Picture of person typing on laptop with programming code visible on the laptop screen

World class computing and information science research at Strathclyde...

The Strathprints institutional repository is a digital archive of University of Strathclyde's Open Access research outputs. Strathprints provides access to thousands of Open Access research papers by University of Strathclyde researchers, including by researchers from the Department of Computer & Information Sciences involved in mathematically structured programming, similarity and metric search, computer security, software systems, combinatronics and digital health.

The Department also includes the iSchool Research Group, which performs leading research into socio-technical phenomena and topics such as information retrieval and information seeking behaviour.

Explore

Constraint directed variable neighbourhood search

Andrew, A. and Levine, J. and Long, D. (2007) Constraint directed variable neighbourhood search. In: Proceedings of the 4th International Workshop on Local Search Techniques in Constraint Satisfaction held at CP 2007. UNSPECIFIED.

Full text not available in this repository. Request a copy from the Strathclyde author

Abstract

Local search algorithms operate by making small changes to candidate solutions with the aim of reaching new and improved solutions. The problem is that often the search will become trapped at sub optimal states from where there are no improving neighbours. Much research has gone into creating schemes to avoid these local optima and various strategies exist mainly based around altering the acceptance function. Another approach is Variable Neighbourhood Search which aims to bypass optima by linearly switching through multiple search neighbourhoods. We propose a new method where the selection of neighbourhoods is dynamically decided dependant on the violations of the problem constraints, Constraint Directed Variable Neighbourhood Search. We compared Constraint Directed Variable Neighbourhood Search to Variable Neighbourhood Search and show that the same search progress can be achieved whilst exploring only a fraction of the states.