A Method for Measuring Congestion Degree of Subway Network Based on PCA Algorithm

In recent years, the level of development of public transport has become an important factor to measure the city construction and development. Subway is an important part of urban public transport. Improving subway's intelligent management level plays an important role in enhancing the convenience and comfort of residents. It is also an important part of building a smart city. In this paper, a new evaluation index of subway congestion degree is proposed, taking the road traffic congestion index as a reference. Based on the massive travel data of Shenzhen metro, the full load ratio between adjacent stations in four weeks is calculated in one-hour granularity, and the comprehensive evaluation model is constructed by principal component analysis algorithm. Finally, according to the full load ratio between adjacent stations at the current time, the current congestion index of subway network is obtained by using the comprehensive evaluation model. According to the congestion index, the subway congestion level is divided into 4 grades. Proposing the subway congestion index will help the subway administration grasp the congestion level of the current subway network. At the same time, it can also guide the public to choose the right travel time according to the current congestion level to alleviate the traffic pressure during the peak period.