Research On The Optimal Problem Of Theoretical Handling Time For Material Supply Of Large Ships

Along with the changes of military structure, the demand of building large ship which can sail on the sea for a long time is increasing. Material supply is an important factor of the endurance and motility. Completing the supply of materials in a shorter time is particularly important. So it has important theoretical and realistic significance to research the optimal problem of theoretical handling time for material supply and discuss the optimal models and computational methods.Based on the background of one specific problem of material supply of large ships,optimal models of theoretical handling time are built in this paper. The optimization is primarily aimed at flow control and transport policy selection. The goal is to make handling time shortest. The process of material supply is abstracted by material type and the number of makeup port, where the optimal problem is divided into four conditions:single material type& single makeup port, single material type& double makeup ports,multi material types& single makeup port, multi material types& double makeup ports.In single material type condition, the problem is attributed to the network maximum flow problem, which is solved by Ford-Fulkerson algorithm through maximizing the system flow. In multi material types condition, the optimization content is the deliver plan from dock to ship. Such problems are belong to NP-hard problem and are similar with Job-shop scheduling problems. Based on the analogy with Job-shop scheduling problem,the optimal model of multi material types conditions are established, where the deliver plan from dock to ship is taken as decision variable and the handling time is taken as objective function. Genetic algorithm is adopted to solve these problems. Matrix encoding is designed and a way of transformation is proposed to make the illegal chromosome legitimize. Besides, the standard genetic algorithm has slow convergence speed and being easily trapped in local optimal, so the improved genetic algorithm of the model is proposed in this paper. Through an example, several specific problems are solved by the algorithms which are put forward by this paper. The results not only verify the effectiveness of the models and algorithms but also show that the improved genetic algorithm has faster convergence speed and better solution than the standard genetic algorithm.