Transit Service Reliability Research Articles

The impacts of the modifiable areal unit problem (MAUP) on social equity analysis of public transit reliability

This study aims to examine the degree to which the results of social equity analysis on public transit reliability are sensitive to the choice of spatial unit of analysis (i.e., modifiable areal unit problem (MAUP)). Using the city of Winnipeg as an example, we investigate the social equity of bus on-time performance (OTP), and pass-up distribution at multiple levels (e.g., stop, route, neighborhood) and compare the results. Neighborhoods are classified as minority vs. non-minority population dominant neighborhoods and transit routes and stops are classified into minority vs. non-minority population serving ones according to the socioeconomics of residents. We conduct an equity assessment by calculating 1) the number of pass-ups, 2) the number of bus deviations (e.g., early arrivals, delays) from the predetermined schedule, and 3) the average deviations time (in seconds) that occurred on minority vs. non-minority serving routes/stops as well as in minority vs. non-minority dominant neighborhoods. We also consider different day types (e.g., weekdays, weekends) and disability status (e.g., wheelchair, regular passengers) in the equity assessment. While the route-level results depict that transit service reliability is equitable, results of the neighborhood and stop-level analysis reveal inequities in the distribution of the OTP and pass-ups in the city. Our findings demonstrate that different levels of spatial aggregation can significantly change the results of social equity analysis of public transit reliability, thereby leading to incomplete conclusions that inadequately capture the inequity due to the existence of MAUP. The results of this paper provide insights to transit authorities, planners, and policymakers for diagnosing the social equity landscape of transit reliability in a more robust manner.

Bus bunching simulation based on the lattice Boltzmann model of traffic flow

Bus bunching is a typical phenomenon observed in daily public transit operations and that significantly reduces the reliability and attractiveness of transit service. Although various control strategies, such as vehicle holding, speed guidance, and signal priority, have been proposed to eliminate or alleviate bus bunching, the impact of road traffic flow on bus bunching has not been well studied. To bridge this gap, this work proposes a new mesoscopic simulation model by incorporating the lattice Boltzmann model to simulate bus bunching considering the impacts of road traffic flow. Simulation results show that by utilizing the new simulation model, different phase diagrams can be observed. Under different conditions of traffic flow-density evolution, i.e., no oscillation, single oscillation, global oscillation, and irregular oscillation, the resulting bus trajectories are significantly different. In addition, an investigation of a real-world bus line in Beijing with global positioning system (GPS)-based vehicle location data shows that some results of the model are qualitatively consistent with the actual characteristics of the bus route. It shows that the proposed new simulation model has great potential for enhancing our understanding of the impacts of road traffic flow on bus bunching.

Vehicle Routing and Scheduling of Flex-Route Transit under a Dynamic Operating Environment

To improve the reliability, responsiveness, and productivity of the flex-route transit service, this paper investigates the vehicle scheduling and routing problem under a dynamic operating environment. First, we discuss the new operating polices after the introduction of intelligent transportation systems (ITSs), including automatic vehicle location (AVL) system, mobile data terminal (MDT), and computer-aided dispatch (CAD) system. Second, a mixed integer programming (MIP) formulation is employed to solve the offline routing problem. Third, an online scheduling scheme is presented to tackle different dynamic events, such as dynamic requests, travel time fluctuations, cancellations of requests, and customer no-shows. Finally, simulation experiments based on a real-life flex-route transit service are conducted to assess the influence of different dynamic events. The results demonstrate that the proposed scheduling scheme is reliable for coping with various dynamic events, and our findings can be used to guide the policy making of flex-route transit services.

Reliability prediction of further transit service based on support vector machine

The requirement for transit reliability grows with the increase of pace of life since unstable bus arrivals can raise the anxiety of waiting passengers. This paper proposes a reliability assessment method to evaluate the reliability of each bus stop on the route and the reliability of bus routes. In reliability prediction, the prediction target is locked by rolling horizon to reduce the interference of other information. In addition, a prediction method of the reliability of further transit service using the accurate online support vector machine is proposed. This prediction can provide more accurate and stable data for the arrival of buses and reduce unnecessary waiting of passengers. Finally, the reliability prediction method proposed is tested with the real data of a bus route in Dalian, China. The results show that the accurate online support vector machine with reasonable parameters can predict the reliability of transit service accurately.

Travel-Time Variability Analysis of Bus Rapid Transit System Using GPS Data

AbstractIn prior studies, the transit service reliability has been looked at from the passenger’s perspective in terms of time spent by the passengers in waiting for their bus to arrive at a stop. ...

Public Transit Service Reliability Assessment using Two-Fluid Model

This study introduces a traffic flow theory-based reliability indicator to evaluate the inter-city public transit service. The two-fluid theory parameter n that measures the road networks’ resilience to changing traffic is used as a new reliability indicator of public bus service. We compared the performance of the new indicator with the three existing reliability indicators, including on-time performance, with a total of 52 different GO Bus routes to ascertain the usability of the new indicator. The GO Bus service is an inter-city bus service operated by Metrolinx in the Greater Toronto and Hamilton areas in Ontario, Canada. We used 1 month of GO Bus GPS data collected in July 2017 to estimate various public transit reliability indictors. We also investigated the relationship between reliability indicators and selected roadway network characteristics, such as route length, freeway ratio, intersection density along a route, and so forth. The study applied a series of statistical analyses, including principal component analysis, correlation analysis, and z-test using Fisher’s z transformation. The results showed that the proposed two-fluid indicator n can be used as a supplementary reliability indicator for assessing road networks’ resilience in regards to changing traffic flow conditions. The new indicator provided additional insights for inter-city bus service operators for initiating communication with roadway governing agencies for the purpose of improving road networks to further improve public transit service reliability.

A Prediction Model for Bus Arrival Time at Bus Stop Considering Signal Control and Surrounding Traffic Flow

The arrival time of bus at the stop is critical data in design of bus operational strategies. Especially for real-time control strategies (i.e., signal priority system), bus arrival time is usually predicted to assess the next bus operation condition in the future (i.e., bus bunching and reliability of transit service) and then can be used as a decision basis of current control actions. The signalized intersection and surrounding traffic are the key factors in bus travel time prediction, but most previous approaches focus on the impact of signal control on bus delay only. This paper proposes a prediction model for arrival time at bus stop under the influence of both upstream signalized intersection and surrounding traffic flow. Considering the affected range of signalized intersection and the dynamic variation of bus speed, bus running processes are evaluated separately (including processes from a given detection point to stop line, through intersection, and from intersection to bus stop). In the proposed models, bus speed is deduced according to the change of traffic density at different locations to reflect the micro-impact of surrounding traffic flow on bus operation. The observed bus travel time is collected from actual investigations at two bus stops incorporating signalized intersection in Jinan City and compared with the predicted travel time in the proposed model. The results show that the proposed model has a low mean relative error. In addition, through the analysis of the maximum relative error, it also can be seen that vehicle queuing with random arrival of vehicles at stop line makes a gap between the prediction and the actual situation, which will be the focus of further research.

Analysis of Bus Travel Time Variability using Automatic Vehicle Location Data

Transit service reliability is a measure of the quality of service offered by the public transit systems. The nature and pattern of the travel time variability can be described by the travel time distribution and it is the prerequisite in the reliability analysis. Many research works have been carried out to understand the travel time distribution, but there are very few studies on the heterogeneous traffic in developing countries like India. In this study, an attempt has been made to analyze the travel time distribution at different spatial and temporal aggregation scales using Automatic Vehicle Location (AVL) data of public bus route in Mysore city. The results suggest that, Generalised extreme value (GEV) distribution is the better fit for travel times at both route and segment levels and the performance of the Generalised extreme value (GEV) distribution is comparatively better in both temporal and spatial aggregation compared to other distributions. This distribution can be utilized by the transit operators in the operation and performance evaluation of transit systems.

The benefits of introducing meeting points into flex-route transit services

In this paper, to improve the reliability of flex-route transit service under uncertain travel demand, a meeting point strategy is proposed in which passengers can be picked up and dropped off either at their reserved stops or at meeting points that are within their acceptable walking distance. A two-stage optimization problem is introduced into the design of flex-route transit operations. Mixed integer programming (MIP) is employed to formulate the problem with a twofold objective: to serve as many requests as possible and to minimize the total trip time of the accepted passengers. A memetic algorithm is adopted to solve the problem in a reasonable amount of time. Simulation experiments based on a real-life flex-route transit service are conducted to assess the benefits of introducing meeting points. The results demonstrate that meeting points can significantly reduce the rejection rate by up to 24% at the cost of walking an acceptable distance. We also find that reducing the passenger fare to compensate for the inconvenience of walking to meeting points constitutes a win–win strategy that benefits both passengers and transit operators.

Understanding the factors that influence the probability and time to streetcar bunching incidents

Bunching is a well-known operational problem for transit agencies and it has negative impacts on service quality and users’ perception. While there has been a substantial amount of literature about understanding the factors associated with bus bunching and strategies used to mitigate the effects of this problem, there has been little research on streetcar bunching. Although bus and streetcar systems share many similarities, one major difference between the two is that streetcars cannot overtake each other. This makes bunching in streetcar networks more critical to the reliability of the system and an important topic that requires more in-depth understanding. This research aims at understanding the factors that are associated with the likelihood of streetcar bunching and to investigate in greater detail the external and internal factors that relate to the time to the initial bunching incident from terminal. To achieve the first goal, the study uses a binary logistic regression model, while it uses an accelerated failure time model to address the second goal. The study utilizes automatic vehicle location system data acquired from the Toronto Transit Commission, the transit provider for the City of Toronto. The models’ results show that headway deviations at terminals are related to both an increase in the probability of bunching and an acceleration of the time to bunching. The discrepancy in vehicle types between two successive streetcars also has the same relationship as headway deviations at terminals. This study offers a better understanding of the factors that are associated with streetcar service bunching, which is an important component of transit service reliability.

Peer-to-Peer Priority Signal Control Strategy in a Connected Vehicle Environment

This paper presents a methodology that enhances the priority signal control model in the multi-modal intelligent traffic signal system (MMITSS). To overcome the range limit of vehicle to infrastructure (V2I) and the intersection geometry message (MAP) distance limits, peer-to-peer intersection communications are utilized to send priority requests from adjacent intersections. Through integrated communication, the peer priority control strategy can create a signal plan for prioritized vehicles that considers longer term (headway) arrival times. Transit vehicles are considered in this study. The longer-term signal plan provides a flexible signal schedule that allows local phase actuation. The peer priority strategy is effective in reducing the number of stops and delay for priority eligible vehicles, while minimizing the negative impact on regular vehicles. To validate the strategy, a simulation experiment was designed to compare fully actuated control, coordination, and MMITSS priority control using two different VISSIM simulation networks (Arizona and Utah). The result shows that the peer-to-peer long term planning strategy can improve transit service reliability while limiting the adverse impact on other traffic.

Estimation of Platform Waiting Time Distribution Considering Service Reliability Based on Smart Card Data and Performance Reports

The estimation of platform waiting time has so far received little attention. This research aimed to estimate platform waiting time distributions on the London Underground, considering travel time variability by using smart card data that were supplemented by performance reports. The on-train and ticket gate to platform walking times were assumed to be normally distributed and were matched with the trip time recorded by the smart cards to estimate the platform waiting time distribution. The stochastic frontier model was used, and its parameters were estimated by the maximum likelihood method. The cost frontier function was used to represent the relation between the travel time recorded in the smart card data as an output and the on-train time and walking time between the ticket gate and the platform as inputs. All estimated parameters were statistically significant, as shown by p-values. Comparing the travel time values estimated by the proposed model with the times recorded recorded in the smart card data shows a goodness-of-fit coefficient of determination of more than 95%. The estimation proved to have quick convergence and was computationally efficient. The results could facilitate improvements in transit service reliability analysis and passenger flow assignment. Matching the obtained distributions with the observed smart card data will help with estimating route choice behavior that can validate current transit assignment models.

A Dynamic Short-Turning Bus Control for Uncertain Demand

This paper formulates a dynamic approach for real-time bus control in uncertain demand. This dynamic approach aims to save the total cost for passengers and operators, while improving transit service reliability. An unfixed rolling horizon was implemented to choose the best dynamic approach. Real-time control predicts two discrete variables (arrival time and bus position) and determines the space-time point of buses. Furthermore, controlled actions include stop skipping and bus holding. The holding time starts when a bus serves a station and depends on previous intervals of passenger boarding and alighting at the station. The stop skipping action allows a bus to skip not only stations with a short-turning exception, but also stations with low demand for boarding that have been alighted in the short-turning segment. Stop skipping and bus holding actions for short-turning service both decrease the travel time of served passengers and the running time of buses, thus improving transit service reliability. A genetic algorithm was applied to solve the problem and the validity of the proposed dynamic approach was tested with four different scenarios. The result of these tests shows that a dynamic short-term bus control can significantly reduce total cost and improve transit service reliability.

A reliability-based transit trip planning model under transit network uncertainty

Transit, although an important public transportation mode, is not thoroughly utilized in the United States. To encourage the public to take transit, agencies have developed systems and tools that assist travelers in accessing and using information. Transit data modeling and trip planner system architecture developments have helped advance these systems, and the recent emergence of transit trip planning algorithms promises further enhancement. Conventional transit trip planning algorithms are usually developed based on graph theory. In order to utilize these algorithms, certain assumptions must be made to support these algorithms (e.g. buses always run on time). However, these assumptions may not be realistic. To overcome these limitations, our study develops an innovative transit trip planning model using chance constrained programming. Unlike previous studies, which only minimized passenger-experienced travel time, our study also considers transit service reliability. Additionally, in-vehicle travel time, transfer time, and walking time are all included as elements of passenger-experienced travel time. Our transit trip planning model avoids the assumptions of previous studies by incorporating transit service reliability and is capable of finding reliable transit paths with minimized passenger-experienced travel time. The algorithm can also suggest a buffer time before departure to ensure on-time arrivals at a given confidence level. General Transit Feed Specification data, collected around Tucson, Arizona, was used to model the transit network using a “node-link” scheme and estimate link-level travel time and travel time reliability. Three groups of experiments were developed to test the performance of the proposed model. The experiment results suggested that the optimal anticipated travel time increased with increasing on-time arrival confidence level and walking was preferred over direct bus transfers that involved out of direction travel. The proposed model can also include additional travel modes and can easily be extended to include intercity trip planning.

The effect of slow zones on ridership: An analysis of the Chicago Transit Authority “El” Blue Line

Transit agencies frequently upgrade rail tracks to bring the system to a state of good repair (SGR) and to improve the speed and reliability of urban rail transit service. For safety during construction, agencies establish slow zones in which trains must reduce speed. Slow zones create delays and schedule disruptions that result in customer dissatisfaction and discontinued use of transit, either temporarily or permanently. While transit agencies are understandably concerned about the possible negative effects of slow zones, empirical research has not specifically examined the relationship between slow zones and ridership. This paper partially fills that gap. Using data collected from the Chicago Transit Authority (CTA) Customer Experience Survey, CTA Slow Zone Maps, and, the Automatic Fare Collection System (AFC), it examines whether recurring service delays due to slow zones affect transit rider behavior and if the transit loyalty programs, such as smart card systems, increase or decrease rider defections. Findings suggest that slow zones increase headway deviation which reduces ridership. Smart card customers are more sensitive to slow zones as they are more likely to stop using transit as a result of delay. The findings of this paper have two major policy implications for transit agencies: (1) loyalty card users, often the most reliable source of revenue, are most at risk for defection during construction and (2) it is critical to minimize construction disruptions and delays in the long run by maintaining state of good repair. The results of this paper can likely be used as the basis for supporting immediate funding requests to bring the system to an acceptable state of good repair as well as stimulating ideas about funding reform for transit.

Genetic Algorithm and Regression-Based Model for Analyzing Fare Payment Structure and Transit Dwell Time

The time that buses spend at stops, also called dwell time (DT), has a direct effect on transit service reliability and operational efficiency. A practical, coherent, and quantitative DT modeling approach is needed to identify the factors that contribute most to DT. Commonly used methods for studying DT to date involve manually collected field data or the use of automatic sensors to gather information on factors influencing DT. These approaches have often suffered from limited sample sizes or the inability to provide information on nonelectronic fare payment methods (e.g., cash payment and prepaid passes), which can contribute significantly to DT. To address these gaps, this study developed a genetic algorithm and regression-based modeling approach first to estimate transit fare transactions that do not have electronic records and then to quantify the effect of a number of factors on DT. Integrating information from multiple data sources, the combined approach of optimization and regression analysis offers a data-driven evaluation of existing fare payment structures and their individual effects on DT. With the 35M bus rapid transit line operated by the Utah Transit Authority as a case study, the method demonstrates the robustness and strong prediction power in DT modeling. Results quantify the magnitude of advantages of offboard over onboard fare collections and offer some insights into the operational effects of station placement, design, and the built environment. The modeling approach is transferable to any transit route or system that is equipped with automatic passenger counters. The fare payment analysis can assist transit agencies with service optimization and performance assessments.

Bus Transit Service Reliability and Improvement Strategies: Integrating the Perspectives of Passengers and Transit Agencies in North America

Transit agencies are consistently trying to improve service reliability and attract new passengers by employing various strategies. Previous literature reviews have focused on either passengers' or transit agencies' perspectives on service reliability. However, none of the earlier reviews have simultaneously addressed these differing perspectives on service reliability in an integrated manner. In response to this gap in the literature, this paper first reviews previous work on passengers' perspectives of transit service reliability and their response to service adjustments made by different agencies. Second, it analyzes transit agencies' plans and reports regarding their reliability goals and used strategies in order to improve service reliability, while looking at the impacts of these strategies on service. Reviewing these two parts together provides a needed contribution to the literature from a practical viewpoint since it allows for the identification of gaps in the public transit planning and operations field in the area of reliability and provides transit planners and decision makers with effective and valuable policy-relevant information.

Models of bus boarding and alighting dynamics

Understanding the dynamics of boarding/alighting activities and its impact on bus dwell times is crucial to improving bus service levels. However, research is limited as conventional data collection methods are both time and labour intensive. In this paper, we present the first use of smart card data to study passenger boarding/alighting behaviour and its impact on bus dwell time. Given the nature of these data, we focus on passenger activity time and do not account for the time necessary to open and close doors. We study single decker, double decker and articulated buses and identify the specific effects of floor/entrance type, number of activities and occupancy on both boarding and alighting dynamics. A linear relationship between average boarding and alighting times and their respective standard deviations is also found, whereas the variability of boarding and alighting time decreases with the number of passengers boarding and alighting. After observing the cumulative boarding/alighting processes under different occupancy conditions, we propose a new model to estimate passenger activity time, by introducing critical occupancy – a parameter incorporating the friction between boarding/alighting and on-board passengers. We conduct regression analyses with the proposed and another popular model for simultaneous boarding/alighting processes, finding that the critical occupancy plays a significant role in determining the regime of boarding and alighting processes and the overall activity time. Our results provide potential implications for practice and policy, such as identifying optimal vehicle type for a particular route and modelling transit service reliability.

Measuring Route-Level Passenger Perceived Transit Service Reliability with an Agent-Based Simulation Approach

This research presents an agent-based method to simulate the day-today bus service process and effectively capture the perception and cognition of passengers under various operational and information conditions. Two indicators, the perceived waiting time and the reliability index, were developed to measure transit service reliability on the basis of the simulation results. Case studies revealed the promising properties of the proposed method for use in measuring transit service reliability from the perspective of passengers. The proposed method was also shown to outperform most traditional models in reflecting passengers’ perception as well as their responses to transit services under information conditions.

Dynamic Dispatch with Advanced Train Control System Data

Transit service reliability, measured by such performance measures as schedule deviation and headway distribution, is a priority for transit agencies. The successful implementation of strategies to keep transit fleets as close to schedule as possible aims to improve customer service and to operate a more efficient system. Common strategies to recover time lost in day-to-day operations include switchbacks, fallbacks, route changes, and reconfigurations. The San Francisco Municipal Transportation Agency, operator of intracity bus and light rail transit in San Francisco, California, uses all of these strategies to remedy unacceptably large headway deviances. Analysis was conducted to determine the potential effects on service reliability of an application of real-time route changes of light rail vehicles at the Embarcadero turnaround. Historical advanced train control system data are applied to a calculator that reassigns the trains on the basis of average headway lengths. The calculator simulates a scenario if such an automatic reassignment system is in place and reports the new schedule deviation and headway distribution. Weekday data for 1 month were applied to the calculator and analyzed to determine whether the reassignment methodology could save significant time in the average headways. The calculator was meant as the first step in the development of a more thorough operational algorithm that would reassign vehicles on the basis of multiple independent variables, to establish better on-time reliability and performance. In preliminary results, the automatic reassignment system showed time savings. Also discussed are current physical and operational constraints to implementation and next steps in model development.