| 摘要: | 研究期間：10208~10307;The banks and ATMs (Automatic Teller Machines) in Taiwan have rapidly enlarged, increasing the risk of robbery during the period of transporting cash. Based on past records of robbery, the main reason of robbery is that the already planned vehicle routes/schedules are usually fixed, making it easy for the bad guys to track and target the vehicle. In practice, the personal experience of the decision maker is used to perform the cash pick-up and delivery vehicle routing/scheduling, which lacks a system optimization, often wasting resources and increasing risk of robbery. Besides, to simplify the complex routing/scheduling process, the carriers often use an average travel time for each trip to construct the vehicle routes/schedules, which neglects the stochastic travel times for trips that often occur in real operations. If stochastic travel times for trips occur during the period of transporting cash, then the already planned vehicle routines/schedules may be disturbed to lose their optimality. Therefore, we will consider the variations in vehicle routes/schedules and the arrival times at each demand point, under deterministic and stochastic travel times, to respectively develop a useful model with a solution algorithm, in order to help carriers reduce the risk of robbery. After a preliminary evaluation, this research is expectedly complicated, so we will propose it as a two-year project. In the first year, we will construct the cash pick-up and delivery vehicle routing/scheduling under deterministic travel times. In this model, to increase the conveyance safety, a new concept of similarity of time and space for vehicle routing/scheduling and a time-space network flow technique are employed to formulate the variant vehicle routing/scheduling problem. Since the model is expected to be an NP-hard problem and can not be optimally solved in a reasonable time for large-scale problems, a solution algorithm will be developed to efficiently solve the problem. In the second year, we will construct the cash pick-up and delivery vehicle routing/scheduling problem under stochastic travel times. To increase the safety during the period of transporting cash, we will also employ the concept of similarity of time and space for vehicle routing/scheduling and a time-space network flow technique to formulate the variant vehicle routing/scheduling problem. The model is also expected to be an NP-hard problem and a solution algorithm will be developed by modifying the solution algorithm developed in the first year to efficiently solve the problem. In addition, a method will be developed to evaluate the performance of the two models in simulated real operations. Finally, conclusions and suggestions will be drawn. |
