Lelitha Devi Vanajakshi

Real-time bus travel time prediction has been an interesting problem since past decade, especially in India. Popular methods for travel time prediction include time series analysis, regression methods, Kalman filter method and Artificial Neural Network (ANN) method. Reported studies using these methods did not consider the high variance situations arising from the varying traffic and weather conditions, which is very common under heterogeneous and lane-less traffic conditions such as the one in India. The aim of the present study is to analyse the variance in bus travel time and predict the travel time accurately under such conditions. Literature shows that Support Vector Machines (SVM) technique is capable of performing well under such conditions and hence is used in this study. In the present study, nu-Support Vector Regression (SVR) using linear kernel function was selected. Two models were developed, namely spatial SVM and temporal SVM, to predict bus travel time. It was observed that in high mean and variance sections, temporal models are performing better than spatial. An algorithm to dynamically choose between the spatial and temporal SVM models, based on the current travel time, was also developed. The unique features of the present study are the traffic system under consideration having high variability and the variables used as input for prediction being obtained from Global Positioning System (GPS) units alone. The adopted scheme was implemented using data collected from GPS fitted public transport buses in Chennai (India). The performance of the proposed method was compared with available methods that were reported under similar traffic conditions and the results showed a clear improvement.

The accuracy of travel time information given to passengers plays a key role in the success of any Advanced Public Transportation Systems (APTS) application. In order to improve the accuracy of such applications, one should carefully develop a prediction method. A majority of the available prediction methods considered the variation in travel time either spatially or temporally. The present study developed a prediction method that considers both temporal and spatial variations in travel time. The conservation of vehicles equation in terms of flow and density was first re-written in terms of speed in the form of a partial differential equation using traffic stream models. Then, the developed speed based equation was discretized using the Godunov scheme and used in the prediction scheme that was based on the Kalman filter. From the results, it was found that the proposed method was able to perform better than historical average, regression, and ANN methods and the methods that considered either temporal or spatial variations alone. Finally, a formulation was developed to check the effect of side roads on prediction accuracy and it was found that the additional requirement in terms of location based data did not result in an appreciable change in the prediction accuracy. This clearly demonstrated that the proposed approach based on using vehicle tracking data is good enough for the considered application of bus travel time prediction.

In recent times, bus arrival time prediction (BATP) systems are gaining more popularity in the field of advanced public transportation systems, a major functional area under intelligent transportation systems. BATP systems aim to predict bus arrival times at various bus stops and provide the same to passenger’s pre-trip or while waiting at bus stops. A BATP system, which is accurate, is expected to attract more commuters to public transport, thus helping to reduce congestion. However, such accurate prediction of bus arrival still remains a challenge, especially under heterogeneous and lane-less traffic conditions such as the one existing in India. The uncertainty associated with such traffic is very high and hence the usual approach of prediction based on average speed will not be enough for accurate prediction. To make accurate predictions under such conditions, there is a need to identify correct inputs and suitable prediction methodology that can capture the variations in travel time. To accomplish the above goal, a robust framework relying on data analytics is proposed in this study. The spatial and temporal patterns in travel times were identified in real time by performing cluster analysis and the significant inputs thus identified were used for the prediction. The prediction algorithm used the Adaptive Kalman Filter approach, to take into account of the high variability in travel time. The proposed schemes were corroborated using real-world GPS data and the results obtained are very promising.

Passenger information systems about bus arrival at bus stops, which is an integral part of any Advanced Public Transportation Systems (APTS) application, is catching the attention of traffic engineers in India in recent years. APTS, by making the public transport system more attractive, will encourage people shift from personal mode to public mode for their transport, thus relieving congestion. There are different approaches of APTS that try to make public transport systems more desirable to commuters and one among them is accurate arrival time prediction at bus stops. This will reduce the wait time and associated uncertainties. There have been many studies reported which looked into the problem of bus arrival time prediction or travel time prediction. However, studies which deal with automatic incorporation of delays explicitly into travel time prediction are limited. Further, studies focusing on travel time prediction under heterogeneous traffic conditions are scarce. The present study attempts to identify delays automatically and explicitly incorporate them in predicting the total travel time of buses under heterogeneous traffic conditions such as those existing in India. The results obtained are corroborated with actual data and found to be promising. © 2009 IEEE.

Accurate travel time information of public transport will help operators to effectively manage and implement their operating strategies and help passengers by reducing the uncertainty about arrival time of buses at bus stops. The reliability of such information provided to passengers greatly depends on the prediction technique used, which in turn, depends on the quality of the input used in the prediction technique. In other words, identifying and using the correct input in the appropriate prediction technique is important. Prediction techniques can be data driven or less data intensive. The first part of this paper presents a systematic statistical approach for identifying the significant inputs for travel time prediction. The second part compares the performance of two popular prediction methods, one being the data driven Artificial Neural Network (ANN) method and the other being a model based approach using the Kalman Filter Technique (KFT) that is less data intensive, to predict bus travel time. The performances of both methods were evaluated using the data obtained from the field. It was found that ANN outperformed KFT in terms of prediction error, if a good database is available, and in case of limited data availability, KFT will be more advantag eous.

Predicting bus arrival times and travel times are crucial elements to make the public transport more attractive and reliable. The present study explores the use of Intelligent Transportation Systems (ITS) to make public transportation systems more attractive by providing timely and accurate travel time information of transit vehicles. However, for such systems to be successful, the prediction should be accurate, which ultimately depends on the prediction method as well as the input data used. In the present study, to identify significant inputs, a data mining technique, namely k-NN classifying algorithm is used. It is based on the similarity in pattern between the input and historic data. These identified inputs are then used for predicting the travel time using a model-based recursive estimation scheme, based on Kalman filtering. The performance is evaluated and compared with methods based on static inputs, to highlight the improved prediction accuracy.