Study Of The Multi-objective Shore AGV-Quay Crane Scheduling Optimization Based On Improved SPEA2

With the rapid development of economic globalization,international trade has become more frequent.Ocean shipping plays a decisive role in the trade.The increase in trade has also increased the competition between ports.Improving the efficiency of port operations and reducing port operating costs has become the constant pursuit of all ports.Intelligent dispatching and exploration of automated terminals have also become important tasks for the further development of ports.Effectively dispatching and utilizing the port's important operating equipment-AG Vs and quay cranes—is critical to the construction of automated docks.It also is directly related to the economic development of a country.Cooperative scheduling optimization for AGV and quay crane has important practical significance.Through the field visit and investigation of Qingdao Port and reading a large number of related literatures,a detailed analysis and study of port scheduling,especially AGV and shore crane scheduling,was conducted.Based on previous research,adding equipment waiting time for shore cranes,considering the initial and final tasks of AGV,considering the conditions of each ship in port three objective functions and several constraints are constructed,and a multi-objective AGV-quay crane coordinated scheduling system is established.According to the multi-objective AGV-shore crane coordinated scheduling optimization model and the actual situation of the port,a multi-objective hybrid online AGV scheduling algorithm(M-HOSA)is proposed.On this basis,SPEA2 is improved through research and analysis of SPEA2.By adding a local search algorithm and some individual update algorithms,improving the crossover strategy,and combinedding with the shore-constraint constraint algorithm,an AGV-quay crane coordinated scheduling algorithm(MSAQ-SPEA2)based on improved SPEA2 is proposed.Finally,using the FlexTerm2.0 experimental platform,the model established in this paper and the proposed algorithm are experimentally verified.By analyzing the experimental results,it is concluded that the model is effective.The algorithm has a significant improvement in operational efficiency,convergence and stability.