Real-time Bus Route State Forecasting Using Particle Filter: An Empirical Data Application

Abstract

Buses on the same route tend to bunch when the system is uncontrolled. This lack of regularity leads to an increase in the average passenger waiting time, increases delays and makes travel times uncertain. A wide variety of solutions have been proposed to maintain accurate bus system performance. Unfortunately, if a strategy is applied permanently, it could detract from the entire transport system efficiency. That is why a transit operator needs an accurate forecast of the route in order to intervene before the bus route is too disrupted to be restored to regularity. This paper aims to predict critical situations in real-time forecasting of a bus route state. To accomplish this, we propose to take advantage of both theoretical and empirical information (model and data) using data assimilation (a particle filter). On one hand, a stochastic dynamic bus model forecasts future bus route states. On the other hand, archived data calibrates the model parameters while real-time data provides information about the actual route state. The methodology is applied to a real case study thanks to the quality data provided by TriMet (the Portland, Oregon transit district). Predictions are finally evaluated by an a posteriori comparison with real data. The results highlight that the method leads to a valid forecast of a bus route state with a 8 minutes time window. This duration is sufficient to predict critical situations, especially bus bunching. Further research would have to consider deterministic travel times from a traffic model instead of the distributions in order to maintain correlation between travel times on links. In that case, the assimilation process would focus on the surrounding traffic flow, also potentially available in the Portland data.

A review of bus arrival time prediction using artificial intelligence

“…The real‐time collected data help transit agency to manage operational control whereas using archived data one can analyze important characteristics and factors that help the transit agency to structure and improve its operational plan. Some studies highlight the importance of archived data to help control operations such as monitor operations for unsafe practices, load variability and improve the service plan such as to characterize unreliability (Barabino et al, 2013; Barabino et al, 2016; Furth et al, 2006; Hans et al, 2015a; Hans et al, 2015b).…”

SVM‐based hybrid approach for corridor‐level travel‐time estimation

“…The main advantage of a PF is that it does not require the process and sensor noise to be Gaussian distributed. The PFs are widely used for various application like real‐time travel‐time estimation [40–43], studies related to traffic parameter estimation [21, 44–49] and vehicular tracking [40, 48, 50, 51]. PFs are also used for estimating the probability density functions (PDFs) of the system states [52].…”

“…PFs are also used for estimating the probability density functions (PDFs) of the system states [52]. Hans et al [40] used PFs to forecast the bus trajectory in real‐time. A stochastic model was used to calculate the dwell time at bus stops and travel‐time at the links.…”

Improving High-Frequency Transit Performance through Headway-Based Dispatching: Development and Implementation of a Real-Time Decision-Support System on a Multi-Branch Light Rail Line

“…Although trips complying with the optimized departure times from the decision-support system improved service quality, only half of the trips were dispatched close to the recommended times. Improving location predictions could help predict service deterioration and thus improve the effectiveness of a control policy, especially in very high-frequency services (22,23). The human factors contributing to non-compliance, including those related to technology and culture, should also be studied in depth.…”