Method for dispatching public transport vehicles

Abstract

The invention discloses a method for dispatching public transport vehicles. The method comprises the following steps of: generating a block collection of departure time point sequence of a public transport vehicle corresponding to each initial departure time point according to a departure time table, the rest time of a driver, the maximum waiting time and the longest working time; generating N chromosomes with finite lengths, and making each chromosome correspond to one initial departure time point; initializing the N chromosomes to obtain the N chromosomes after initializing; making one initial departure time point in the chromosome correspond to one block; intersecting and mutating the current N chromosomes successively, and selecting the N chromosomes after mutating and the N chromosomes before intersecting to obtain the N chromosomes after selecting; intersecting, mutating and selecting to preset times so as to obtain an optimal chromosome; and regulating the optimal chromosome, dispatching the public transport vehicles according to each departure time point, cover times corresponding to the departure time point and the block after regulating. According to the method for dispatching the public transport vehicles, disclosed by the invention, the practicability and the generating efficiency of a public transport vehicle dispatching scheme can be increased.

Description

technical field [0001] The invention relates to vehicle dispatching technology in the field of traffic, in particular to a bus dispatching method. Background technique [0002] Vehicle scheduling problem (VSP) is an important field in transport scheduling. According to the constraints involved, VSP can be roughly divided into five categories: mileage constraints, model constraints, full-load constraints, non-full-load constraints, and time window constraints. The vehicle scheduling problem has a very wide range of practical application scenarios, but due to the complexity of the actual scenarios, the vehicle scheduling problem is divided into many specific problems according to different application scenarios. [0003] For the VSP with time window constraints, that is, the vehicle scheduling problem when the departure time point is determined, there are two main methods for vehicle scheduling processing, one is manual scheduling, and the other is scheduling by vehicle sched...

AI Technical Summary

Problems solved by technology

[0004] Manual scheduling requires a lot of manpower, resulting in huge operating costs and low efficiency

[0005] However, the vehicle scheduling schemes adopted by the existing vehicle scheduling systems are mainly divided into two types. The first is the sequential method, that is, the vehicle scheduling problem is solved first, and then the personnel scheduling problem is solved. Scheduling factors, the second is the integration method, that is, to simultaneously dispatch vehicles and personnel

[0006] The sequential method does not consider human factors when solving the vehicle scheduling problem, such as the driver's rest time, meal time, etc., so that the optimization of vehicle scheduling may limit the subsequent personnel scheduling, and even cannot produce an effective personnel scheduling solution, which is practical Sex is relatively low

[0007] The integration method tries to solve the personnel scheduling problem while solving the vehicle scheduling problem. In this way, when the solution model is established, the number of variables will increase a lot, so it is very difficult to solve, and the efficiency of generating the vehicle scheduling plan is too low. For example, Freling The algorithm proposed by et al. takes 86 minutes to solve the vehicle scheduling problem of a line, and the fastest one takes about 40 minutes. In practical applications, it is often encountered that the vehicle scheduling scheme needs to be regenerated due to various temporary problems. For example, if some vehicles fail, a new vehicle scheduling plan needs to be regenerated immediately. If the generation of the vehicle scheduling plan takes too long, it will undoubtedly affect the normal operation of all vehicles, thus limiting the practicability of this method

[0008] To sum up, when solving the vehicle scheduling problem, the bus scheduling schemes in the prior art generally have the problems of low practicability and low generation efficiency

Images