Monash University, Institute of Transport Studies: World Transit Research (WTR)
Not a member yet
11112 research outputs found
Sort by
Addressing Data Latency in GTFS (General Transit Feed Specification) Realtime to Improve Transit Signal Priority
No full textTransit signal priority (TSP) is a strategy that provides preferential treatment at signalized intersections. TSP reallocates green time to reduce the delay of transit vehicles at traffic signals. To be effective, a transit vehicle (bus) must communicate its location to the traffic signal to make the reallocation of time beneficial. In GTFS (General Transit Feed Specification) Realtime, latency poses a significant challenge for the implementation of GTFS-based TSP. Using data from four transit agencies, this research identifies issues with current GTFS Realtime feeds and proposes a solution using machine learning algorithms to address latency compensation. Experimental results demonstrate that the performance of two machine learning models surpasses the baseline approach, which relies on hourly means for bus speeds and dwell times. This paper tackles multiple issues related to existing GTFS data, enhancing the practicality and feasibility of GTFS-based adaptive TSP. In contrast to conventional approaches focusing on estimation of bus arrival time, this paper emphasizes estimation of bus location and presents an effective method to compensate for latency and improve estimation of bus location and dwell time
Rethinking Transit Safety: Understanding and Addressing Gender-Based Harassment and Enhancing Safety on San Francisco’s Muni Transit System
No full textThe goal of this research is to support prioritizing safety on the Muni system from a gender equity lens. This work is specifically aimed to inform the Safety Equity Initiative of the San Francisco Municipal Transportation Agency (SFMTA). Using a mixed-method approach, we surveyed Muni transit riders (n = 1,613) to explore their travel behaviors, experiences with gender-based harassment, and perceptions of safety while riding Muni. We find that the pervasiveness of gender-based harassment on the Muni public transit system is significant, with 67% of our sample reporting that they have experienced harassment in the last six months. Perceptions of safety are also quite low, with 68% always or often feeling safe while riding Muni during the daytime and only 32% feeling safe at nighttime. We found that certain populations are disproportionately victimized while riding transit with statistically significant differences across both perception of safety and experiences of harassment between women versus men, gender minorities versus cis gender people, transit dependent riders versus those who have access to a private vehicle, and white versus non-white riders. Based on these findings we provide transformative recommendations to address the high rates of harassment among certain groups of Muni riders. The recommendations are organized into three categories: service changes, infrastructure improvements, and campaigns and advocacy. This work adds to the existing knowledge about gender-based harassment in the transit environment while also specifically informing the Safety Equity Initiative of the SFMTA
Dynamic adjustment strategy of electric bus operations: A spatial branch-and-bound method with acceleration techniques
No full textElectric bus systems frequently encounter operational instability, resulting in delays, bunching and disturbed charging schemes. Advanced technologies like sensoring and wireless connectivity strongly support dynamic adjustments to improve the stability of electric bus systems, fostering flexibility responsiveness to changing operating conditions. This article explores the dynamic adjustment problem of electric bus operations to jointly generate bus charging schemes and timetable adjustments in a real-time decision-making process. We propose a mixed-integer nonlinear programming model for each decision stage, explicitly considering factors such as vehicle overtaking, passenger load, capacity limitations, and charging behaviors. To efficiently solve this problem, we design a spatial branch-and-bound method with multiple acceleration techniques of logical inference for bound contraction, bilinear-specific branching, and parallel computation. Indeed, the original mixed-integer nonlinear programming problem can be split into a series of mixed-integer quadratic programming problems with reduced domains. The acceleration techniques proposed could be easily customized to address other mixed-integer nonlinear programs in such branch-and-bound-based schemes. Computational experiments validate the effectiveness of the proposed adjustment strategy yielding feasible solutions that enhance headway regularity, energy savings and service quality. Additionally, our solution method efficiently tackles real-world scenarios of significant complexity, with up to 12 vehicles running 8 loops on a route comprising 54 stops, with an average of 5.11-seconds computational time, outperforming both the common commercial solver and standard spatial branch-and-bound approaches in computational speed
Bridging the gap: A social equity analysis of intra-city transit access to inter-city rail in Canada
No full textTo support the decarbonization of inter-city transport and reduce the risk of marginalized populations being excluded from long-distance travel, intra-city transit systems should provide equitable access to inter-city rail network nodes. This study evaluates transit-based accessibility to inter-city rail stations in seven Canadian cities along the Québec City–Windsor corridor through an equity lens. Specifically, we evaluate accessibility inequalities using Gini coefficients and three accessibility ratios based on income, race, and age. While Gini coefficients reveal no clear link between city size and inequality, the three socioeconomic status-based accessibility ratios indicate that marginalized population groups face significant inequalities in the two largest cities: Toronto and Montréal. In these two cities, low-income individuals, visible minorities, and older citizens experience inequitable accessibility, with longer transit travel times to inter-city rail stations. These findings highlight the uneven distribution of transit access to inter-city rail services in larger cities, potentially deepening social exclusion for vulnerable population groups and hindering the transition to sustainable inter-city travel. By examining intra-city transit access to inter-city rail through a social equity lens, this study offers valuable insights into the social equity of intermodal connectivity in Canada and provides a framework for similar assessments in other geographic contexts
Opening the tracks: The impact of rail liberalization on ridership growth in the Czech Republic
No full textThe Czech Republic has carried out dedicated rail liberalization. Two major lines, Prague–Ostrava and Prague–Brno, were opened for on-track competition and entered by private operators. These entries led to fare declines, frequency increases, and quality improvements. Ridership increased significantly, but it has not been clear what part of the ridership growth has been caused by competition and what part by other factors. We used data about Czech rail ridership that consists of 12 long-distance connections from Prague to regional centres. Four were routes with competition and eight were routes operated solely by the incumbent. This design enables differentiation of what part of the ridership increases was caused by competition and what part by other factors. The main result is that after controlling for economic growth and travel-time improvements, ridership on lines with competition grew by 5 % p.a. more than it did on lines without competition
Optimization of electric bus vehicle scheduling and charging strategies under Time-of-Use electricity price
No full textWith the growing awareness of environmental protection and energy conservation, more and more cities choose to utilize electric buses (EBs) in their public transit systems. Due to the limitations of battery technology, many fully charged EBs are not enough to complete their daily tasks, which must be charged twice or more times per day. Besides, many cities encourage the off-peak electricity power consumption, and the charging cost of EBs during peak hours is often extremely high. From an economic viewpoint, it is thus one practical and urgent problem on how to decide the fleet size of EBs and organize their charging schedules for the bus companies. To solve this problem, this manuscript builds a mixed-integer programming (MIP) model via taking the scheduling and charging constraints of EBs into consideration. Also, one dynamic label setting-based branch and price (DLS-BP) algorithm is proposed accordingly, whose efficiency is further verified and compared with two heuristic algorithms via some numerical experiments
Is fare-free public transport effective in improving air quality? Evidence from Fuzhou, China
No full textFare-free public transport (FFPT) programs are gaining increasing popularity worldwide as a policy tool to mitigate the negative externalities associated with automobile usage. However, evidence regarding their effectiveness in reducing air pollution, a major automobile externality, remains limited. In this study, we empirically examine the effect of FFPT programs on air quality in Fuzhou, a provincial capital city in China, based on a quasi-experimental design. Using difference-in-differences models, we find that Fuzhou’s FFPT program reduces fine particulate (PM2.5) concentrations by 0.332 µg/m3 (2.1 %) in the short run. Furthermore, the program leads to an increase of 129,486 rides (49.8%) in daily subway ridership. Back-of-the-envelope calculations indicate that the health benefits brought by the FFPT program through air quality improvements, including reduced mortality and healthcare expenditures, amount to about 3.09 billion Chinese yuan (or 478.92 million US dollars) annually, which is six times the loss of fare revenue. These findings highlight the potential of fare-free public transport as a sustainable urban transport policy in urban China and contribute to a better understanding of its cost-effectiveness
Predicting travel demand of a bike sharing system using graph convolutional neural networks
No full textPublic transportation systems play a crucial role in daily commutes, business operations, and leisure activities, emphasizing the need for effective management to meet public demands. One approach to achieve this goal is by predicting demand at the station level. Bike-sharing systems, as a form of transit service, contribute to the reduction of air and noise pollution, as well as traffic congestion. This study focuses on predicting travel demand within a bike-sharing system. A novel hybrid deep learning model called the gate graph convolutional neural network is introduced. This model enables prediction of the travel demand at station level. By integrating trajectory data, weather data, access data, and leveraging gate graph convolution networks, the accuracy of travel demand forecasting is significantly improved. The Chicago city bike-sharing system is chosen as the case study. In this investigation, the proposed model is compared to the base models used in previous literature to evaluate their performance, demonstrating that the main model exhibits better performance than the base models. By utilizing this framework, transportation planners can make informed decisions on resource allocation and rebalancing management
Assessing the Effect of Negative Externalities of Urban Transport on Travel Satisfaction for Work Trips
No full textResearch on the importance and performance of work trip attributes for users of different transportation modes—such as public transit, two-wheelers, and four-wheelers—and their impact on travel satisfaction remains limited. This gap is particularly evident when considering the influence of negative externalities of urban transport on travel satisfaction in developing countries. This study seeks to fill this gap by performing an importance-performance analysis of work trip attributes and evaluating the impact of these attributes on travel satisfaction through the use of an ordered hybrid choice model. Findings underscore the significant impact of negative externalities on work trip-related travel satisfaction across all mode users. Additionally, a higher travel cost is negatively associated while a lower travel time is positively associated with travel satisfaction. Higher-income individuals and four-wheeler users exhibit relatively high travel satisfaction compared with two-wheeler and public transit users. Furthermore, the analysis of work trip attributes reveals that the most crucial factors are the access time for public transit users, travel time reliability and travel time delay for four-wheeler users, and the risk of road crashes for two-wheeler users. The study\u27s findings offer insights for policymakers and planners when prioritizing strategies within the urban transport sector in the Mumbai Metropolitan Region (MMR). On a global research scale, the study advocates for the inclusion of negative externalities in travel satisfaction research, emphasizing their pivotal role in shaping the travel experience
Coordinating ride-pooling with public transit using Reward-Guided Conservative Q-Learning: An offline training and online fine-tuning reinforcement learning framework
No full textThis paper introduces a novel reinforcement learning (RL) framework, termed Reward-Guided Conservative Q-learning (RG-CQL), to enhance coordination between ride-pooling and public transit within a multimodal transportation network. We model each ride-pooling vehicle as an agent governed by a Markov Decision Process (MDP), which includes a state for each agent encompassing the vehicle’s location, the number of vacant seats, and all pertinent information regarding the passengers on board. We propose an offline training and online fine-tuning RL framework to learn the optimal operational decisions of the multimodal transportation systems, including rider-vehicle matching, selection of drop-off locations for passengers, and vehicle routing decisions, with improved data efficiency. During the offline training phase, we develop a Conservative Double Deep Q Network (CDDQN) as the action executor and a supervised learning-based reward estimator, termed the Guider Network, to extract valuable insights into action-reward relationships from data batches. In the online fine-tuning phase, the Guider Network serves as an exploration guide, aiding CDDQN in effectively and conservatively exploring unknown state–action pairs to bridge the gap between the conservative offline training and optimistic online fine-tuning. The efficacy of our algorithm is demonstrated through a realistic case study using real-world data from Manhattan. We show that integrating ride-pooling with public transit outperforms two benchmark cases—solo rides coordinated with transit and ride-pooling without transit coordination—by 17% and 22% in the achieved system rewards, respectively. Furthermore, our innovative offline training and online fine-tuning framework offers a remarkable 81.3% improvement in data efficiency compared to traditional online RL methods with adequate exploration budgets, with a 4.3% increase in total rewards and a 5.6% reduction in overestimation errors. Experimental results further demonstrate that RG-CQL effectively addresses the challenges of transitioning from offline to online RL in large-scale ride-pooling systems integrated with transit