Artificial Neural Networks for Forecasting Passenger Flows on Metro Lines

• Published in: Sensors (Basel, Switzerland) Vol. 19; no. 15; p. 3424
• Main Authors: Gallo, Mariano, De Luca, Giuseppina, D'Acierno, Luca, Botte, Marilisa
• Format: Journal Article
• Language: English
• Published: Switzerland MDPI AG 05.08.2019; MDPI
• Subjects: Algorithms; artificial neural networks; Back propagation; Computer simulation; Counting; Deep learning; Dynamic loads; Forecasting; Intelligent transportation systems; Methods; metro; Monitoring systems; Neural networks; Passengers; Public transportation; Roads & highways; Schedules; Sensors; Short term; Subway stations; Traffic congestion; Traffic flow; transportation; Transportation networks; user flow forecast; Italy; artificial neural networks; user flow forecast; metro; transportation
• ISSN: 1424-8220; 1424-8220
• DOI: 10.3390/s19153424

Forecasting user flows on transportation networks is a fundamental task for Intelligent Transport Systems (ITSs). Indeed, most control and management strategies on transportation systems are based on the knowledge of user flows. For implementing ITS strategies, the forecast of user flows on some network links obtained as a function of user flows on other links (for instance, where data are available in real time with sensors) may provide a significant contribution. In this paper, we propose the use of Artificial Neural Networks (ANNs) for forecasting metro onboard passenger flows as a function of passenger counts at station turnstiles. We assume that metro station turnstiles record the number of passengers entering by means of an automatic counting system and that these data are available every few minutes (temporal aggregation); the objective is to estimate onboard passengers on each track section of the line (i.e., between two successive stations) as a function of turnstile data collected in the previous periods. The choice of the period length may depend on service schedules. Artificial Neural Networks are trained by using simulation data obtained with a dynamic loading procedure of the rail line. The proposed approach is tested on a real-scale case: Line 1 of the Naples metro system (Italy). Numerical results show that the proposed approach is able to forecast the flows on metro sections with satisfactory precision.