A Destination Prediction Model for Individual Passengers in Urban Rail Transit

Abstract

Urban rail transit, as an important part of public transportation, plays an important role in solving the problems of large urban public traffic flow and road congestion. Real-time estimation of the destination of each individual passenger who has entered a metro station but remains in the system is of great significance for real-time passenger tracking, service recommendation, and other related applications. Traditional technology is mainly using statistical and probabilistic methods to predict user's destination, which are based on the travel information of individual passengers. However, individual passenger travel is not only affected by itself, but also affected by group travel, and other contextual factors, especially when individual users have a small amount of history travel information. Therefore, these methods cannot get best results for the prediction of destination. This paper proposes a method called as DCM based on multi-nominal logit model with highly explanatory. DCM extracts three types of features, including group features, individual features, and context features, then feed these features to a multi-nominal logit model to predict the destinations of individual passengers. We tested our method based on real data over two months in Shenzhen, China which is collected by automatic toll system, and the result shows that our method is superior to existing methods.

Cite this work

No public copy is stored for this record, so only the metadata is cited.

Stored only when both papers are in the corpus — 6 citations exist upstream.

Papers that cite the same references — the relational fallback while embeddings are still backfilling.