An Adaptive Long-Term Bus Arrival Time Prediction Model with Cyclic Variations

Real-time bus arrival information systems at transit stops can be useful to passengers for efficient trip planning and reducing waiting times. The accuracy of such systems depends upon the ability of the model to account for variations in the data series and to adjust according to changing traffic conditions. Many of the existing studies on passenger information systems have modeled the system based on stationary relations, not taking into account the cyclic variations in data, which is often suitable for demonstration purposes but not for long-term implementation. The present study models the changing relationships using Double-Seasonal Holt-Winter’s Exponential Smoothing approach, which allows for self-updating parameters at four levels. It accounts for both long-term and short-term seasonal fluctuations in data while maintaining the dynamic treatment of real-time bus information. The model also takes into account delays using the real-time running information of the bus and incorporates it into subsequent forecasts for better accuracy. Real-time data from GPS transmitters in buses were used for validation of the proposed model. The results show that the proposed model performs better than the currently-used elementary field methods and is able to forecast bus travel times with a reasonable accuracy. Introduction Traffic congestion is a problem that affects transportation networks in many major cities in terms of reduced mobility and system reliability. In practice, particularly in transit-oriented cities, passengers often are faced with a choice between different modes of transport—for instance, buses and metro rail in Delhi. Bus systems often suffer a disadvantage, as rail and other transportation networks often have fixed schedules that are fairly accurate. In densely-populated cities (e.g., Mumbai, Delhi, and Kolkata), with heavy passenger demands, real-time Passenger Information System (PIS) on arrival times of buses can be An Adaptive Long-Term Bus Arrival Time Prediction Model with Cyclic Variations Journal of Public Transportation, Vol. 18, No. 1, 2015 2 very helpful to the commuters. This can prove to be useful, particularly at major bus stops with large numbers of transfer passengers to other modes (rail, BRT etc.). The availability of real-time transit arrival information can help in efficient trip planning and making smart choices for travel. Further, real-time information can attract potential transit users to the system due to the inherent advantages of the system. In recent times, advances in the field of Information and Communication Technologies (ICT) have opened up new avenues for integrating real-time information in passenger information systems. Recently, transit agencies also have started implementing ICT-related technologies in their operations to help in improving operational efficiency and the quality of information (Yu 2011). Such advanced systems are lacking in most Indian cities, and the few that have been implemented are very elementary in nature and have focused largely on technology demonstration. However, the reliability of such systems greatly depends on the ability of the system to account for different variations in the data series and to adjust to dynamic traffic conditions. Hence, there is an urgent need to develop a robust real-time bus arrival time prediction model that can be applied under dynamic Indian conditions (Vanjakshi et al. 2009; Ramakrishna et al. 2006; Padmanaban 2010) Literature Review In the recent past, various sophisticated techniques and algorithms have been employed to forecast bus travel time or arrival time by using Global Positioning Systems (GPS)or bus triangulation data. These may be categorized as elementary (non-recursive) or advanced (recursive algorithms) based on the way they process the input data. Elementary Algorithms A variety of elementary approaches have been applied in the past studies to forecast travel times. Some of these include: