Integrating Bluetooth and smart card data for better estimation and prediction of bus speed on arterial corridors with low frequency buses

Data driven based travel speed (or travel time) short term prediction models require accurate estimation of the historical time series with equally spaced data points. The availability of the bus speed time series data points depends on the bus frequency and other operational factors such as on-time performance. Low frequency bus routes, coupled with bad on-time performance can result in time series with number of missing values (or irregular interval of data points). Addressing the above need, this paper explores the relationship between bus and car speed and utilises the understanding to better estimate bus travel speed time series and its application for short term prediction of bus speed. With a case study on a Brisbane corridor, the car speed is estimated using Bluetooth MAC Scanner (BMS) and bus speed is estimated using Automatic Fare Collection data (Go card). The findings are encouraging and the results of the integration of the two data sources indicate around 3% improvement in the bus speed estimation compared to the case where the time series gaps are filled with linear interpolation. Furthermore, the prediction results are also improved for different prediction horizons. This paper will assist transit operators to exploit Bluetooth data to augment the performance of low frequency buses by estimating and predicting more accurate bus speed (travel time)