Density based problem space search for the capacitated clustering p-median problem
This paper proposes a clever and efficient constructive algorithm for the Capacitated Clustering Problem and a new search space for this problem based on perturbation vectors to the data vectors. Our density constructive algorithm finds 7 optimal solutions for benchmarks from the literature. Our problem space search extends the Storers problem space search idea to provide an alternative search space for the problem which is generated by applying our high quality constructive algorithm to the perturbated data. We also propose a general framework for building efficient heuristics, by adding elements of adaptive computation and periodic construction-deconstruction concepts within the constructive heuristics.
Citation : Ahmadi, S. and Osman, I.H. (2004) Density based problem space search for the capacitated clustering p-median problem. Annals of Operations Research, 131(1-4), pp. 21-43.
ISSN : 1572-9338
Research Group : Software Technology Research Laboratory (STRL)