A Bi-level Multi-objective Facility Location Model for Smart Dustbins with Customer Utility

Considering the customer utility in competitive environment, this paper develops a bi-level multi-objective location model for recycling facility location. The upper level decides on the smart bins location of follower enterprise to reduce the total cost and ensure coverage maximization, and the customers as the lower level decide on which kind of facility and which one to provide service based on the competition between the traditional and smart bins to maximize the customer utility. Concerning the complexity of the proposed model, this paper develops an improved NSGA-II to solve it. A special coding rule is employed. Uniform crossover and inversion mutation operator are used to improve the effectiveness of NSGA-II. Finally, based on new instances generated from the existed single objective MCLP benchmark data sets (22 instances), the experimental results indicated that the improved NSGA-II had competitiveness and a better performance in convergence than MOHSA, with the improved algorithm achieving the best average gaps of 0.88% and 3.87% and improving the best-known solutions in some instances. Besides five groups of simulation instances proved that the model proposed and algorithm improved could solve the facility location problem considering customer utility in competition environment in a reasonable time to get pareto solutions. The results of sensitivity analysis proved that the objective values were positively correlated with three important parameters, which suggested that reasonable scale and incentive mechanism could affect the system over the long-term.