Demonstration and Evaluation of Intermittent Bus Lane Strategy
Nicolas Chiabaut, Universite de LyonShow Abstract
Anaïs Barcet, Universite de Lyon
As cities around the world grow rapidly, efficiency of public transport is heavily deteriorated by traffic jam, generating increase of travel times and making them unpredictable. Local authorities and operators are therefore compelled to implement strategies to improve the performances of public transport networks and especially bus lines to increase their attractiveness. Very innovative, the concept of intermittent bus lane may lead to important increase of bus system performance while limiting the reduction of the capacity devoted to general traffic. Frequently studied by analytical paper, practical demonstrations of intermittent bus lane strategy are not numerous. This paper tries to fill this gap by proposing the results and the lessons learned of a real-field demonstration in Lyon, France. After a detailed presentation of the case study, effects of intermittent bus lane strategy on traffic conditions are evaluated. Then, analysis of impacts on bus systems performance are carefully performed and shows that intermittent bus lane can be a promising strategy.
Transit Stop Consolidation
Maaza Mekuria, Hawaii Department of TransportationShow Abstract
A. Jeff Becker, Regional Transportation District (Denver)
Douglas Monroe, Regional Transit District Denver
Transit service is modeled using transit lines and the associated access network used by the customer. Proposed operational improvements maybe measured and compared using a set of travel characteristics and costs to see if beneficial strategies maybe obtained. A complete network analysis of several routes in the Regional Transit District (RTD) of Denver is analyzed using a constrained network optimization model by introducing stops that are kept in the result set because they are deemed necessary from policy and analyst points of view. The resulting optimization is compared with the free for all result. Using a flexible and comprehensive network model that takes into account of the variation in the demographic characteristics of the transit service area, we present results that inform stop consolidation decisions. Three routes are analyzed and results are presented using block and parcel level demographic data and actual street network. The stop consolidation results show that it is possible to serve the optimally using and transit network data.
Integrated Model for Easing Congestion in Metro Trains
Amit Jain, Indian RailwaysShow Abstract
Pradip Kumar Sarkar, School of Planning and Architecture
The peak hour demand for metro rail (urban rail) systems is usually met by supply side measures such as operating more number of trains per hour in the peak period. The peak period demand can be managed by adopting demand side measures such as time-based fare pricing strategies also called differential fares strategy. In this study, differential fares strategy has been adopted as a demand side measure and increase in trains (or cars) per hour has been adopted as a supply side measure to develop an integrated elasticity model. The model can be applied to determine different combinations of peak and non-peak hour fares and trains (or cars) per hour to ease congestion in peak period in metro trains. Level of Service (LOS) inside the trains has been defined to measure congestion. The values of elasticity of demand for fares for passengers have been estimated from customer survey using stated preference method. The value of elasticity for services has been worked out from regression analysis of the data for increase in ridership of Delhi metro, India with increase in availability of accommodation offered by Delhi metro over a period of time. The integrated model has been applied to Delhi metro to estimate peak and non-peak hour fares with incremental increase in trains per hour to achieve an LOS of C inside the trains. It is concluded that suitable combination of demand and supply side measures shall be adopted to ease peak hour congestion in metro trains.
Modeling Passenger Ferry Service Quality Considering Observed and Latent Variables
Md. Minhajul Islam Khan, Ahsanullah University of Science and Technology (AUST)Show Abstract
Md. Hadiuzzaman, Bangladesh University of Engineering and Technology
Tanmay Das, Sonargaon University
Fahmida Rahman, Bangladesh University of Engineering and Technology
Tahmida Hossain Shimu, Military Institute of Science and Technology (MIST)
Development and Applications of Demand Responsive Transit System–Based Optimization Models
Cheng-Chieh Chen, National Dong Hwa UniversityShow Abstract
Tzu-Yu Kao, National Dong Hwa University
Providing a public transportation with quality service is critical to attract more passengers to the system. Due to low residential density with wide service area in Eastern Taiwan, conventional transit operators tend to provide a relatively low service frequency system, while also resulting in a significant fraction of passengers unwilling to utilize transit systems. This paper develops three different optimization models devoted to managing operations in demand responsive transit system (DRTS) networks, from seats assignment, vehicles dispatching, to routing tasks. The study starts from managing demand occurred along the fixed routes (i.e. similar as bus) and then considers demand occurred within or even outside the service area (i.e. similar as taxi or dial-a-ride service.) The first model starts by optimizing passengers’ seat assignment, which is done primarily by integrating passengers’ origins and destinations. A second model determines which type of vehicles should be dispatched. Finally, a set of passengers’ characteristics is considered by the proposed third model for vehicle routing problem with simultaneously pick-up and drop-off requests, especially for different fare schemes inside and outside the DRTS service area.
Identifying and Visualizing Operational Bottlenecks for Public Transport Considering Queue Length, Bus Load Profiles, and a Sharper Data Source
Alejandro Schmidt, Centro de Desarrollo Urbano SustentableShow Abstract
Jaime Moya, Centro de Desarrollo Urbano Sustentable
Diego Cruz, Directorio de Transporte Público Metropolitano - Santiago
Juan-Carlos Munoz, Pontificia Universidad Catolica de Chile
An accurate identification of speed related problems depends on both quality and precision of tools used to analyse the available operational data, which should consider impacts on users and operational costs. This study presents an extension of the tool developed by Bucknell et al. (2017), which allows to identify and visualise bottlenecks on the transit system operation based on bus speed profiles per route. Three major improvements have been made to their methodology: (i) more detailed data, with GPS signals every 250 metres and every 15 minutes, (ii) queue length estimation for identified bottlenecks, and (iii) incorporation of bus load profiles in the analysis. These extensions allow a better identification of bottlenecks and a more precise measure of their impact on bus riders. The model is applied to Santiago’s transit system, and a comparison between the results obtained with this new methodology and the output of the former tool is presented. Differences among the spatial distribution of the opportunities in the city and the critical areas are discussed. The new tool allows a better identification and prioritisation of the opportunities in the city, allowing to allocate resources in those bottlenecks with the greatest potential increase in social welfare.
Measuring Spontaneous Accessibility for Iterative Transit Planning
Matthew Laquidara, Public Transit AnalyticsShow Abstract
Public transit planners rely on measurements of network performance to anticipate the impact of changes to a transportation system. Accessibility-based measurements emphasize how well a transportation system allows individuals to reach desired opportunities, rather than maximizing network properties such as capacity. This paper presents a measurement of accessibility for transit customers making unanticipated, spontaneous trips. Measuring this Spontaneous Accessibility necessitated the development of a software tool that can analyze a transit network throughout an entire day, over a complete municipal boundary or transit agency service area, at very fine spacial granularity, and without some of the simplifying assumptions made by previous studies. The tool is used to study Spontaneous Accessibility within the city of Seattle over a one year period featuring the opening of a light rail extension and restructures of bus service. Studies of this nature require only limited data sources but produce precise results, and thus can be utilized to measure iterative refinement of the transit network. Furthermore, techniques from the discipline of information theory provide insight into ways to reduce the computational demands, giving planners the ability to consider more alternatives.
Factors Affecting Bus Users’ Satisfaction in Times of Crisis
Dimitrios Efthymiou, Technical University of MunichShow Abstract
Constantinos Antoniou, Technical University of Munich
Yannis Tyrinopoulos, Technological Educational Institute of Athens
Eleanna Skaltsoyanni, Technological Educational Institute of Athens
This paper continues the research effort that the authors begun in 2008 on the quality factors that affect adoption of public transportation and retention of its users. The objective of this paper is to explore the impacts of the deep, 7-year economic crisis, on the perceptions of public transport users, with bus as their main transport mode, about service quality. Data from three user-satisfaction surveys that were conducted in Athens in 2008 (pre-crisis), 2013 (mid-crisis) and 2017 (deep-crisis) are used for the analysis. Mann Whitney/Wilcoxon test is applied for the distribution comparison of the responses between the pairs of consecutive years (2008-2013 and 2013-2017), in order to measure the change in users’ preferences. Ordered logit models are developed for the user satisfaction and shift to public transportation after the beginning of the crisis.
The results of the analysis indicate that the satisfaction about quality attributes, such as service frequency, conditions at the stations and information provision, are important contributors of the total satisfaction, verifying the results of Tyrinopoulos and Antoniou (2008) and Efthymiou et al. (2014), but their impact varies over time. Despite the general decrease of commuting activities due to increase of unemployment, the shift to public transportation has increased. More specifically, demographic characteristics, such as age, occupation and gender, as well as qualitative factors, such as overall quality of service, service production, transfer quality, ticket services and environmental consciousness, have affected the decision of people to shift to and from public transportation.
Optimal Design of Transit Demand Management Strategies
Zhenliang Ma, Northeastern UniversityShow Abstract
Haris Koutsopoulos, Northeastern University
Yunqing Chen, Shandong University
Travel demand management (TDM) is used for managing congestion in urban areas. While TDM is well studied for car traffic, its application in transit is still emerging. A well-structured transit TDM approach can help agencies better manage the available system capacity when the opportunity and investment to expand are limited. However, transit systems are complex and the design of a TDM scheme, deciding when, where, and how much discounts or surcharges are implemented, is not trivial. This paper proposes a general framework for optimal design of TDM promotion schemes to facilitate providing the right promotion to the right users at the right time in the right place. The framework consists of two major parts: assignment and optimization. The assignment updates the origin-destination (OD) demand, assigns it to the network and estimates performance metrics given the TDM scheme. The optimization allocates resources to maximize TDM performance in a cost effective way by better targeting users. The optimal design of TDM strategies is greatly facilitated by the availability of smart card (automatic fare collection, AFC) data. The proposed approach is demonstrated with data from a busy subway system. The case study demonstrates the value of the method, compares the effectiveness of different strategies, and also highlights that the potential of various strategies may be rather limited.
Transit Vehicle Performance Analysis for Service Continuity or Termination: A Data Envelopment Analysis (DEA) Approach
Seyed Kiavash Fayyaz Shahandashti, University of UtahShow Abstract
Xiaoyue Cathy Liu, University of Utah
Ran Wei, University of Utah
Public transit agencies aim to improve services while reducing operating costs. Transit performance analysis, as the main approach to assess operating cost and revenue, has received a lot of attention in recent decades. Most of such studies focus on macro-level performance analysis by comparing across transit agencies or within a transit agency across different parts of its operation. The macro-level analysis assumes that bus drivers and vehicles have identical performance in terms of production and resource consumption. Yet they can vary significantly and directly influence service reliability and operational efficiency. As a result, micro-level vehicle performance analysis is needed for operation optimization. In this paper, we introduce an innovative and effective use of Data Envelopment Analysis (DEA) approach to estimate, project, and compare the operational efficiency of each transit vehicle. Using Utah Transit Authority (UTA)’s paratransit fleet as a case study, our study demonstrates the varying cost structures and operational efficiencies over time associated with different vehicle types. We show that such variations and time series analysis can be used to guide vehicle procurement and service continuity/termination prioritization, which further leads to significant cost savings and service reliability improvement. The proposed approach is replicable to any transit fleet with available maintenance and operation data. The proposed method provides transit agencies with data-driven analytics to facilitate decision making process.
Development of a Modified Bus Stop Capacity Model
Faheema Hisham, Queensland University of TechnologyShow Abstract
Jonathan Bunker, Queensland University of Technology
Ashish Bhaskar, Queensland University of Technology
Capacity estimation of a transit line is an essential aspect to evaluate its operation, reliability and performance. Although bus stops are identified as capacity bottlenecks, research done to improve stop capacity have been scarce. A new approach towards capacity estimation is proposed in this study by incorporating traffic blockage and interference between buses by building upon the basic relationship contained in Transit Capacity and Quality of Service Manual (TCQSM). The difference between the traditional TCQSM and the modified model is that, the TCQSM accounts the above mentioned effects as mere factors that reduce capacity and the modified method quantifies these impacts. Furthermore, sensitivity analysis results are provided to compare both models. The proposed model provides a better understanding of stop operations and can be useful to transit planners in improving bus stop operations and providing better service reliability.
Optimizing Train Service Plans to Coordinate Transport Capacity for Urban Rail Transit Lines
Sijie Li, Tongji UniversityShow Abstract
Ke Han, Imperial College London
Wei Zhu, Tongji University
In view of big passenger flow volume and high passenger risk at transfer stations during the peak period, this paper studied the coordination method of urban rail transit network transportation organization from the perspective of capacity matching. The change law of passenger flow was analyzed, and the calculation methods of train remaining carrying capacity, waiting passenger demand and the largest number of people gathered on the platform were determined. The concept of capacity coordination degree (CCD) was proposed, used to describe the matching degree between traffic demand and transport capacity of each line. Based on this, taking the optimal comprehensive CCD of the transfer station as the goal, the first train departure time and train departure interval as decision variables, and guarantee of passenger safety within station as the main constraint, a nonlinear integer programming model of train service plans collaborative optimization was established, and the genetic algorithm was designed. A case study of a two-line intersecting network was carried out. The results show that, after the use of capacity coordination scheme, the total number of running trains increases by only 1, the number of remaining passengers reduces by 68.44%, comprehensive CCD is closer to 1, and the largest number of people gathered in big passenger flow directions decreases by 11.77% and 19.68%, respectively. Transport supply can better meet the passenger demand in all directions, effectively improving the interests of both passengers and operators.
Transit Riders and the Subjectively Shortest Path: A Look into Transit System Resilience and User Choice Preferences
Jacqueline NowakShow Abstract
Alireza Khani, University of Minnesota, Twin Cities
Schedule-based shortest path and multicriterion shortest path algorithms are combined to investigate whether transit riders choose to take the shortest path between their origin and destination, a subjectively shortest path, or neither. A label setting algorithm is applied to transit schedule data along with five distinct time weighting schemes chosen to represent extreme attitudes toward trip attributes: high in-vehicle time, high waiting time, high walking time, and high transfer penalty. Results show a significant amount of overlap between paths generated for different weighting schemes, despite their extremity. This could indicate limited route options due to low redundancy and therefore low network resilience. Paths extracted from automated fare collection data align most closely with a baseline scenario with no extreme time weighting (i.e. the shortest travel time path). Future work will include investigation of intermediate weighting schemes and clustering analysis of passengers.
Commuter Travel Cost Estimation at Different Levels of Crowding in Suburban Rail System: A Case Study of Mumbai
Prasanta Sahu, Birla Institute of Technology and ScienceShow Abstract
Gajanand Sharma, Birla Institute of Technology and Science Pilani
Anirban Guharoy, BITS Pilani
Mumbai suburban train system (MSRS) is one of the most crowded train systems in the world. The system offers two types of train services: Fast and Slow. It carries about 8 million passengers on a typical weekday. This research attempted to value travel attributes such as waiting time, in-vehicle time and crowding level using the behavioural data obtained through stated preference experiment on 896 MSRS train users. Actual on board crowding images from a train coach were considered to perceive crowding more realistically by the local train user. Multinomial logit modelling technique was used for estimating commuter travel cost (time) at different levels of crowding. Model estimation results showed that there is increase in total perceived in-vehicle travel time with the increase in level of crowding. Traveling in a crowded seating condition increases travel cost by 0.81 minutes per one minute travel. Standing up to 10 minutes in crowded condition with standee density of 7-9 passengers per m 2 increases travel cost by 0.94 minutes per one minute travel. Similarly, standing up to 20 minutes in such a condition will increase travel cost by 1.26 minutes per one minute travel. Super dense crush crowding situation with standee density 14-16 passengers per m2 increase travel cost by 2.82 minutes per minute travel. A crowded seat impose user to perceive 81% increase in in-vehicle travel time, whereas this perception becomes 282% more during travel in super dense crush crowding than normal travel conditions. Generalized travel cost becomes maximum in super dense crush crowding. The rate of decrease in utility decreases after a certain time of standing in crowded conditions. Females tend to perceive more decrease in utility due to crowding as compared to that of males. They prefer slow train over the fast train due to crowding. Similar observations were noted for higher income (more than ₹45000 per month) group users. Effect analysis suggested that the females are more sensitive to travel attributes as compared to that of male. Users belonging to business community prefers slow train, possibly due to flexible working hours. Users with age greater than 36 years prefer less crowded train; this behaviour might be attributed to physical constraints, tendency to trade off comfort with other attributes and predisposition towards comfort. Travellers with trip length more than 24 km were more sensitive to standing time in ‘super dense crush’ load condition as compared to users travelling for a trip length less than or equal to 24 km. Reduction in seating capacity of the 9-coach trains could be an effective policy measure to negotiate overcrowding and minimize the generalized travel cost. Presented discussions in this research is important to policy makers and planners in Mumbai Railway Vikas Corporation and to monitor, measure and develop programs for the MSRS operation service quality. In summary, the study finding will be useful for developing a policy framework to deal with issues related to level of service improvement for suburban rail system in India and other developing economies.
Constructing a Nationally Comparable Measure of Transit Performance Using GTFS Data
Timothy Welch, Georgia Institute of Technology (Georgia Tech)Show Abstract
Amit Kumar, Georgia Institute of Technology (Georgia Tech)
Alyas Widita, Georgia Institute of Technology (Georgia Tech)
Sabyasachee Mishra, University of Memphis
A well-functioning public transportation system is critical to the movement of people and goods across a city. But determining how the transit system is performing can be a challenge. The complexity of analyzing transit networks with their typically large spatial footprint and multi-modal nature is only made more difficult by a traditional lack of robust transit data. Historically, this unavailability of transit data has limited the potential for the development and adoption of comprehensive public transportation performance measures. The proliferation of GTFS data in recent years has opened the door to a new generation of transit assessment. In this paper, we develop an analytical approach for building a comprehensive transit-system performance model using GTFS datasets. Multiple transit networks are analyzed with a robust set of connectivity measures at the stop and route level. We then compare the level of transit connectivity across systems, providing a new national transit performance ranking model.
Attitudinal Segmentation of Public Transit Riders with Clustering and Bayesian Regression to Understand Heterogeneity in Satisfaction
Joel Huting, Twin Cities Metro TransitShow Abstract
Eric Lind, Metro Transit, Minneapolis-St. Paul
Kim Ky, Twin Cities Metro Transit
Rebecca Freese, Twin Cities Metro Transit
Public transit rider satisfaction is well-studied in the academic literature and transit industry. Numerous studies have focused on the factors that drive overall satisfaction and thus provide ample insights to transit agencies on investment priorities. However, there is less published research on the difference in satisfaction across transit mode (light rail, commuter rail, bus), bus route-type (express, arterial bus rapid transit, local service), or demographic groups. This study builds the body of research by providing a comprehensive assessment of public transit rider satisfaction among Metro Transit riders in the Minneapolis/St. Paul metropolitan area. Additionally, it proposes a methodology for analyzing surveys that addresses the categorical and interdependent nature of survey data – a process that employs Gower’s distance and a partitioning around medoids (PAM) clustering algorithm to segment riders based on attitudes along with a Bayesian logistic regression model to profile the unique identified clusters. Light rail, arterial bus rapid transit, express, and particularly commuter rail riders were much more likely to be satisfied when compared to local bus riders. Satisfaction tended to increase with age, low and high-income riders were more satisfied than middle income riders, people of color tended to have slightly lower satisfaction than white riders, while riders who reported having a disability were somewhat more satisfied. Transit reliant riders tended to be less satisfied, whereas new transit riders (less than two years of riding experience) were more satisfied than more experienced riders. Riders who had experienced various forms of street harassment on transit were less satisfied.
Perceived Service Quality in Urban Rail Transit: A Comparison of Structural Equation Models
Amirali Soltanpour, Amirkabir University of TechnologyShow Abstract
Mahmoud Mesbah, Amirkabir University of Technology
Meeghat Habibian, Amirkabir University of Technology
Promoting transit services requires a deep understanding of service quality and in turn factors that affect passenger satisfaction. There are objective and subjective means to evaluate service quality. Since people are attracted to transit services based on their own subjective perception, the judgment provided by transit users is what matters. However, there are various approaches in the literature to model service quality in which different lists of factors are identified which in turn affect passenger satisfaction. This paper aims to test the transferability of these approaches to another city. The common essence of the compared approaches is to evaluate the relationship between perceived service quality and customer satisfaction using a structural equation model (SEM).
Four classes of approaches were tested to identify the factors affecting service quality: 1) recommendations of previously published models from the literature, 2) an established customer satisfaction theory, 3) adopting an exploratory approach, and 4) a formal model development procedure which combined the previous three approaches.
Tehran Metro Line 3, which is a heavy urban rail transit, was used as a case study. Three hundred validated responses were collected through a customer satisfaction survey. The results indicate that none of the first three approaches were directly transferable to our case study. The best model is developed when the entire modeling procedure is repeated using a wide range of affecting factors and gradually narrowing them down through model development. The final model consisted of four latent variables, namely major services, comfort, security and minor services, the first two of which had the largest effect on service quality.
A Psychological Investigation on Private Vehicle Users Toward Public Transport Usage in a Developing City
Ngoc Nguyen, Nagoya UniversityShow Abstract
Tomio Miwa, Nagoya University
Psychological determinants have received much attention in explaining choice behaviors, including decisions on using travel modes. However, this issue is still poorly understood in the context of developing cities, where many residents refuse to use public transport. Take Ho Chi Minh City, Vietnam as a case study, we aim to investigate psychological determinants relating to decisions on using the current system of buses and the being developed system of public transport. A web-based survey was conducted on 1030 private vehicle users in June 2017. A structural equation modeling approach was then used to capture hypothesized psychological determinants through a three-step framework. The results confirm the existence of five psychological determinants including “awareness,” “car passion,” “bus dissatisfaction,” “agreement to public transport projects,” and “new public transport dissatisfaction.” Four individual characteristics contribute to structuring the psychological determinants include the status of being students, driving a motorcycle, driving a car, and working five days or more per week. In addition, the “awareness,” “agreement to public transport projects,” and “bus dissatisfaction” were independently observed to have impacts on decisions toward using current buses or/and new public transport. These findings could be the basis for further analyses on models of travel behaviors and on transport policies in developing cities.
Method of Characterizing Passenger Congestion State in Bus Transit
ChangPeng Guo, Tongji UniversityShow Abstract
Jing Teng, Tongji University
From the travel perception of bus passengers, the definition of passenger congestion (PC) in bus transit was proposed characterizing both time congestion and space congestion of passenger flow. Then theoretical calculation model of passenger congestion in bus transit was built based on interval. Moreover, the function relations among change of space congestion, time congestion and passenger congestion were concluded. In this paper, relevant influencing factors of PC state in bus transit were analyzed and the method for analysis and evaluation on effectiveness of improvement strategies was formed via sensitivity analysis on characteristic variables. Evaluation and supervision on operation service state of bus transit corridor are theoretically supported by research findings.
Analysis of Waiting Time Perception of Bus Passengers Provided with Mobile Service
Po-Chieh Wang, National Taiwan UniversityShow Abstract
Yu-Ting Hsu, National Taiwan University
Smartphones and relevant mobile service have greatly influence people’s daily lives, and smartphone addicts can be commonly seen at transit stops/stations, playing with their smartphones while waiting for buses/trains. This research presents a holistic perspective for analyzing the effects of smartphone usage on transit passengers, which also considers the effects at a psychological level. Such effects may be manifested as the reduction of perceived waiting time at stops/stations against the negative emotionality induced by long waiting, such as boredom and tediousness, so as to result in improved travel experience. An on-site survey is designed and implemented over the bus system in Taipei, Taiwan, to collect the revealed responses of bus passengers in regard to waiting time perception and smartphone usage, particularly for travel-irrelevant mobile service. The survey data are modeled and analyzed in both numerical and verbal representation of perceived waiting time by using a multiple linear regression model and a cumulative proportional odds logistic model, respectively. A finite mixture model is further employed to investigate the potential heterogeneity of waiting time perception related to using smartphones for travel-irrelevant mobile service. The analysis results highlight that using smartphone for travel-irrelevant mobile service may lead to the reduction of perceived waiting, and the effect can be more significant for young passengers and when waiting time is prolonged. Such findings can contribute to the holistic consideration of passenger behavior in transit system planning and associated information/service provision strategies.