Hybrid model for prediction of bus arrival times at next station

Effective 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. Copyright © 2010 John Wiley & Sons, Ltd.     

Bus arrival time prediction at bus stop with multiple routes

Provision of accurate bus arrival information is vital to passengers for reducing their anxieties and waiting times at bus stop. This paper proposes models to predict bus arrival times at the same bus stop but with different routes. In the proposed models, bus running times of multiple routes are used for predicting the bus arrival time of each of these bus routes. Several methods, which include support vector machine (SVM), artificial neural network (ANN), k nearest neighbours algorithm (k-NN) and linear regression (LR), are adopted for the bus arrival time prediction. Observation surveys are conducted to collect bus running and arrival time data for validation of the proposed models. The results show that the proposed models are more accurate than the models based on the bus running times of single route. Moreover, it is found that the SVM model performs the best among the four proposed models for predicting the bus arrival times at bus stop with multiple routes.     

Bus running time prediction using a statistical pattern recognition technique

Abstract

Given that real-time bus arrival information is viewed positively by passengers of public transit, it is useful to enhance the methodological basis for improving predictions. Specifically, data captured and communicated by intelligent systems are to be supplemented by reliable predictive travel time. This paper reports a model for real-time prediction of urban bus running time that is based on statistical pattern recognition technique, namely locally weighted scatter smoothing. Given a pattern that characterizes the conditions for which bus running time is being predicted, the trained model automatically searches through the historical patterns which are the most similar to the current pattern and on that basis, the prediction is made. For training and testing of the methodology, data retrieved from the automatic vehicle location and automatic passenger counter systems of OC Transpo (Ottawa, Canada) were used. A comparison with other methodologies shows enhanced predictive capability.     

Online Prediction of Travel Time: Experience From a Pilot Trial

Abstract

This study was designed to present an online model which predicted travel times on an interurban two-lane two-way highway section on the basis of field measurements. The study included two parts: an evaluation of the performance of the model, and an examination of the possibility to improve the model in case of unsatisfactory performance. The model was based on MLP neural networks. The main results of the evaluation showed that the prediction model outperformed a non-predictive system. However, the model for one section had not performed as well during the trial period as was expected. This might be due to a slight change in the congestion phenomenon. After further development, the findings showed that the model could be improved considerably with new data. The main implication was that even a simple prediction model improves the quality of travel time information substantially, compared to estimates based directly on the latest measurements.     

Bus control strategies in corridors with signalized intersections

This paper proposes a new dynamic bus control strategy aimed at reducing the negative effects of time-headway variations on route performance, based on real-time bus tracking data at stops. In routes with high demand, any delay of a single vehicle ends up causing an unstable motion of buses and producing the bus bunching phenomena. This strategy controls the cruising speed of buses and considers the extension of the green phase of traffic lights at intersections, when a bus is significantly delayed. The performance of this strategy will be compared to the current static operation technique based on the provision of slack times at holding points. An operational model is presented in order to estimate the effects of each controlling strategy, taking into account the vehicle capacity constraint. Control strategies are assessed in terms of passenger total travel time, operating cost as well as on the coefficient of headway variation. The effects of controlling strategies are tested in an idealized bus route under different operational settings and in the bus route of highest demand in Barcelona by simulation. The results show that the proposed dynamic controlling strategy reduces total system cost (user and agency) by 15–40% as well as the coefficient of headway variation 53–78% regarding the uncontrolled case, providing a bus performance similar to the expected when time disturbance is not presented.     

International experience with speed limits during and prior to the energy crisis of 1973–4

This paper presents an artificial neural network (ANN) based method for estimating route travel times between individual locations in an urban traffic network. Fast and accurate estimation of route travel times is required by the vehicle routing and scheduling process involved in many fleet vehicle operation systems such as dial‐a‐ride paratransit, school bus, and private delivery services. The methodology developed in this paper assumes that route travel times are time‐dependent and stochastic and their means and standard deviations need to be estimated. Three feed‐forward neural networks are developed to model the travel time behaviour during different time periods of the day‐the AM peak, the PM peak, and the off‐peak. These models are subsequently trained and tested using data simulated on the road network for the City of Edmonton, Alberta. A comparison of the ANN model with a traditional distance‐based model and a shortest path algorithm is then presented. The practical implication of the ANN method is subsequently demonstrated within a dial‐a‐ride paratransit vehicle routing and scheduling problem. The computational results show that the ANN‐based route travel time estimation model is appropriate, with respect to accuracy and speed, for use in real applications.     

Evaluating the added-value of online bus arrival prediction schemes

Online predictions of bus arrival times have the potential to reduce the uncertainty associated with bus operations. By better anticipating future conditions, online predictions can reduce perceived and actual passenger travel times as well as facilitate more proactive decision making by service providers. Even though considerable research efforts were devoted to the development of computationally expensive bus arrival prediction schemes, real-world real-time information (RTI) systems are typically based on very simple prediction rules. This paper narrows down the gap between the state-of-the-art and the state-of-the-practice in generating RTI for public transport systems by evaluating the added-value of schemes that integrate instantaneous data and dwell time predictions. The evaluation considers static information and a commonly deployed scheme as a benchmark. The RTI generation algorithms were applied and analyzed for a trunk bus network in Stockholm, Sweden. The schemes are assessed and compared based on their accuracy, reliability, robustness and potential waiting time savings. The impact of RTI on passengers waiting times are compared with those attained by service frequency and regularity improvements. A method which incorporates information on downstream travel conditions outperforms the commonly deployed scheme, leading to a 25% reduction in the mean absolute error. Furthermore, the incorporation of instantaneous travel times improves the prediction accuracy and reliability, and contributes to more robust predictions. The potential waiting time gains associated with the prediction scheme are equivalent to the gains expected when introducing a 60% increase in service frequency, and are not attainable by service regularity improvements.     

A genetic algorithm-based method for improving quality of travel time prediction intervals

The transportation literature is rich in the application of neural networks for travel time prediction. The uncertainty prevailing in operation of transportation systems, however, highly degrades prediction performance of neural networks. Prediction intervals for neural network outcomes can properly represent the uncertainty associated with the predictions. This paper studies an application of the delta technique for the construction of prediction intervals for bus and freeway travel times. The quality of these intervals strongly depends on the neural network structure and a training hyperparameter. A genetic algorithm–based method is developed that automates the neural network model selection and adjustment of the hyperparameter. Model selection and parameter adjustment is carried out through minimization of a prediction interval-based cost function, which depends on the width and coverage probability of constructed prediction intervals. Experiments conducted using the bus and freeway travel time datasets demonstrate the suitability of the proposed method for improving the quality of constructed prediction intervals in terms of their length and coverage probability.     

A dynamic route choice model for public transport networks with boarding queues

Abstract

The concepts of optimal strategy and hyperpath were born within the framework of static frequency-based public transport assignment, where it is assumed that travel times and frequencies do not change over time and no overcrowding occurs. However, the formation of queues at public transport stops can prevent passengers from boarding the first vehicle approaching and can thus lead to additional delays in their trip. Assuming that passengers know from previous experience that for certain stops/lines they will have to wait for the arrival of the 2nd, 3rd, …, k-th vehicle, they may alter their route choices, thus resulting in a different assignment of flows across the network. The aim of this paper is to investigate route choice behaviour changes as a result of the formation and dispersion of queues at stops within the framework of optimal travel strategies. A new model is developed, based on modifications of existing algorithms.     

Estimating origin‐destination matrices from link volume counts: a simple intercity auto travel demand model for Sri Lanka

The cost of nation wide travel surveys is high. Hence in many developing countries, planners have found it difficult to develop intercity transportation plans due to the non availability of origin‐destination trip matrices. This paper will describe a method for the intercity auto travel estimation for Sri Lanka with link traffic volume data.

The paper outlines the rationale of selecting the district capitals of Sri Lanka as its “cities,” the methodology for selecting the intercity road network, determination of link travel times from express bus schedules and the location of link volume counting positions.

Initially, the total auto travel demand model is formulated with various trip purpose sub‐models. This model is finally modified to a simple demand model with district urban population and travel times between city pairs as the exogenous variables, to overcome statistical estimation difficulties. The final demand model has statistics within the acceptable regions.

The advantages of a simple model are discussed and possible extensions are proposed.     

Using automatic passenger counter data in bus arrival time prediction

Artificial neural networks have been used in a variety of prediction models because of their flexibility in modeling complicated systems. Using the automatic passenger counter data collected by New Jersey Transit, a model based on a neural network was developed to predict bus arrival times. Test runs showed that the predicted travel times generated by the models are reasonably close to the actual arrival times.     

A demand model for recreational travel

A model is developed to describe and to predict the patterns of regional recreational travel. The model is designed in such a manner to allow its calibration and use without the need to conduct extensive travel surveys in a large region. To allow its use for prediction, the model is based on a causal structure and attempts to derive recreational travel demand from behavioural variables. The main hypothesis of the model is that the amount of recreational travel a recreation area attracts is affected by the accessibility of this area to points of demand potential and by its attractiveness relative to the recreation areas.

The calibration is founded on actual data on recreational travel to national forests in California, U.S.A. It is found in the calibration that accessibility to demand potential is the single most important determinant of recreational travel attraction. A simple relationship is derived to relate travel to each national forest to the relative accessibility of the forest. The model is calibrated and statistically validated.

It is suggested that when constructing travel demand models simplicity be sought, even at the risk of the loss of some explanatory power. In the calibration of such models statistical significant is more important than the ability to reproduce observed patterns.     

Analytical model for calibrating delay at congested bus stops

Abstract

A model is proposed to calculate the overall operating and delay times spent at bus stops due to passenger boarding and alighting and the time lost to queuing caused by bus stop saturation. A formula for line demand at each stop and the interaction between the buses themselves is proposed and applied to different bus stops depending on the number of available berths. The application of this model has quantified significant operational delays suffered by users and operator due to consecutive bus arrival at stops, even with flows below bus stop capacity.     

Quantifying the effects of driver non-compliance and communication system failure in the performance of real-time bus control strategies

Control strategies that prevent bus bunching allow for improvement to the level of service offered by a transit corridor as well as reducing travel time and its variability, thus providing higher reliability to the user. Several optimization models based on the use of real-time information have been shown to achieve this, through the planning of holding of the buses at bus stops. In the majority of the cases the benefits of these models have been estimated assuming ideal operational conditions while only few of them have been tested in real conditions. However, neither the simulation experiment, nor the real implementations have quantified the effects of real-life phenomena that harm the performance of the system, preventing it from achieving the full potential of these control schemes.This paper examines three phenomena that may occur during the operation of a bus service, which would limit the effectiveness of a holding-based control strategy in the sense that some of the planned holdings might not be executed. These phenomena are drivers non-compliance, failure of communication systems with buses, and the combination of both. The objective is to estimate the negative impact these phenomena can have on the benefits of the strategy, and to identify possible measures that could help operators and decision makers to reduce this impact. Both objectives are achieved using the real-time holding model developed by Delgado et al. (2012), which is tested in a simulation environment.     

Passenger travel characteristics and bus operational states: a study based on IC card and GPS data in Yinchuan,China

ABSTRACT

In recent years, public transport has been developing rapidly and producing large amounts of traffic data. Emerging big data-mining techniques enable the application of these data in a variety of ways. This study uses bus intelligent card (IC card) data and global positioning system (GPS) data to estimate passenger boarding and alighting stations. First, an estimation model for boarding stations is introduced to determine passenger boarding stations. Then, the authors propose an innovative uplink and downlink information identification model (UDI) to generate information for estimating alighting stations. Subsequently, the estimation model for the alighting stations is introduced. In addition, a transfer station identification model is also developed to determine transfer stations. These models are applied to Yinchuan, China to analyze passenger flow characteristics and bus operations. The authors obtain passenger flows based on stations (stops), bus lines, and traffic analysis zones (TAZ) during weekdays and weekends. Moreover, average bus operational speeds are obtained. These findings can be used in bus network planning and optimization as well as bus operation scheduling.     

Real time queue length estimation for signalized intersections using travel times from mobile sensors

We study how to estimate real time queue lengths at signalized intersections using intersection travel times collected from mobile traffic sensors. The estimation is based on the observation that critical pattern changes of intersection travel times or delays, such as the discontinuities (i.e., sudden and dramatic increases in travel times) and non-smoothness (i.e., changes of slopes of travel times), indicate signal timing or queue length changes. By detecting these critical points in intersection travel times or delays, the real time queue length can be re-constructed. We first introduce the concept of Queue Rear No-delay Arrival Time which is related to the non-smoothness of queuing delay patterns and queue length changes. We then show how measured intersection travel times from mobile sensors can be processed to generate sample vehicle queuing delays. Under the uniform arrival assumption, the queuing delays reduce linearly within a cycle. The delay pattern can be estimated by a linear fitting method using sample queuing delays. Queue Rear No-delay Arrival Time can then be obtained from the delay pattern, and be used to estimate the maximum and minimum queue lengths of a cycle, based on which the real-time queue length curve can also be constructed. The model and algorithm are tested in a field experiment and in simulation.     

Machine learning applications in activity-travel behaviour research: a review

ABSTRACT

This paper reviews the activity-travel behaviour literature that employs Machine Learning (ML) techniques for empirical analysis and modelling. Machine Learning algorithms, which attempt to build intelligence utilizing the availability of large amounts of data, have emerged as powerful tools in the fields of pattern recognition and big data analysis. These techniques have been applied in activity-travel behaviour studies since the early ’90s when Artificial Neural Networks (ANN) were employed to model mode choice decisions. AMOS, an activity-based modelling system developed in the mid-’90s, has ANN at its core to model and predict individual responses to travel demand management measures. In the dawn of 2000, ALBATROSS, a comprehensive activity-based travel demand modelling system, was proposed by Arentze and Timmermans using Decision Trees. Since then researchers have been exploring ML techniques like Support Vector Machines (SVM), Decision Trees (DT), Neural Networks (NN), Bayes Classifiers, and more recently, Ensemble Learners to model and predict activity-travel behaviour. A large number of publications over the years and an upward trend in the number of published articles over time indicate that Machine Learning is a promising tool for activity-travel behaviour analysis and prediction. This article, first of its kind in the literature, reviews these studies and explores the trends in activity-travel behaviour research that apply ML techniques. The review finds that mode choice decisions have received wide attention in the literature on ML applications. It was observed that most of the studies identify the lack of interpretability as a serious shortcoming in ML techniques. However, very few studies have attempted to improve the interpretability of the models. Further, some studies report the importance of feature engineering in ML-based studies, but very few studies adopt feature engineering before model development. Spatiotemporal transferability of models is another issue that has received minimal attention in the literature. In the end, the paper discusses possible directions for future research in the area of activity-travel behaviour modelling using ML techniques.     

What Moves Us? An Interdisciplinary Exploration of Reasons for Traveling

Abstract

We review a number of theories of motivation, and typologies of motivations, in psychological theory and in application to a variety of specific contexts, including shopping, eating, leisure, tourism, and travel. A recurring theme is the distinction between extrinsic (instrumental, utilitarian, functional) and intrinsic (autotelic, hedonic, experiential) motivations. We suggest that travel is a behavior to which intrinsic motivations apply, and that focusing exclusively on the extrinsic motivations to travel runs the risk of substantially underestimating the demand for travel, and the resistance to policies attempting to reduce it or to technologies (notably, information and communication technologies) expected to (partly) replace it. We offer a number of suggestions for improving standard travel surveys to help obtain the data needed to explore intrinsic motivations more fully. As better data become available, travel behavior models can be refined to partly account for such motivations. We believe that the resulting insights will be extremely valuable to policy-makers, planners, and behavioral scholars.     

Valuing crowding in public transport: Implications for cost-benefit analysis

This paper investigates the valuation of crowding in public transport trips and its implications in demand estimation and cost-benefit analysis. We use a choice-based stated preference survey where crowding levels are represented by means of specially designed pictures, and use these data to estimate flexible discrete choice models. We assume that the disutility associated with travelling under crowded conditions is proportional to travel time. Our results are consistent with and extend previous findings in the literature: passenger density has a significant effect on the utility of travelling by public transport; in fact, the marginal disutility of travel time in a crowded vehicle (6 standing-passengers/m²) is 2.5 times higher than in a vehicle with available seats. We also compare the effects of different policies for improving bus operations, and the effect of adding crowding valuation in cost-benefit analysis. In doing that, we endogenise the crowding level as the result of the equilibrium between demand and supplied bus capacity. Our results indicate that important benefits may be accrued from policies designed to reduce crowding, and that ignoring crowding effects significantly overestimate the bus travel demand the benefits associated with pure travel time reductions.     

Using survival models to estimate bus travel times and associated uncertainties

Transit agencies often provide travelers with point estimates of bus travel times to downstream stops to improve the perceived reliability of bus transit systems. Prediction models that can estimate both point estimates and the level of uncertainty associated with these estimates (e.g., travel time variance) might help to further improve reliability by tempering user expectations. In this paper, accelerated failure time survival models are proposed to provide such simultaneous predictions. Data from a headway-based bus route serving the Pennsylvania State University-University Park campus were used to estimate bus travel times using the proposed survival model and traditional linear regression frameworks for comparison. Overall, the accuracy of point estimates from the two approaches, measured using the root-mean-squared errors (RMSEs) and mean absolute errors (MAEs), was similar. This suggests that both methods predict travel times equally well. However, the survival models were found to more accurately describe the uncertainty associated with the predictions. Furthermore, survival model estimates were found to have smaller uncertainties on average, especially when predicted travel times were small. Tests for transferability over time suggested that the models did not over-fit the dataset and validated the predictive ability of models established with historical data. Overall, the survival model approach appears to be a promising method to predict both expected bus travel times and the uncertainty associated with these travel times.