Abstract
Developing public transport is an effective way to solve traffic congestion and improve travel efficiency. Improving bus service quality can attract passengers to travel by public transport. In the past, as the bus arrival time is unknown, and the buses often arrive inaccurate, passengers feel anxious and the quality of public transport service declined. Even though some bus stations equipped with electronic bus stop boards, the predicted bus arrival time is often inaccurate. Therefore, in order to convenient for people travel by public transit, this paper puts forward the method of bus travel time prediction based on the Markov chain which considers the spatial-temporal characteristics of the bus travel time. The prediction method can improve the quality of the bus service, help the travelers to make travel planning and reduce the waiting time. The algorithm is verified by the actual operation data of No.114 bus line in Harbin. The results show that the prediction error is small, and the algorithm is easy to implement.
Highlights
Background of algorithmThe operation of the bus is related to temporal and spatial
As there are many internal and external influence factors of the bus system and the influence factors are quite complex, the existing bus arrival time prediction systems are inaccurate, and some errors even reach more than 200%
The travel time between bus stops can be used as the state of a specific bus route system
Summary
The problem of urban traffic is becoming more and more prominent. Traffic congestion, traffic accidents, environmental pollution, energy consumption and other issues have become the serious urban diseases. Many large cities in China have carried out research and practice on intelligent electronic bus stop board and public transport service information platform. It includes: (1) Entity electronic bus stop board. By developing computer and mobile phone client software, the users can inquire bus information in real time. It includes operation status, the bus stop numbers of the coming bus, distance and time of the vehicles. The bus stop numbers of the coming bus, distance and time of the vehicles It provides real-time services for transit users. The method of bus arrival time prediction mainly includes, historical averages and real-time GPS data weighted prediction method by Sun [2], neural network model [3, 4], Calman filter [5, 6], support vector machine (SVM) model [7] and so on
Sign-in/Register to view highlights & summary Sign-in/Register to access full text options 
Free) 
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a call
