A hybrid DEA-based K-means and invasive weed optimization for facility location problem

Abstract

In this paper, instead of the classical approach to the multi-criteria location selection problem, a new approach was presented based on selecting a portfolio of locations. First, the indices affecting the selection of maintenance stations were collected. The K-means model was used for clustering the maintenance stations. The optimal number of clusters was calculated through the Silhouette index. The efficiency of each cluster of stations was determined using the Charnes, Cooper and Rhodes input-oriented data envelopment analysis model. A bi-objective zero one programming model was used to select a Pareto optimal combination of rank and distance of stations. The Pareto solutions for the presented bi-objective model were determined using the invasive weed optimization method. Although the proposed methodology is meant for the selection of repair and maintenance stations in an oil refinery Company, it can be used in multi-criteria decision-making problems.