Hybrid PSO-WDBA method for the site selection of tailings pond

Abstract

The disposal of tailings is a vital process in mines. This study aims to address site selection issue of tailings pond using a suitable method. First, as different types of evaluation criteria exist, hybrid decision making information is considered. That is, crisp numbers are used to express quantitative values, while picture fuzzy numbers (PFNs) are suggested to describe qualitative information. Moreover, new comparison methods of PFNs are introduced to overcome the existing limitation. Considering the situation where weight information is partly known, a nonlinear optimization model is constructed. Then, to calculate criteria weights, the particle swarm optimization (PSO) algorithm is modified by introducing penalty functions. Afterwards, the weighted distance based approximation (WDBA) method is extended with hybrid evaluation information to rank alternatives. Last, the hybrid PSO-WDBA technique is applied in selecting the best tailings pond site in a gold mine. The effectiveness and advantages of this method are also verified through detailed discussions. The results show that the proposed evaluation model is dependable and effective, and can provide references for the site selection and management of tailings pond.