Simulation-based bus scheduling optimization method and system and storable medium

Abstract

The invention discloses a simulation-based bus scheduling optimization method and system, a storable medium and a storable medium, and relates to the technical field of public transportation, and the method comprises the steps: collecting passenger flow instrument data; fitting a passenger arrival rule based on the passenger flow instrument data, establishing a passenger flow simulation generation model fitting an actual situation, and obtaining simulated passenger flow data; simulating the running state of a line based on the simulated passenger flow data; and gradually optimizing the departure time of each train number along the gradient direction of the independent variable based on the simulated passenger flow data and the running state of the line, obtaining a target function, and obtaining an optimal scheduling scheme. According to the steepest rise-Bayesian optimization algorithm constructed by the method, the advantage of high precision of the Bayesian optimization algorithm is integrated, the defect of low efficiency of a response surface analysis method is avoided, the timeliness of the optimization model constructed based on a simulation mode is greatly improved, the universality is high, and departure scheme optimization scenes of different actual conditions can be directly fit.

Description

technical field [0001] The present invention relates to the technical field of public transportation, and more specifically relates to a simulation-based bus scheduling optimization method, system and storage medium. Background technique [0002] For bus scheduling frequency optimization, theoretical research can be divided into the following three perspectives: [0003] (1) Genetic Algorithm: Calculate the possibility of a non-stop strategy based on real-time passenger demand or generate a schedule to minimize passenger travel time and the number of vehicles required for operation; [0004] (2) Goal planning: use mixed integer linear programming model, heuristic algorithm or multi-objective optimization model, bi-level programming model for transmission synchronization planning of bus schedule; [0005] (3) Simulation software: Based on various simulation software, a unified dimensional optimization model that considers the interests of multiple parties is established. ...

AI Technical Summary

Problems solved by technology

[0006] However, in practical applications, manual scheduling optimization is still the mainstream. Decision makers formulate relatively reliable preliminary plans based on past experience. However, because scheduling based on experience is too subjective, on the one hand, it takes a lot of time to feed back and adjust the plan. On the other hand, The adjustment results are often inconsistent with the actual situation, the optimization quality is low, and the bus operation process is an extremely time-consuming behavior. To analyze, evaluate and give feedback on certain behaviors in the bus operation, it usually takes a few weeks or It will be reflected after several months, that is, the timeliness of the obtained optimization plan is low

Moreover, many existing departure time optimization schemes are mostly based on specific routes, which are optimization schemes for specific situations, and have poor versatility.

The current forecasting algorithms such as Bayesian optimization algorithm and response surface analysis have shortcomings. The former focuses on the result but takes a long time, while the latter focuses on the process but has low precision, and it is difficult to find the optimal result.

[0007] To sum up, regarding the bus schedule, many scholars and experts have proposed various optimization schemes from different angles. However, in practice, due to the impact of various uncertain factors on bus scheduling , simply using the analytical model often has poor results, poor versatility, low quality of manual scheduling, and long time consumption leading to low timeliness, etc.

Images