Hybrid model for prediction of bus arrival times at next station

Abstract

SUMMARYEffective prediction of bus arrival times is important to advanced traveler information systems (ATIS). Here a hybrid model, based on support vector machine (SVM) and Kalman filtering technique, is presented to predict bus arrival times. In the model, the SVM model predicts the baseline travel times on the basic of historical trips occurring data at given time-of-day, weather conditions, route segment, the travel times on the current segment, and the latest travel times on the predicted segment; the Kalman filtering-based dynamic algorithm uses the latest bus arrival information, together with estimated baseline travel times, to predict arrival times at the next point. The predicted bus arrival times are examined by data of bus no. 7 in a satellite town of Dalian in China. Results show that the hybrid model proposed in this paper is feasible and applicable in bus arrival time forecasting area, and generally provides better performance than artificial neural network (ANN)-based methods.

Bus arrival time prediction using mixed multi-route arrival time data at previous stop

“…They compared the performance of these two models and found both had their own advantages. Similar to Chien's et al (2000) study, Yu et al (2010) presented a hybrid model and found his model generally provides better performance than conventional ANN method. More recently, Lin et al (2013) and Khetarpaul et al (2015) proposed hybrid arrival time prediction models combining ANN and clustering methods to capture the traffic fluctuations and determine the parameter inputs of different clusters more clearly.…”

“…The previous studies also prove that situation 1 is more popular (Shalaby, Farhan 2004;Zheng et al 2012;Yu et al 2006Yu et al , 2010Yu et al , 2011. However, situation 2 has two unique advantages:…”

“…In model calibration step, the historical bus arrival time data picked up of all the selected segments are divided into two parts: 80% are training data that are used for model calibration, and the rest 20% are adopted to test the prediction performance of each model. For SVM algorithm, previous studies (Yu et al 2006(Yu et al , 2010 suggested that Radial Basis Function (RBF) kernel was efficient for bus arrival time prediction. So in this paper, RBF is also selected as kernel for SVM.…”

A Hybrid Model Based on Support Vector Machine for Bus Travel-Time Prediction

“…SVM was developed by Vapnik [14,15], which is characterized by a specific type of learning algorithms. It has been successfully applied to solve some classic problems, such as incident detection [16], traffic-pattern recognition [17], passenger head recognition [18], and travel time prediction [7,8,[19][20][21]. SVM is used to find regular pattern and makes use of them to analyse the unknowns.…”

Dependence of Parking Pricing on Land Use and Time of Day

“…According to statistics on Beijing's business zones, the average radius of such a zone is approximately 1 km. The average riding time for a trip by bus or subway from a parking location outside a business zone to a travel destination in the business zone was calculated as 8.00 min for parking during peak hours and 6.00 min for parking during off-peak hours based on the Beijing parking survey data [36][37][38][39]. In addition, according to the parking survey data, the average walking time (from the parking location to the destination) after parking in a business zone is approximately 3.14 min, and the average time required to walk from outside a business zone to a travel destination in the business zone was set at 8.00 min [40][41][42].…”