Detecting Transportation Modes Using Smartphone Data and GIS Information: Evaluating Alternative Algorithms for an Integrated Smart-Phone based Travel Diary Imputation

Smartphones with embedded Global Positioning System (GPS) technology provide an opportunity to passively collect individuals’ travel trajectory data, which can be utilized to identify key aspects of travel behaviour such as transport modes. This paper presents a mode detection algorithm developed to infer transport modes from location data collected using smartphones. The algorithm is designed to classify single-mode trip segments into one of six transport modes, including walk, bicycle, car, bus & streetcar, subway, and regional train. There are 31 features selected as inputs to the algorithm, which are divided into two categories: general features and transit-specific features. General features, including speed, acceleration, distance, and duration, are needed to classify all available modes. Transit-specific features representing proximity to the nearest transit lines are designed to enhance the classification of the three transit modes. Four tree-based ensemble learning models are tested and evaluated: Random Forest (RF), Gradient Boosting Decision Tree (GBDT), eXtreme Gradient Boosting (XGBoost), and Light Gradient Boosting Machine (LightGBM). The results show that LightGBM performs the best in terms of detection accuracy and computation time. A case study is conducted to test the effectiveness of the algorithm in a proposed travel survey system. It demonstrates that the algorithm can correctly detect trip segments and their modes from smartphone location data and present them in the form of map-based travel diaries.

Language

  • English

Media Info

  • Media Type: Digital/other
  • Features: Figures; Maps; References; Tables;
  • Pagination: 16p

Subject/Index Terms

Filing Info

  • Accession Number: 01764115
  • Record Type: Publication
  • Report/Paper Numbers: TRBAM-21-00415
  • Files: TRIS, TRB, ATRI
  • Created Date: Dec 23 2020 11:20AM