Interactive Metaheuristics
“Interactive metaheuristics for optimization-based decision support systems” is a project funded by the DFG (Deutsche Forschungsgemeinschaft). The objective of this project is to develop efficient and practical models, algorithms and software tools for the design of interactive heuristic optimisation methods for decision support systems. The project is conducted at the Institute of Computer Science of the University of Osnabrück. Read more…
Hyper-heuristic for CHeSC
An Evolutionary Programming Hyper-heuristic (EPH) has been proposed for the Cross-Domain Heuristic Search Challenge 2011. The proposed method combines an evolutionary programming approach and co-evolution. The solving process of EPH consists in evolving a population of solutions by applying heuristics sequences. The heuristics sequences are also in a population and evolve according to their performances.
A small application that allows to visualize the process of the EPH hyper-heuristic as been implemented and presented during the OR53 Annual Conference 2011. Read more and download demo…
Agent Metaheuristics Framework
AMF (Agent Metaheuristics Framework) aims at analyzing existing algorithms, and facilitating the design of hybrid or new metaheuristics. It proposes an organizational model of metaheuristics that can be used to describe both population-based metaheuristics and trajectory methods. In this model, a metaheuristic is viewed as an organization composed of a set of roles which interact in order to find an optimal solution. Read more…