Vehicle scheduling algorithm based on human-vehicle binding mode

Abstract

The invention provides a vehicle scheduling algorithm based on a human-vehicle binding mode. The vehicle scheduling algorithm is implemented by the steps of: (1) constructing a vehicle scheduling model, and solving the vehicle scheduling model to obtain a vehicle scheduling scheme; (2) regarding the vehicle scheduling scheme as an initial solution for a driver scheduling problem, and iteratively performing local search improvement; (3) introducing a destructive reconstruction method into the local search process for disturbing a current solution; (4) iteratively performing the step (2) and thestep (3) for Niter times, wherein Niter is a number of local search iterations; (5) iteratively performing the steps from (1) to (3) for Mstart times, wherein Mstart is a number of algorithm multiplestartup times; (6) improving a feasible shift chain obtained in the steps (2), establishing an SCP model, and acquiring a superior shift chain combination. The vehicle scheduling algorithm adopts theiterative local search algorithm, and introduces the destructive reconstruction method into the local search process for disturbing the current solution, thereby solving the problems of low labor efficiency of drivers, high driver cost, unbalanced driver tasks and possible increase of vehicle numbers in the traditional algorithm.

Description

technical field [0001] The invention relates to a vehicle scheduling algorithm, in particular to a vehicle scheduling algorithm based on a human-vehicle binding mode. Background technique [0002] Public transportation is an important mode of transportation for the public to travel, and it is also an important means to solve the problem of urban traffic congestion. In the public transportation operation system, vehicle scheduling is an important issue. How to effectively solve this problem is crucial to the operation efficiency and service quality of the bus company. [0003] Aiming at the problem of bus scheduling, scholars at home and abroad have conducted research on the problem of bus scheduling, and there are some research results on the problem of bus scheduling. The methods to solve the problem of vehicle scheduling can be divided into two types, namely optimization method and heuristic algorithm. The optimization method uses traditional optimization models such as ...

AI Technical Summary

Problems solved by technology

However, the existing foreign solutions and algorithms are difficult to adapt to the problem of vehicle scheduling and driver scheduling in my country. The reason is that the relationship between drivers and vehicles in foreign public transport companies is not fixed, and one driver can drive multiple vehicles in one day; Domestic public transport companies usually adopt the "person-vehicle binding" model, that is, one driver drives the same vehicle on the same working day

However, optimization algorithms are used to solve such problems in China, and they are better than manual scheduling, but there are also some shortcomings, that is, the algorithms used are not universal and do not take into account the Husky vehicles. The cost cannot reflect the needs of practical applications; at the same time, the current domestic vehicle scheduling is basically based on experience, which is not scientific, and leads to low labor efficiency and high driver costs. Unbalanced driver tasks, possibly increased number of vehicles

Images