Improving metaheuristic performance by evolving a variable fitness function

Description

In this paper we study a complex real world workforce scheduling
problem. We apply constructive search and variable neighbourhood search
(VNS) metaheuristics and enhance these methods by using a variable fitness
function. The variable fitness function (VFF) uses an evolutionary approach to
evolve weights for each of the (multiple) objectives. The variable fitness
function can potentially enhance any search based optimisation heuristic where
multiple objectives can be defined through evolutionary changes in the search
direction. We show that the VFF significantly improves performance of
constructive and VNS approaches on training problems, and "learn" problem
features which enhance the performance on unseen test problem instances.

Links & Downloads

1. Dahal KP, Remde SM, Cowling PI et al (2008) Improving metaheuristic performance by evolving a variable fitness function. In: Evolutionary computation in combinatorial optimization. 8th European Conference (EvoCOP 2008) Naples, Italy, March 26-28, 2008: 170-181.
2. http://hdl.handle.net/10454/2498
3. https://doi.org/10.1007/978-3-540-78604-7_15

Tags

Variable fitness function

Evolution

Heuristic

Meta-heuristic

Workforce scheduling problem

Additional Fields

Identifer	oai:union.ndltd.org:BRADFORD/oai:bradscholars.brad.ac.uk:10454/2498
Date	January 2008
Creators	Dahal, Keshav P., Remde, Stephen M., Cowling, Peter I., Colledge, N.J.
Publisher	Springer Verlag
Source Sets	Bradford Scholars
Language	English
Detected Language	English
Type	Conference paper, Accepted manuscript
Rights	© 2008 Springer Verlag. Reproduced in accordance with the publisher's self-archiving policy. Original publication is available at http://www.springerlink.com