This session includes intelligent transportation systems, transit signal priority, as well as modeling and optimization for bus operation, design, and application. Intelligent transportation systems are making the bus transit system operation smarter and more efficient. Many transit agencies have adopted a number of different intelligent transportation systems and technology in order to enhance and/or supplement their transit services to the public. Audiences who are interested in bus transit operations, intelligent transportation systems, and technology are welcome to join this session.
A Sensitivity Analysis on the Transit Signal Priority Requesting Threshold and the Impact on Bus Performance and General Traffic
Michael Sheffield, Wall Consultant Group (WCG)Show Abstract
Grant Schultz (firstname.lastname@example.org), Brigham Young University
David Bassett, PricewaterhouseCoopers LLP
Dennis Eggett, Brigham Young University
An analysis was performed to evaluate the impact of changing the transit signal priority (TSP) requesting threshold on bus performance and general traffic using field-generated data exclusively. Route 217, a conventional bus route that utilizes a dedicated short-range communication (DSRC)-based TSP system as part of its normal, day-to-day operations was analyzed over a 3-month period from May 2019 through August 2019. The requesting thresholds evaluated for Route 217 were 3, 2, and 0 minutes, which stipulate how far behind schedule the bus must be in order to request TSP. For each requesting threshold, bus performance was evaluated through on-time performance (OTP), schedule deviation, travel time, and dwell time, while the traffic analysis was performed by evaluating split failure, change in green time, and the frequency at which TSP was served. A combination of observational and statistical analyses concluded with convincing evidence that OTP, schedule deviation, and travel time improve as the requesting threshold approaches zero with negligible impacts to general traffic. As the requesting threshold changed from 3, to 2, to 0 minutes, OTP increased 2.0 and 2.5 percent, respectively; mean schedule deviation improved 15.9 and 20.9 seconds, respectively; and travel time decreased at 75 percent of timepoints. Meanwhile, negative impacts to traffic occurred if an increase in split failure was measured after TSP was served, a phenomenon observed a maximum of once every 43 minutes. Thus, it is concluded that bus performance improves as the requesting threshold approaches zero with inconsequential impacts to general traffic.
Safety Effects of Transit Signal Priority Using the Full Bayesian Approach
MD Sultan Ali, Florida International UniversityShow Abstract
Angela Kitali, Florida International University
John Kodi, Florida International University
Priyanka Alluri, Florida International University
Thobias Sando, University of North Florida
Transit signal priority (TSP) is a strategy that prioritizes the movement of transit vehicles through a signalized intersection to provide better transit travel time reliability and minimize transit delay. Although TSP is primarily intended to improve the operational performance of transit vehicles, it may also have substantial safety benefits. This study explored the potential safety benefits of the TSP strategy deployed at various locations in Florida. An observational before-after full Bayes (FB) approach with a comparison-group was adopted to estimate the crash modification factors (CMFs) for total crashes, rear-end crashes, sideswipe crashes, and angle crashes. The analysis was based on 12 corridors equipped with the TSP system and their corresponding 29 comparison corridors without the TSP system. The deployment of TSP was found to reduce total crashes by 7.2% (CMF = 0.928), rear-end crashes by 5.2% (CMF = 0.948), and angle crashes by 21.9% (CMF = 0.781), and these results are statistically significant at a 95% Bayesian credible interval (BCI) except for the rear-end crashes. On the other hand, sideswipe crashes increased by 6% (CMF = 1.060) although the increase was not significant at a 95% BCI. Overall, the results indicated that TSP improves safety. The findings of this study may present key considerations for transportation agencies and practitioners when planning future TSP deployments.
Residents’ Attitudes Toward a Bus Network Redesign in Chattanooga, Tennessee
Abubakr Ziedan, University of TennesseeShow Abstract
Cassidy Crossland, University of Tennessee, Knoxville
Candace Brakewood, University of Tennessee, Knoxville
Philip Pugliese, Chattanooga Area Regional Transportation Authority
Harrison Ooi, University of Tennessee
Many transit agencies are considering or implementing bus network redesigns. This is one of several approaches that are being implemented to counteract a trend of decreasing transit ridership in recent years in the United States. In light of this growing trend, this study investigates local residents’ attitudes toward a bus network redesign in Chattanooga. This study used survey data collected from the city of Chattanooga as part of their bus network redesign process. This study estimated three logit models to explore residents’ attitudes toward different bus network redesign philosophies, transit network improvements, and willingness to pay for transit improvements and expansion. Three main findings were revealed. First, respondents that ride the bus prefer access to more places over frequent bus service while non-riders prefer more frequent service. Second, this study showed that younger generations are more supportive of transit improvements than older age groups. Third, people living near bus routes and those with higher income are more willing to pay for transit improvements. The findings of this study should inform transit agencies that are considering or in the process of redesigning their bus network.
Implementation Sequence Optimization for Dedicated Bus Lane Projects
Murat Bayrak, Pennsylvania State UniversityShow Abstract
S. Ilgin Guler (email@example.com), Pennsylvania State University
Paul Schonfeld, University of Maryland, College Park
Transportation projects are often implemented in phases, and the total project duration can span years. Optimization of the sequence in which transportation projects are implemented can decrease the severity of disruptions caused by construction, reduce the total cost of projects, and increase the present value of the benefits of the project. This paper presents a method for optimizing the sequence and location of dedicated bus lane implementations. The proposed method is based on a bi-level optimization framework. The lower-level simulates the traffic using a link transmission model to evaluate car and bus delays, while the upper-level determines the optimum locations of bus lanes and the optimum implementation sequence of a bus lane configuration. If budget or other resource constraints are binding, the optimum sequence uniquely determines the optimum schedule. The solution method is evaluated for a test network, and an analysis of sensitivity to different demand patterns and different bus lane configurations is conducted. First, the optimum locations of bus lane implementations are determined for an illustrative test network. Results show that, for the tested optimum set of bus lane locations, optimization of the bus lane implementation sequence does not yield significant benefits. Considering implementing bus lanes on every possible link, prioritizing the implementation of bus lanes on links with large delay savings decreases the total cost of implementation. The results of the non-uniform demand pattern scenario are similar and also prioritize the implementation of bus lanes at locations that most benefit the network.
Deadhead Minimization with a Flexible Facility Locator Tool
Kara Todd, Georgia Institute of Technology (Georgia Tech)Show Abstract
Freyja Brandel-Tanis, Georgia Institute of Technology (Georgia Tech)
Daniel Arias, Georgia Institute of Technology (Georgia Tech)
Kari Watkins, Georgia Institute of Technology (Georgia Tech)
As transit agencies expand, they may outgrow their existing bus storage and service facilities. When selecting a site for an additional bus facility, an important consideration is the change in bus deadhead time, which impacts the agency’s operating costs. Minimizing bus deadhead time is the subject of many studies. However, transit agencies are not consistently equipped with this information when choosing a new bus depot site. This study presents a flexible tool for bus facility location. It evaluates each candidate site based on a given bus network and existing depots, calculating the network minimum deadhead time for each potential new set of facilities. Importantly, the tool is also designed to be ready for use by any transit agency, no matter their resources. It is created with open-source software and uses only General Transit Feed Specification (GTFS) and other data inputs readily available to all transit agencies in the United States. The tool is demonstrated through a case study with the Metropolitan Atlanta Rapid Transit Authority (MARTA), which is considering a new bus depot as it builds its Bus Rapid Transit network. The case study used current MARTA bus GTFS data, existing depot locations, and vacant properties from Fulton County, Georgia. The tool evaluated 17 candidate sites and found that the winning site would save 29.8 deadhead hours on a typical weekday, which translates to more than $12,000 daily based on operating cost assumptions. This tool provides important supplemental information to transit agencies evaluating sites for a new bus depot.
Developing a Bus Eco-driving Strategy with Consideration of Holding Control
Xumei Chen (firstname.lastname@example.org), Beijing Jiaotong UniversityShow Abstract
Qianwen Ye, Beijing Jiaotong University
aihua Fan, Xuchang University
Yixin Zhang, Beijing Jiaotong University
Lei Yu, Texas Southern University
Intersection and bus stop are typical areas that need eco-driving to improve the efficiency and environmental benefit for bus transit system. For near-side bus stops which are located a short distance upstream of a signalized intersection, whether the holding control strategy can be used to enhance the eco-driving effects is still an unexplored issue worthy of attention. In this context, a study on developing the bus eco-driving strategy considering holding control is conducted in this paper. A two-stage speed guidance for bus on the segment from bus stop to downstream intersection is proposed. The eco-driving model for bus considering holding control is developed by minimizing total costs of bus travel time and fuel consumption. Seven bus stations of the Bus Route 15 and the corresponding intersection downstream in Beijing are selected as the case study. The results show that the developed eco-driving strategy considering bus holding has a significant effect in reducing the average fuel consumption, decreased by over 26%, and average number of stops and passenger travel time, decreased by 38% and 27%, respectively, representing that passenger travel efficiency is not compromised under the environmental-friendly driving strategy. The sensitivity analyses of three parameters, including the coefficient of VOT (value of travel time), distance between the stop and downstream intersection, and maximum bus speed limit, indicate that the distance between the bus stop and intersection has significant impacts on the optimized speed trajectory.
Evaluation of a Bus Collision Avoidance Warning System in University Towns
Noah Goodall (Noah.Goodall@VDOT.Virginia.gov), Virginia Department of TransportationShow Abstract
Peter Ohlms, Virginia Department of Transportation
Towns and small cities with major universities represent a unique safety challenge for transit operators, whose routes may vary from suburban and rural settings to dense campuses with high volumes of vulnerable road users (VRUs). Further complicating operations, pedestrian and cyclist volumes are highly seasonal depending on the universities’ academic calendars, requiring operators to adjust to variable conditions, often week-by-week. This study represents the first evaluation of a bus-VRU collision avoidance and warning system (CAWS) in university towns. A CAWS was installed on nine buses at three Virginia transit agencies serving large universities in Blacksburg, Harrisonburg, and Lynchburg. Municipal populations of these localities range from 45,000 to 80,000, and university enrollments range from 20,000 to 110,000. Data was collected in stealth mode without alerting drivers for at least two months, and in live mode with driver alerts for at least six months. Surveys were collected from 41 operators after driving buses with the CAWS in live mode. With few exceptions, CAWS produced statistically significant reductions of between 9 and 80% in warnings and alerts across most routes (Welch’s t-test, p < 0.05). Similar reductions were found when isolating weekdays, weekends, and days with classes in session. Operators had mixed reactions to the system, with 43% finding it not helpful and 90% citing distractions. These results align with findings from previous studies in that the CAWS improved safety surrogates yet was unpopular with many operators.
Ridership Prediction of New Bus Routes at Stop Level by Modelling Socio-economic Data using Supervised Machine Learning Methods
Yatri Patel, University of Tennessee, ChattanoogaShow Abstract
Connor Firat, University of Tennessee, Chattanooga
Tegan Childers, University of Tennessee, Chattanooga
Mina Sartipi, University of Tennessee, Chattanooga
Predictive modeling is key to studying passengers' behavior in transportation research. Modelling the public transport system can be used to estimate present and future demand and users’ trend toward public transport services. Machine learning techniques have proven to be better at recognizing the patterns and relations in the data. While, the traditional techniques are aimed at forming casual relationships and are unable to recognize patterns in the data. This paper seeks to predict the ridership at stop level for the new bus routes using the socio-economic data, building data, and ridership data of the existing routes at stop level. Neural networks (NN), a machine learning method has been applied to build predictive models. Ridership of the existing routes has been used to train and validate the model performance, which is able to predict the public transport ridership of the new routes. This model can be used by public transport agencies and relevant government organizations to predict the public transport demand for new commuters before introducing any new changes in the public transit system.
Impact of Public Transit Bus System Redesign on Baltimore City Public School Students
Fathy Elgendi, Maryland Transportation AuthorityShow Abstract
Young-Jae Lee, Morgan State University
Celeste Chavis, Morgan State University
Baltimore City Public School System (BCPSS) follows an open choice model for middle and high schools, and student transportation is provided by the Maryland Transit Administration (MTA). Studies have shown that BCPSS students traveling by public transit have travel times longer than Baltimore residents' work commutes. Public transit routes are designed for access to jobs that were historically located in the Central Business District (CBD) and not to serve students and schools which are more dispersed throughout the city. The objectives of this research is to evaluate the effect of the MTA transit service redesign on BCPSS students. This was achieved by performing a travel time analysis to schools pre and post network redesign and identifying residential locations (aggregated to traffic analysis zones (TAZs)) and school pairs with insufficient service based on key performance metrics.
Multi-objective Optimization Approach for Feeder Bus Routes Based on Service Area Determination and Stops Selection
Shanshan Liu (email@example.com), University of Illinois, Urbana-ChampaignShow Abstract
Xiucheng Guo, Southeast University
Meina Zheng, Southeast University
The optimal feeder bus is effective to solve the first and last mile problem. This research proposes a comprehensive approach to design the service area, stop layout and route layout of feeder bus in sequence. Based on community detection and the attractive scope of metro station, the service area of feeder bus can be determined. Considering the potential feeder demand, feeder bus stops are selected from the set of the entrance and exit of residential area and the existing bus stations. And the demand allocation method is developed to distribute stop demand into road sections. From the perspective to the feeder bus passengers, a multi-objective optimization model for feeder bus routes is established. The upper-level objective is to maximize the demand for feeder bus service and the lower-level objective is to minimize the travel cost of feeder bus passengers. The research takes the feeder bus system in the downtown area in Suzhou, China to apply the proposed optimization approach. The results illustrate that 36 community bus line layout plans are generated.
Impacts of Lane Transit District's EmX Bus Rapid Transit (BRT) on Area Residential Property Values
Victoria Perk, University of South FloridaShow Abstract
Martin Catala, University of South Florida
Vanko Antonov, SpartaServe Consulting
Bus rapid transit (BRT) has been growing in popularity in the United States and the mode’s impacts on property values and land uses need to be better understood. Economic theory suggests, and literature has shown, that people are willing to pay higher housing costs to reduce their transportation costs of accessing areas of economic activity. Does high-quality BRT service reliably provide such access and, thereby, increase residential property values? The hypothesis is that property values are higher closer to BRT stations, reflecting a premium for the access provided by the BRT service. The literature reveals that, to date, relatively little work has been done on U.S. BRT systems’ impacts on property values using robust techniques. Further, because every BRT system is different, it is helpful to analyze additional systems to estimate how modern U.S. BRT may affect surrounding property values. This research contributes to the body of literature on this topic by examining Lane Transit District’s EmX BRT service (Eugene, Oregon) using econometric modeling to estimate changes in property values associated with the BRT. The findings indicate that the EmX BRT does positively impact surrounding single-family home sale prices. Results are statistically significant yet, as expected, relatively small in magnitude. An interesting finding is that the impact of the EmX stations on property values increased in each of the three years examined in this study (2005, 2010, and 2016). These results provide further insight into how BRT can enhance the livability and economic development in a community.
Multi-Modal Traffic Flow in Shared Bus-Bike Lanes: A Scoping Literature Review in 1 Comparison with Baltimore SBBL Infrastructure
Istiak Bhuyan, Morgan State UniversityShow Abstract
Celeste Chavis, Morgan State University
Chappelle Branch, Morgan State University
Providing separate bicyclists from high‐volume traffic improves safety but may not be always possible for a congested city. Sharing the lane with buses can deliver similar outcome but the Shared Bus-Bike Lanes (SBBL) are relatively new concept in the U.S. This study evaluates the existing literature on SBBL operations in the U.S. for better understanding of design, safety, and delay on selected performance measures. Baltimore had implemented 10 corridors of SBBLs in 2017 after complete overhaul of their transit system. This study also captures 55 hours of high definitions video data on the selected SBBL locations and evaluate the performance measures. The study will be beneficial for future design, safety 8 aspects, policy implementation, and effectiveness of SBBLs throughout the country.
Developing an Optimal Integrated Single Framework Algorithm for the Multi-Level School Bus Network Problem
Amirreza Nickkar, Morgan State UniversityShow Abstract
Young-Jae Lee (YoungJae.Lee@morgan.edu), Morgan State University
Since the last decades, many scholars attempted to improve the school bus routing problem by making it more realistic and practical. In the States, school bus fleet are assigned to serve students in three levels high school, middle school, and elementary school sequentially however, in the past studies each of these stages in the problems considered separately. Respect to recent improvements in intelligent transportations, necessity of having a unit framework for routing of school buses in both time windows of morning and afternoon, has been increased more than ever.This study introduces a novel integrated school bus problem that considers sequential operation of fleet for all three levels in a unit framework. An algorithm based on the simulated annealing (SA) method was developed to find the optimal routes based on minimization of school buses’ operating cost and total student traveling time. An example on a hypothetical network was developed and tested to demonstrate the developed algorithm. The algorithm successfully handled the integration of school buses’ optimal route generation while it met all constraints.The results showed that the routings by the integrated single framework algorithm can save the total costs by 4.5% to 12.4% compared to the routings with the separated level algorithm. Also, it showed that the total costs of the integrated routing framework for different morning and afternoon time windows are 8.28% less than the same routings (identically reversed) for morning and afternoon time window.
PRINCIPLES FOR SETTING SINGLE OR MULTILINE BUS HOLDING CONTROL BASED ON NETWORK CHARACTERISTICS
Georgios Laskaris (firstname.lastname@example.org), Delft University of TechnologyShow Abstract
Oded Cats, Delft University of Technology
Erik Jenelius, KTH Royal Institute of Technology
Marco Rinaldi, Universite du Luxembourg
Francesco Viti, Universite du Luxembourg
Public transport networks often include one or more sets of common consecutive stops between different lines to offer more capacity in busier segments, or to allow transfers. In such networks, both single line and multiline control can in principle be applied. In this study, we investigate the effect of both the size of different segments of the network and the characteristics of demand distribution on the performances of single line and multiline control. After introducing the key elements that characterize networks with overlapping segments, two sets of scenarios (a stop set size and a demand-based scenario) are conducted on different network configurations, for both control schemes. Results show that the choice between the two control alternatives is more sensitive to demand distribution than to the lines’ topology. Passenger groups traversing different stop sets are the most consequential in terms of chosen control strategy’s optimality. The results suggest applying multiline shared transit corridor control for corridors given that those stops account for at least 50% of the total number of boarding passengers.
Covid-19 and Changes in Route-Level Transit Ridership in a University Town
Sagar Patni, University of FloridaShow Abstract
Sivaramakrishnan Srinivasan, University of Florida
Juan Suarez, University of Florida
COVID-19 has affected public transportation worldwide, resulting in an unprecedented decline in transit demand. The broad goal of this study is to examine the impacts of Covid-19 on route-level transit ridership in the university town of Gainesvile, Florida, USA. In particular, the study sought to examine how the impacts varied based on the land use along the route, and, further, how these impacts have varied over time (March – June of 2020). Ridership data were assembled for the transit routes for the months of March – June for the years 2018-2020. Land use patterns along the routes were assembled using GIS-based datasets and overlay procedures. Monthly ridership per vehicle revenue hour is considered as the dependent variable in a linear-regression model. Overall, the models captured how COVID-19 had a differential impact on different transit routes (with varying land use patterns) even after controlling for routine seasonal variations caused by the university. For the pre-COVID years, the models predict a demand pattern consistent with the UF schedule. For the year 2020, the model captures a major reduction in ridership between April and March and between May and April. The demands for June were comparable to that of May. The greatest drops in ridership were for routes that are predominantly university-serving compared to routes that serve other parts of Gainesville. The proposed framework can be extended using data for future months of 2020 to examine how the ridership trends will continue for the rest of this year.
Optimal Dynamic Demand Responsive Feeder Bus Network Design for a Short Headway Trunk Line
Amirreza Nickkar, Morgan State UniversityShow Abstract
Young-Jae Lee (YoungJae.Lee@morgan.edu), Morgan State University
Improving accessibility is one of the major issues in suburban transportation. With recent technological advancements, it is expected that demand responsive feeder transit services can improve mobility in urban and suburban areas where the accessibility to public transit is limited. When the headway of the rail service is long enough for feeder buses to come back by the next train, then the feeder network algorithm is rather easy, because the maximum feeder service cycle time is determined by the rail headway, and matching between feeder buses and the trains is not necessary. But if the headway of the rail service is not long enough for the feeder buses to return before the next train, then the algorithm should find not only matching between passengers and feeder buses, but also matching between feeder buses and trains. In this research, an algorithm for the optimal flexible feeder bus routing for a short headway trunk line, which also considers relocation of buses for multiple stations and trains, is developed using a simulated annealing (SA) algorithm. An example with a 5-minute headway rail trunk line was developed and tested to demonstrate the developed algorithm. The algorithm successfully handled relocating the buses when the optimal bus routings were not feasible with the available buses at certain stations. Furthermore, the developed algorithm considered the multiple relocations of vehicles while minimizing total cost, including total vehicle operating costs and total passenger in-vehicle travel time costs.
DISCLAIMER: All information shared in the TRB Annual Meeting Online Program is subject to change without notice. Changes, if necessary, will be updated in the Online Program and this page is the final authority on schedule information.