Headway Adherence Research Articles

Deep Reinforcement Learning Two-Way Transit Signal Priority Algorithm for Optimizing Headway Adherence and Speed

Transit Signal Priority (TSP) is a broadly used traffic signal control strategy designed for reducing transit delays at signalized intersections. Although recent TSP systems began to consider more objectives, TSPs that addressed transit reliability issues commonly focused on improving schedule adherence and were only able to reduce schedule delays by expediting buses. Buses running ahead of the schedule were not considered. This paper proposed a dual-objective two-way TSP algorithm (D2 TSP) using Deep Reinforcement Learning (DRL). D2 TSP concurrently optimizes transit delays and reliability (i.e., headway adherence) by expediting late buses or delaying early buses. Further, the DRL agents were enhanced with a coordination algorithm for an optimized solution balancing opposite directions. This D2 TSP reacts adaptively and efficiently to real-time bus performance using data provided by readily available technology (loop detector) at low communication frequencies. We trained and tested this algorithm in a stochastic microsimulation environment in Aimsun Next that modelled a transit route segment with reliability issues in the City of Toronto. The performance of D2 TSP was compared with four baseline scenarios, one without TSP, one with the current TSP algorithm used in the field in the City of Toronto, one conditional TSP with an arrival prediction model, and one using DRL agents with a First-Come-First-Served logic. D2 TSP demonstrated its advantages in providing an efficient and balanced solution in reducing headway variability and travel time for both directions.

Charging operations in battery electric bus systems considering fleet size variability along the service

This paper analyzes the new operations induced by the energy range limitation in battery electric bus (BEB) systems to ensure the service provision along the whole day. An operating model has been developed to estimate the number of resources to be deployed (extra vehicles and charging facilities) for two charging schemes: charging at bus garage facility and opportunity charging at on-street chargers. Firstly, the model calculates the entrance and exit time of buses in service assuming an unlimited battery capacity, caused by demand and headway variations in different time periods along day. Then, based on the previous results, the model estimates the moment when vehicles will run out of energy and proposes a charging schedule as well as new vehicle movements between chargers, bus garage and routes to maintain the targeted headway. Consequently, the model is able to calculate the number of resources needed and the total cost for a given route and battery size, allowing cost and performance comparison to diesel and hybrid powertrains. The estimated results in two bus routes in Barcelona indicate that the purchasing cost of BEB vehicles (670,000 euros/veh for articulated, 500,000Euros/veh for standard vehicles, both depreciated over a lifetime of 15 years), battery packs (231 Euro/kWh purchased, depreciated over a lifetime of 10 years), as well as the investment in chargers, are still high to clearly justify their cost-competitiveness against diesel or hybrid counterparts. Nevertheless, if the acquisition cost of both vehicles and battery pack are only reduced by 9%, the BEB technology may present cost savings for transit agencies. It only happens if the fleet size is as large as the corresponding to diesel and hybrid technologies. Opportunity charging is resulted to be more cost- efficient than charging at bus garage for small battery packs, routes with acceptable time headway adherence and, specially, when fast chargers are located very close to terminal stops. Nevertheless, when the real bus route design and operation do not fulfill these standards, charging scheme at bus garage is preferable.

Timetable and Headway Adherence Assessment using the Gini Coefficient and the Lorenz Curve

Adherence to the schedule is of prime importance in public transport. This paper presents a specific application of the Gini coefficient, well known indicator in economics, for the headway adherence assessment. The paper shows that Lorenz curve, which is usually used to define mathematically the Gini coefficient, is a good indicator of the users' waiting time when it is based on the bus schedule. When it is computed on the basis of the ratio of observed headway to the schedule, it is a powerful visual tool that can be used by operators to detect the existence of irregularities on a bus line at a glance. An equation gives, in an idealistic case, the impact of any single traffic disturbance on the Gini coefficient, making this coefficient comprehensive. A detailed analysis is developed, based on the bus proportions according to the headway adherence level. These proportions are obtained from new indices coming from the derivative of the Lorenz curve. The values of these indexes alert the operator of any adherence disturbance. The examination of the Lorenz curve takes more time, but is worthwhile, giving the types of the irregularities The application of these indicators is carried on real-time data from the New Delhi bus network.

The Effect of Transit Signal Priority on Bus Rapid Transit Headway Adherence

We report the results of an experiment to evaluate the impact of transit signal priority (TSP) on headway adherence for a bus rapid transit (BRT) system in Provo / Orem, Utah. The bus requests TSP based on its unpublished schedule, but users perceive only a headway. Quantile regression models estimated on raw timepoint data from the BRT system reveal that TSP significantly improves headway adherence, after controlling for peak times, direction, and cumulative trip dwell time. We also find that requiring the bus to be 2 minutes late before requesting TSP improves headway adherence more than allowing all buses to request TSP.

Using Microsimulation to Estimate Effects of Boarding Conditions on Bus Dwell Time and Schedule Adherence for Passengers with Mobility Limitations

AbstractBoth bus dwell time and bus headway adherence are critical to the serviceability of the bus transit system. It is acknowledged that passengers with physical disabilities may greatly impact ...

Tailored Wakeby-type distribution for random bus headway adherence ratio

This paper addresses an interesting and practical bus headway adherence issue for a given public bus route with a number of bus stops. It first defines the random headway adherence ratio (HAR) at a particular bus stop of a specific bus route as the ratio of difference between actual bus headway and scheduled headway with respect to the scheduled headway. This study proceeds to customize a four-step procedure to estimate a probability distribution that can describe the random HAR at each bus stop of the bus route by using the automatic vehicle location (AVL) data. Our real case studies with 44,025 HAR data show that the 19 existing probability distributions including Lognormal, Gamma, Beta and Wakeby are unable to well fit these HAR data. This study thus proposes a tailored Wakeby-type distribution with five parameters. After deriving two fundamental propositions for the tailored Wakeby-type distribution, a tangible L-moment based method to estimate those parameters involved the tailored Wakeby distribution is presented. The tailored Wakeby-type distributions can meet our expectation via our real case studies. Finally, applications of the tailored Wakeby-type distribution derived for the random HAR are conducted.

A connected-vehicle-based dynamic control model for managing the bus bunching problem with capacity constraints

ABSTRACTThis paper describes a connected-vehicle-based system architecture which can provide more precise and comprehensive information on bus movements and passenger status. Then a dynamic control method is proposed using connected vehicle data. Traditionally, the bus bunching problem has been formulated into one of two types of optimization problem. The first uses total passenger time cost as the objective function and capacity, safe headway, and other factors as constraints. Due to the large number of scenarios considered, this type of framework is inefficient for real-time implementation. The other type uses headway adherence as the objective and applies a feedback control framework to minimize headway variations. Due to the simplicity in the formulation and solution algorithms, the headway-based models are more suitable for real-time transit operations. However, the headway-based feedback control framework proposed in the literature still assumes homogeneous conditions at all bus stations, and does not consider restricting passenger loads within the capacity constraints. In this paper, a dynamic control framework is proposed to improve not only headway adherence but also maintain the stability of passenger load within bus capacity in both homogenous and heterogeneous situations at bus stations. The study provides the stability conditions for optimal control with heterogeneous bus conditions and derives optimal control strategies to minimize passenger transit cost while maintaining vehicle loading within capacity constraints. The proposed model is validated with a numerical analysis and case study based on field data collected in Chengdu, China. The results show that the proposed model performs well on high-demand bus routes.

Bunching and Headway Adherence Approach to Public Transport with GPS

Emerging GPS devices enable new data collection opportunities for transit performance monitoring. In addition to the fact that GPS devices can replace labor-intensive survey techniques, they also collect traffic information throughout transit routes that the traditional fixed loop detectors cannot. If public transit vehicles are equipped with GPS devices, it is possible to monitor the performance of public transit services throughout the routes and alert the associated authorities at significantly lower costs about potential problems for corrective actions to be taken. Transantiago, a major public transit service provider in Santiago, Chile, has recently installed GPS sensors on all its vehicles which provides an excellent venue on which an innovative transit monitoring methodology can be modeled and applied. This paper first conducts current-status analysis on distributions of headways throughout a route in Santiago by processing extensive raw GPS data from transit vehicles. Then, unique transit headway adherence indices are developed with respect to the expected passenger waiting time and are presented in forms of two-dimensional tempo-spatial graphs. The analysis of real-life data collected from bus GPS probes in Santiago, Chile indicates that GPS devices in transit buses can effectively provide the proposed performance measures throughout the route on a daily basis.

Evaluation of the pre-detective signal priority for bus rapid transit: coordinating the primary and secondary intersections

Since the traditional transit priority strategy can only adjust signal timing in a limited range and is not suitable for all kinds of signal timing designs, it cannot provide enough priority for Bus Rapid Transit (BRT). In addition, traditional transit priority strategy has caused serious interferences with other traffic. This study proposes a pre-detective signal priority strategy for BRT with coordination between primary and secondary intersections. By pre-detecting, the time buses arrive at the primary intersection, the signal timing of both the primary and secondary intersections, along with the offsets, are adjusted simultaneously, based on the common length and the green ratio of each phase. In this method, the signal cycle constraints are clarified, and the bus control coordination between intersections has been taken into consideration. In this paper, one direction traffic is taken as a study example to testify the effectiveness of this method. The study uses the data collected from Changzhou, China, and a microscopic traffic simulation software PTV VISSIM with four simulation scenarios defined: no signal priority, traditional signal priority, pre-detective signal priority and pre-detective signal priority with coordination. This paper selects a set of indicators to evaluate the traffic operation for both public transit and private traffic. Results show that pre-detective signal priority with coordination is the most effective, with the total bus intersection delay decreases by 67.4% and the bus headway adherence declines by approximately 40% at all the primary and secondary stations of BRT line 1. Moreover, the negative effects that could happen with providing signal priority for BRT, such as increasing the delay and length of queue of private traffic at the intersections, are significantly reduced.

Transit Travel Time Reliability

Transit agencies traditionally have used vehicle movement to assess service quality. However, new technologies such as automatic fare collection systems, automatic passenger counters, and automatic vehicle location devices are giving agencies a wealth of data from which critical insights can be gained about customer experience. Trip-level travel time distributions are used to explore the potential for a new customer-focused measure of service quality. The proposed method builds on recent transit research and ties in to lessons learned over the past 20 years of highway reliability research. An example of delay incidents illustrates the disconnect between the vehicle-focused tools that currently are available to transit agencies for the evaluation of service quality (e.g., minutes of train delay, headway adherence) and what a customer cares about (i.e., travel time to the destination). This disconnect is addressed with the use of actual trip times to outline a new measure: percentage of customers with a travel time of less than x min. Historical travel time distributions create the opportunity to understand better the duration and magnitude of an incident. Possible applications of information about travel time reliability are identified; these include improved understanding of the causes of delay, incident management, and communication of service status to customers.

Calibration of a Public Transport Performance Measurement System for Switzerland

Performance evaluation is fundamental to transport analysis and operations. This paper describes a quality evaluation framework based on level of service (LOS) developed for urban public transport in Switzerland. The framework focuses on four key indicators: on-time performance, headway adherence, speed, and passenger loads. The framework allows users to assess service quality at different levels by measuring the indicators at the single element level and aggregating scores over larger sets of elements. Because the number of indicators is relatively small compared with other frameworks, application of the method is easier, but the method also provides less detailed results. An earlier paper presented a conceptual approach for the framework. This paper extends that work by describing the approach in more detail and focusing on calibration of the LOS framework and application of the framework to a real example in Zurich, Switzerland. The research is part of a larger effort designed to develop a consistent multimodal LOS evaluation system for Switzerland. This system would include nonmotorized traffic, public transport, and road-based transport.

Adaptive Optimal Control for Designing Automatic Train Regulation for Metro Line

Automatic train regulation (ATR) plays an important role in maintaining the service quality of a metro in relation to schedule and headway adherence. However, maintaining service quality in an optimal way with less capacity utilization of infrastructure, particularly in an environment where disturbances occur frequently, is a challenge. Intrinsically, designing ATR is a real time optimal control problem with high nonlinearity, heavy constraints, and stochastic characteristics. Dual heuristic programming (DHP) was successfully employed to design ATR; however, the influence of traffic modeling error on the ATR performance was observed. In this paper, the adaptive optimal control (AOC) method is developed to improve the DHP design in regard to modeling errors as well as optimality, and an ATR designed using AOC method was developed and evaluated. The evaluation shows that the AOC method is able to find a near-optimal solution more rapidly and accurately than the DHP method. Moreover, the ATR designed using the AOC method improves both schedule and headway adherence with less capacity utilization and is more robust against disturbances as well as traffic modeling errors resulting from passenger flow fluctuations.

Development of Measurement System for Public Transport Performance

Accurate measurement systems of performance are important for all businesses, including public transport. The importance of such systems will grow with the advent of liberalization in the European transportation sector (i.e., the opening of transit service provision for tendering), as service provision contracts will require agreed-on measures for performance and service. This paper describes research on developing a public transport level-of-service (LOS) standards framework for Switzerland. The standards framework is intended to reflect influences on capacity and operational quality in urban road-based transit and relies on four main measures: speeds, passenger loads, on-time performance, and headway adherence. This paper outlines the selection of specific measures for evaluating public transport LOS and the proposed framework. The framework provides a consistent method for evaluating public transport LOS on all levels of analysis, from single elements (e.g., bus stops) to networkwide analysis. The paper also describes how information from automated onboard data collection systems can be used for detailed service analysis. This calibration procedure is illustrated in an example from the Zurich public transport system. Challenges faced (especially with respect to the selection and filtering of relevant data) and future development of the approach are described. While the proposed LOS framework is being prepared for Switzerland, the experience and approach described should help others developing and improving performance management systems for public transport.

An Improved Model for Headway‐Based Bus Service Unreliability Prevention with Vehicle Load Capacity Constraint at Bus Stops

This paper presents an improved model for improving headway‐based bus route service reliability at bus stops using real‐time preventive operation control, taking into account dynamic interaction among random passenger demand, stochastic driving conditions of route segments, and vehicle load capacity constraint. In this model, the real‐time information of passenger demand and vehicle operation is involved to predict the imminent unacceptable headway deviation, in the case of which some in‐time preventive control strategies are deployed according to the given control rules. As a case study, a single fixed bus route with high‐frequency services was simulated and different scenarios of real‐time preventive operation control were performed. Headway adherence and average passenger wait time were used to measure bus service reliability. The results show that the improved model is closer to the real bus route service, and using real‐time information to predict potential service unreliability and trigger in‐time preventive control can reduce bus bunching and avoid big gap.

Evaluation of urban public transportation efficiency in Kutahya, Turkey

In this study, the public transport system in the city of Kutahya and its current transportation problems were firstly mentioned. Then, the transit quality of service in Kutahya was evaluated from various aspects such as transit availability, comfort and convenience. For this purpose, at first, transit availability was examined in terms of service frequency, hours of service and service coverage. Secondly, the comfort and convenience provided by the transit system in Kutahya was studied. For this reason, the overall crowding levels within the vehicles, headway adherence, and transit-automobile travel time were considered. In order to carry out most of these analyses, the procedures in Transit Cooperative Research Program (TCRP) Report 100 were followed. However, determining the level of service of public transport in terms of transit-automobile travel time by comparing the public transport travel times with automobile travel times constituted the essence of this study. In order to accomplish this aim, the feature of “driving directions” in Google Maps was used in terms of being a reference to automobile usage. So, the travel times on all the bus transit lines in Kutahya were compared with private car travel times determined by Google Maps for the same routes. By founding the difference between the public transport and car travel times, the public transport service level in terms of transit-automobile travel time was evaluated. Finally, several recommendations for increasing public transport usage were given.

Service Reliability Analysis of High Frequency Transit Using Stochastic Simulation

Aiming at decreasing average passenger waiting time at stops, time headway adherence is used to analyze and measure the route-level transit service reliability in terms of high frequency bus route services. Considering the influence of passenger demand fluctuation and the stochastic nature of running time on road segments on transit service reliability, a fixed high frequency bus route was set to be the example route. The stochastic simulation method was used to analogy the process of its bus services, time headway variation, and average passenger waiting time at stops under different passenger demand and running time fluctuation. The results show that bigger running time fluctuation and more uneven passenger arrival lead to higher headway variation and longer average passenger waiting time at stops; the accumulative headway variation tends to be higher which causes worse service reliability at stops downstream.

Florida Metropolitan Planning Organization Reports on Transit Capacity and Quality of Service: First-Year Evaluation

As an application of the transit quality-of-service framework presented in the first edition of the Transit Capacity and Quality of Service Manual (TCQSM), the Florida Department of Transportation required all metropolitan planning organizations (MPOs) in the state where fixed-route transit service operates to analyze those services on the basis of the six measures identified in the TCQSM: service frequency, hours of service, service coverage, passenger loading, reliability (on-time performance and headway adherence), and transit versus automobile travel time. A first-year evaluation compiles the analyses provided by the participating MPOs and provides an assessment of the aggregate performance of the transit systems. A larger part of the study focused on the examination of the actual process used by the MPOs and transit systems to evaluate their services. Changes recommended to improve and refine the process for future years are presented, based on the first-time experiences of the MPOs. This evaluation serves as a model for other areas in the country interested in applying the customer-oriented assessment of transit based on the TCQSM.

Dynamic Bus Arrival Time Prediction with Artificial Neural Networks

Transit operations are interrupted frequently by stochastic variations in traffic and ridership conditions that deteriorate schedule or headway adherence and thus lengthen passenger wait times. Providing passengers with accurate vehicle arrival information through advanced traveler information systems is vital to reducing wait time. Two artificial neural networks (ANNs), trained by link-based and stop-based data, are applied to predict transit arrival times. To improve prediction accuracy, both are integrated with an adaptive algorithm to adapt to the prediction error in real time. The bus arrival times predicted by the ANNs are assessed with the microscopic simulation model CORSIM, which has been calibrated and validated with real-world data collected from route number 39 of the New Jersey Transit Corporation. Results show that the enhanced ANNs outperform the ones without integration of the adaptive algorithm.