Health diagnosis of bus operation based on multi‐source data

Abstract

Based on the multi-source data available for bus operations, this paper proposes a health diagnosis system for single-line bus operation systems from two aspects: The operation efficiency and stability. Firstly, the index weight has been defined and calculated based on the Entropy Method. The composite index of bus operation has been obtained, and the health classification standards that pertain to efficient and effective bus operations have also been constructed (Very Healthy, Healthy, Sub-Healthy, and Unhealthy). Secondly, the more efficient machine learning method has been used in order to establish the classification algorithm training set. The effect of the k-Nearest Neighbour and Decision Tree Classification Model has also been compared and analysed in this particular study. Finally, a bus line in Foshan is taken as a case study to verify the effectiveness of the method. This paper can effectively improve the diagnosis efficiency and accuracy by introducing the artificial intelligence algorithm into bus operation diagnosis. It provides a foundation for the development of bus operation health diagnosis decision support system with the function of “bus disease” prevention and treatment.