of Operational Research and Systems at Warwick Business School, University of Warwick, UK

March 27, 2017 - Celestijnenlaan 200A, Heverlee (Leuven)

Abstract

Simulation has become an invaluable tool to analyze complex systems, and is used across disciplines including engineering, manufacturing and social sciences. However, optimization based on simulation models is challenging, not only because running a simulation to evaluate a solution is computationally expensive, but also because most discrete event simulation models are stochastic, turning the optimization problem into a stochastic problem.
In this talk, I will present two new algorithms to tackle optimization based on simulation. The first aims at efficiently identifying the best system from a small set of alternatives by distributing simulation runs in a way that maximizes the value of information collected. The second is an enhanced version of simulated annealing which allows searching large search spaces efficiently even if the evaluation function is noisy.

Biography

Peggy ValckeJuergen Branke is Professor of Operational Research and Systems at Warwick Business School, University of Warwick, UK. He received his PhD from the University of Karlsruhe, Germany, in 2000 and has been an active researcher in the area of nature-inspired optimization since 1994. He has published over 150 articles in international journals and conferences on various topics such as multiobjective optimization, handling of uncertainty in optimization, dynamically changing optimization problems, and the design of complex systems, in particular in the area of scheduling and logistics. Prof. Branke is Area Editor of the Journal of Heuristics, Associate Editor of IEEE Transactions on Evolutionary Computation and the Evolutionary Computation Journal.

 

 


Practical info