Research On Logistics Scheduling And Optimization Of Resource Based Enterprises

The production process in resource-based enterprises usually has many characteristics such as massive transportation tasks,limited transportation facilities,strict transportation requirements,high transportation costs and so on.These characteristics make the logistics scheduling in production process of resource-based enterprises complex and difficult,and unscientific scheduling can lead to substantial cost waste.Therefore,how to improve the enterprises' logistics scheduling capability,improve efficiency,reduce costs and increase profits under the current resource conditions is a problem that enterprises need to solve urgently.In this thesis,the distribution of product oil in resource-based enterprises and mine production are taken as the research background,and the vehicle scheduling problem is studied by focusing on the problem of vehicle scheduling of two stage logistics of product oil and vehicle scheduling in open-pit mines.Based on the analysis of the actual problem background,the continuous time scheduling model is established,and the improved algorithm is designed.From these work,the two stage logistics of product oil and vehicle scheduling capacity of open-pit mines is improved,transportation costs are reduced,and transportation efficiency are improved.The main contents of this thesis are as follows:1)The vehicle scheduling problem of two stage logistics of product oil is modeled and optimized.First,the event based continuous time modeling method is designed and implemented.The continuous-time model for batch,semi-continuous and continuous processes are studied,and the optimum selection method for number of event points is provided.The validity of the model and the selection method of event points are verified by standard examples.Second,according to the characteristics of the two stage logistics of product oil,a decomposition optimization strategy is designed,namely the problem is broken down into two stages including distribution assignment and vehicle scheduling.Distribution assignment decides the delivery tasks of product oil,and vehicle scheduling problem decides the scheduling of delivery tasks.Based on the continuous time model,the mathematical model of vehicle scheduling problem of two stage logistics for product oil is established,and the validity of the model is verified by numerical experiments using GAMS solver.2)The vehicle scheduling problem of open-pit mines is modeled and optimized.The transport process,characteristics and requirements of vehicle scheduling problem in open-pit mines are analyzed,and the mathematical model of open-pit mines vehicle scheduling problem using continuous time modeling is established.The validity of the model is verified by numerical experiments using GAMS solver.For large-scale examples,an approximate dynamic programming model is established by Approximate Dynamic Programming(ADP)algorithm based on Q-Learning.Numerical experiments of different extraction methods of feature vector and update methods of coefficient vector are carried out,and the results are compared with the results by using GAMS.The experimental results show that the ADP algorithm based on Q-Learning designed in this thesis can effectively solve the large scale open-pit mines vehicle scheduling problem.3)Based on the model and algorithm above,the vehicle scheduling algorithm optimization module of a decision support system of open-pit mines production is designed and developed.The module can meet the requirements of enterprise information management of production equipment,scheduling information configuration and the vehicle scheduling optimization,and a reasonable scheduling scheme is worked out,finally efficient and low-cost ore transportation with trucks is achieved.