Department of Industrial Engineering, Faculty of Technology and Engineering, East of Guilan, University of Guilan, Guilan, Iran
School of Business Management, University of Rahbord Shomal, Guilan, Iran
Developing an appropriate plan for distribution department is significant because of its influence on company's other costs and customers' satisfaction. In this study, a new bi-objective mix-integer linear programming model developed for scheduling two groups of experienced and inexperienced distribution staff based on capillary marketingin Pak Pasteurized Dairy Products Company of Guilan province in order to reduce costs and increase sales along with customer satisfaction. Several constraints are taken into account at the model.The model solving results using the epsilon constraint method, which provides a set of Pareto responses and solved by GAMS software, shows efficiency of the model to solve small-size problems. In order to evaluate the validity of model in large scale problem, with respect to NP-hardness of the problem, a multi objective water flow-like optimization (MOWFO) algorithm was expanded. For evaluation the suggested method, several problems were expanded and the efficiency of the method was compared with a multi-objective invasive weed optimization (MOIWO) algorithm based on the planned factors. For better algorithms performance, their input parameters were set using RSM technique; furthermore, in order to compare parameters statistically, the Tukey’s 95% confidence interval method was used. The results show the superiority of MOWFA compared to MOIWO algorithm in comparison indicators.