Optimizing Uncapacitated Facility Location Problem with Cuckoo Search Algorithm based on Gauss Distribution

Abstract

The objective of this study was to assess the capability of the Gauss distribution-based Cuckoo Search algorithm (GCS) in solving the Uncapacitated Facility Location Problem (UFLP). UFLP is an optimization problem that there are number of locations available to be built a facility so that it can serve number of customers, assuming each facility has no limits to serve customers and only a single facility is allowed to provide services to each customer. The objective function of UFLP is to minimize the combined costs of constructing facilities in an area and providing services to customers. UFLP falls under the category of NP-Hard Problems, where the computation complexity increases with the size of the data. The Cuckoo Search algorithm, which mimics the breeding behavior of Cuckoo birds, has been extensively used to tackle optimization problems. GCS was introduced to overcome the weaknesses of Cuckoo Search algorithm in terms of computational time and search accuracy. GCS used Gaussian distribution instead of Levy Flight which based on Levy distribution. In this study, the GCS algorithm was implemented using JavaScript and the dataset used was obtained from ORLib. The research outcomes showed that the GCS algorithm could achieve optimal result in all dataset.