no150-152

[151] Long-term Variations of Spatio-temporal Characteristics for Railroad Usage using Traffic IC Card Data

○Mio Hosoe¹, Masashi Kuwano¹, Kosuke Miyazaki² (1. Tottori University, 2. Kagawa National College of Technology)

Keywords:Long-term Variations, Information Compression, Structural Changes, Smart Card, Public Transportation

It is important to grasp public transportation usage and its variation. The purpose of this study is to reveal long-term variation patterns of railroad users from smart card data which is used Kotoden train in Kagawa prefecture, Japan. This study proposes the method by penalized non-negative tensor factorization and piecewise liner approximation model. As a result applied the proposed method, this study succeeded in specifying characteristic movement patterns which are the combinations of the riding day, the origin station and the destination station of each trip. Also, this study revealed long term variation of movement patterns.