Solving Truck Scheduling Optimization Problem in Multi- Door Cross Dock with Learning Effect and Deteriorating Jobs Using Social Engineering Optimizer

In general, each supply chain consists of three main stages of procurement, production and distribution. The use of the cross-docking system is a new strategy at the distribution stage to improve customer response time by moving products directly from pickup trucks to delivery trucks. Generally, for an activity to be done both machine and human resources are needed. Many researchers have already developed numerous planning methods for cross-docking systems, but human resource constraints have largely ignored. In this paper, for the first time, we examine the problem of truck scheduling in multi-door cross-dock considering the learning effects and the deterioration of tasks to fill the gap between theoretical planning models and what is happening in the real world. We have proposed a mixed integer programming model for this problem. According to the research literature, with increasing the size of the problem, the complexity of integer programming model is expanding rapidly so that the exact methods can hardly achieve the optimal solution. To solve large-scale problems, five meta-heuristic algorithms are used including Genetic Algorithms (GA), Imperial Competitive Algorithm (ICA), Keshtel Algorithm (KA), and Social Engineering Optimization (SEO). Finally, the numerical results obtained from all meta-heuristic algorithms are analyzed. We compare the meta- heuristic algorithms based on the best, average, Rpd and time criteria. As a result, the SEO and KA algorithm performed better than the other algorithms in terms of solution quality.