Research On Models And Algorithms For Agricultural Machinery Scheduling Problem With Time Window

With the rise of the land transfer policy,farmers have transferred land contractual management rights,forming a large-scale business entity,which promotes the formation of agricultural machinery service organizations and the emergence of agricultural machinery socialization services.At present,agricultural machinery operations present problems such as imbalance between supply and demand,backward scheduling,low operational efficiency,and unreasonable resource allocation.Achieving scientific and rational agricultural machinery scheduling is an urgent problem that agricultural machinery service organizations need to solve.The agricultural machinery scheduling problem is essentially a special kind of vehicle scheduling problem,which belongs to the multi-objective combination optimization problem.Based on the actual characteristics of agricultural machinery operation scheduling,this paper established a model for agricultural machinery scheduling problem with time window,and proposes an agricultural machinery scheduling method based on improved genetic algorithm.The model and algorithm are verified by test data and actual instance data.The main contents are as follows:(1)In view of the agricultural machinery service mode represented by agricultural machinery cooperatives,considering the influencing factors such as farmland operation points,agricultural hangars,agricultural machinery,space and time,combined with the research of existing vehicle scheduling problems and the actual characteristics of agricultural machinery operations,a model for agricultural machinery scheduling problem with time window was established.Compared with the variant models for traditional vehicle scheduling problem,the model for agricultural machinery scheduling problem with time window proposed in this paper integrates multiple constraints such as multi-depots,vehicle heterogeneity,time window,agricultural machinery matching rules and multi-objectives,which makes the model adapt to a wider range of scenarios.(2)Based on the cluster first,route second algorithm,a two-stage algorithm for solving the agricultural machinery scheduling problem with time window is proposed.The algorithm is divided into clustering stage and scheduling stage.In the clustering stage,a new farmland task clustering method is proposed,which takes the spatial distribution characteristics and time window distribution characteristics of the task into account.Besides,the agricultural machinery operation matching rules were considered,as well as the agricultural machinery library resource limit feasibility.In the scheduling stage,a genetic algorithm based on improved push-forward route construction heuristic and local search technology was designed.In the framework of genetic algorithm,chromosomes were properly encoded.The evolution process included elitism,selection,crossover,mutation and education.Several mutation operators and local search operators were designed.(3)The algorithm was verified by the test data set and agricultural machine scheduling actual case.An agricultural machinery scheduling system with time window was designed and implemented.Combining the test data set and the actual case of agricultural machine operation scheduling,the feasibility and effectiveness of the agricultural machine scheduling model and the algorithm are verified.Based on the software layering principle,an agricultural machinery scheduling system was designed and implemented.The system consists of information management module,agricultural machinery scheduling module,visualization module and input/output module.