Research on intelligent bus station vehicle throughput capacity enhancement via enhanced genetic algorithm

As critical nodes constraining corridor-wide transport capacity, station vehicle throughput capacity enhancement as the key strategy for overall corridor improvement. This paper proposes an Enhanced Genetic Algorithm-based method for intelligent bus station vehicle throughput capacity enhancement.Firstly, this method puts forward an intelligent connected bus berth prediction method based on the relationship between headway and vehicle service time at the station. Then, it explores the impact of vehicle berth prediction on passengers' behaviors from entering the station to boarding the bus, establishes a Station Pedestrian Flow Model (SPFM), and determines the optimal queuing area for passengers based on Enhanced Genetic Algorithm (EGA). Finally, it analyzes the average passenger passing time at the station and the passenger flow capacity of the station under two scenarios.The results show that: in different passenger flow density scenarios, the pedestrian passing time at the platform with berth prediction can be reduced by 27.2%-33.9% compared with that at the platform without prediction; the passenger flow capacity of the platform with prediction can be increased by 21.5%-31.6% compared with that without prediction. The berth prediction method can alleviate the problem of prolonged average passenger passing time at the station caused by the increase in inbound passenger flow, improve the station vehicle throughput capacity of the platform, and enhance public transport system operational efficiency.