Robust Bus Scheduling Considering Transfer Synchronizations
Konstantinos Gkiotsalitis, University of TwenteShow Abstract View Presentation
Oskar Eikenbroek, Universiteit Twente
Oded Cats, Delft University of Technology
In this study, we model the bus scheduling problem considering transfer synchronizations. Our mathematical model accounts for the variability of the travel and dwell times of bus trips, the regularity of individual bus lines, and the regulatory constraints related to the schedule sliding prevention and the layover time limits. To perform a synchronization of multiple bus lines that is robust to travel time and dwell time variabilities, we tackle this problem using the minimax decision rule of 2-player games where one player selects the optimal dispatching times for some specific travel and dwell time noise, whereas the other player selects the worst-case travel and dwell time noise for a given dispatching time solution. In a validation of this approach in two bus lines in Stockholm using 1 month of actual vehicle location and passenger counting data, we demonstrate the potential improvement in terms of service regularity and increased synchronizations. Finally, we examine how the performance of bus schedules in common-case scenarios is affected when their robustness to extreme deviations is increased.
Designing Bus Rapid Transit Systems: Lessons for Service Reliability and Operations
Oded Cats, Delft University of TechnologyShow Abstract View Presentation
Robert Ishaq, Technion Israel Institute of Technology
Cities worldwide are looking for expanding the capacity of their public transport system while considering budget limitations. Bus Rapid Transit (BRT) systems are increasingly considered as alternatives for designing a mass public transport in mid-size cities in developed countries. While operations have been recognized as an important success factor, previous studies have focused on infrastructure design and planning principles of BRT. We study the operations of BRT related to service reliability and service utilization and derive lessons for planning and operations. The study is cantered on the performance of the Matronit BRT system in Haifa, which is the first mass transit network of its kind in Israel. The inter-relation between service reliability, fleet management and service utilization are analysed. The speed and reliability improvements attained by the infrastructure and technological priority measures need to be complemented with control instruments to yield further gains for both service users and service provider.
Structural, Economic, and Environmental Analysis of Guadalajara Bus Network
Orlando Barraza Aguilar, Universitat Politecnica de CatalunyaShow Abstract View Presentation
Sergio Garmendia, Rijksuniversiteit Groningen
Miguel Estrada, Universitat Politecnica de Catalunya
This paper provides a model for designing bus network considering the impacts on environment, operating cost, structure and accessibility. This methodology was applied to the city of Guadalajara, Mexico. In order to compare the new design generated by model to the existing bus network in this city based on a set of performance metrics. We proposed a transit system design consisting on a grid center with bifurcation in the peripheries. This method relies on continues models which took in consideration a set of variables summarized in 3 different components: environment, user and agency component. Each of them capture key features such as the environmental impact, level of service to users and operational cost. The model generated an idealized layout that was modified to the urban reality taking into consideration a previous design. This method gave information about carbon dioxide (CO 2 ) emissions, operation cost, and structure of the proposal network. The method also encompasses the estimation of some network sciences proprieties. It is based on an estimation of the grade of accessibility which measures the efficiency of travel time. A dissolved network data set was generated and calculated its travel time matrix cost. The result was an alternative network with a better overall performance reducing from 283 routes to 40, having a drop of 58% in the operation cost which currently is 76,819 USD/ hour of service, a reduction of 57% in CO 2 emissions and an improvement in the travel time of 20%.
Optimal Berth Allocation at Bus Terminal Integrated with Passenger Choice of Overlapping Routes
Hui Jin, Soochow UniversityShow Abstract View Presentation
Haiming Hao, Ningbo Institute of Industrial Technology Chinese Academy of Sciences
Xiaoguang Yang, Tongji University
Dong Zhang, Dalian University of Technology
Bus terminal features multiple berths with overlapping routes, where passengers have to decide which bus route to take and which berth to wait on. When overlapping routes are assigned to different berths, passengers may struggle, sometimes with heavy luggage, from the current berth to others to catch the newly arriving bus. To facilitate passenger waiting and to avoid disorder movement at bus terminal, route assignment to berths is modeled with mixed integer nonlinear programing, with the objective to minimize passengers’ transfer distance between optional berths, represented with passenger demand multiplied by distance between berths serving overlapping routes. Constraints of service and waiting capacity of each berth are established to secure service quality. Then genetic algorithm (GA) combined with Greedy Descend Hill Climber is developed to solve the modeling. Case study shows that the proposed method is capable of significantly reducing passengers’ transfer distance with more rapid and stable convergence than standard GA. Moreover, the proposed berth assignment manages to promote cost-effective terminal construction and management as fewer berths are required to serve the routes. This is emphasized with sensitivity analysis that reducing bus dwell time performs better than enlarging berth waiting area to reduce passenger transfer distance. This research provides insights to design/redesign and management of bus terminal and to enhance bus service quality.
Optimal Operational Strategies for Multiple Bus Lines Considering Passenger's Preferences
Chunyan Tang, Dalian Maritime UniversityShow Abstract
Avishai Ceder, Technion - Israel Institute of Technology
Ying-En Ge, Shanghai Maritime University
Na Wu, Chang'an University
A public transit system with multiple fixed bus lines faces non-uniform fluctuated passenger demand. This non-uniformity warrants the use of public transit operational strategies, to attain efficiency. This study proposes a methodology for optimizing integration of multiple bus lines operation by applying five operational strategies: full route operation, short turn, limited stop, deadheading, and a mixture of either two or three of the latter three strategies. The operational strategies developed improve the efficiency of bus-lines operation and accommodate the observed passenger demand in the most favorable manner possible. That is, through the consideration of passenger’s preferences across the offered minimum-cost operational strategies. The methodology is firstly applied to a sample problem, and secondly - to a real life, case study of multiple bus lines in Dalian, China. The obtained results demonstrate that the effectiveness of combined strategies is higher than that of any other types of strategies. In the real life bus lines case, it saves four vehicles in comparison with the full route operation scenario. The anticipated number of vehicles is further reduced by three, by the introduction of deadheading-trip strategy resulting in a greater public transit system efficiency.
Analytical Modeling of Semi-Flexible Transit System: Effect of Service Delivery Type on Operating Cost Efficiency and Critical Passenger Demand
Sushreeta Mishra, University of ManitobaShow Abstract View Presentation
Babak Mehran, University of Manitoba
A methodology is proposed to compare the operational efficiency of a fixed route bus transit (FRBT) and a semi-flexible transit (SFT) system for a low demand existing bus route. The service delivery systems adopted for SFT are Contract-Out Taxi (COT), and In-House Transit (IHT). Rigorous and approximate operating cost models are developed to define the critical passenger demand at which it is advisable to “switch” from one transit system to another to minimize the operating cost. The approximate solutions are obtained by applying the method of continuous approximation; typically, intended to provide handy but powerful decision support tool to solve complex transit decision problems. In essence, the study findings would be helpful for decision-makers in identifying the passenger demand where taxis or in-house transit could replace or supplement each other to optimize the operation of public transit along a low demand bus route. The application of the proposed methodology is demonstrated for a low demand existing bus route in the City of Regina, Canada.
A Compatibility-Based Approach for Routing and Scheduling the Demand Responsive Connector
Yunxue Lu, Southeast UniversityShow Abstract
Hao Wang, Southeast University
Li Wenquan, 东南大学
The routing and scheduling problem of the Demand Responsive Connector (DRC) has been generally included into the field of Vehicle Routing Problem. Routing algorithms are proposed without fully considering the characteristics of DRC, so their performances in DRC operation are discounted. This study concentrates on building feeder plans to serve people traveling to and from transit hub. The interaction of passengers taking the same bus is deeply analyzed to clarify the time constraints faced by feeder buses. For the purpose of reducing operating cost and passenger inconvenience, this paper proposes a compatibility-based heuristic algorithm to solve the vehicle routing and scheduling problem of DRC. A passenger subset termed as the minimum incompatible set (MICS) is specially constructed to initialize feeder routes, and the schedule optimization procedure is designed to reduce number of vehicles. Several factors critical to the real-word operation are considered in our study, such as double time window constraints (both at the beginning and end of passenger travels), the flexibility of feeder plan, and the maximum number of vehicles. The proposed algorithm is validated on numerical instances of 400, 600, 800 and 1000 passengers. Simulation results show that the algorithm can effectively reduce the number of vehicles with acceptable increase of passenger inconvenience, and the algorithm efficiency is also considerably improved. In addition, sensitivity analysis is conducted to help service providers to evaluate the trade-off between the operation cost and level of service. Keywords : Demand Responsive Connector, Routing and Scheduling, Compatibility-based Approach, Insertion Heuristic Algorithm
The Application and Performance of LSSVM for Short-Term Passenger Flow Forecasting with Diverse Input
Dongya Li, Southeast UniversityShow Abstract
Xuedong Hua, Southeast University
Wei Wang, Southeast Hospital
Pandi Wang, Southwest Jiaotong University
This paper focuses on the short-term passenger flow forecasting using LSSVM. With several kinds of input (HD, OF, and HD-OF), we find out the optimal operating conditions of LSSVM and evaluate its performance compared with other three machine learning algorithms (BP, RBF and SVM). The comparison of different input indicates how input affecting the forecasting results, and the comparison of different methods indicates how LSSVM performing in some cases with other methods as contrasts. A bus line in Guangzhou was selected as a case study. The simulating analyses identify some worthwhile findings for traffic operating, shown as following: a) even with small quantities of information as input, LSSVM can perform well with high accuracy and robustness (MAE is 109 and RMSE is 136); b) too many input variables of LSSVM are not suggested, since too many variables may lead to poor performance; c) LSSVM has the shortest time consuming compared with BP, RBF and SVM, which is significantly important for short-term forecasting to support the traffic operation decision; d) RBF always performs well but with much long running time, while LSSVM has obviously shorter running time, and performs well if the input variables are not excessive. These findings are very helpful for operators to choose the most applicable method for passenger flow forecasting in different cases. Moreover, the research results of this paper provide some references for the further studies on short-term passenger flow forecasting using LSSVM.
An Inventory of Bus Stop Amenities Guidelines at U.S. Transit Agencies
Samuel Jensen, University of ArizonaShow Abstract
Arlie Adkins, University of Arizona
Keith Bartholomew, University of Utah
Ja Kim, University of Utah
Bus stops are one of the few capital investments visible to the public at many large transit agencies and form an important part of the rider experience. While the importance of quality bus stop amenities has been well established, the amenities provided can vary both between and within agencies due to budgetary constraints, local jurisdiction and neighborhood involvement, and the age of the stop. This paper examines the guidelines used by large transit agencies in the United States in deciding where to place bus stop amenities and identifies common criteria used in this decision-making process. The study shows that bus stop amenity guidelines are a common tool at used at large transit agencies and that a number of criteria are named in the guidelines, often varying based on local level concerns – the most common being a ridership threshold. Finally, this paper identifies potential best practices in bus stop amenity placement policies.
A POPMUSIC Algorithm for Customized Bus Network Design
Yining Hu, Kyoto UniversityShow Abstract View Presentation
Nobuhiro UNO, Kyoto University
Customized bus (CB) is a kind of dynamic bus service which can assemble passengers with similar travel routes into one single bus based on changing demand. It allows passengers to send requests before a trip, provides seat for each of them and satisfies their requests of departure and arrival time. Usually the application of CB requires high level of demand between a pair of Origin and Destination (OD) with relatively long travel distance. However, sufficient demand for starting the service is not always generated for every CB line, leaving the passengers dependent on CB stranded. In this paper, we propose a new form of CB that combines small demands between different short-distance ODs to create a longer CB route, extending the utilization of CB in low demand areas. A hybrid heuristic is used to solve small-sized routing problems of the new CB model, and a metaheuristic approach (POPMUSIC) is proposed to obtain the solution more efficiently in large-sized CB network. Computational results show that the proposed hybrid heuristic can obtain satisfying solutions with high efficiency in small-sized problems, while in large-sized problems both the computational efficiency and solution quality decrease; and the proposed POPMUSIC can obtain better solution with higher efficiency in large-sized problems than the proposed hybrid heuristic if given a proper sub-problem size.
Multi-Objective Optimal Allocation of Wireless Bus Charging Stations Considering Costs and the Environmental Impact
Yuval Hadas, Bar Ilan UniversityShow Abstract
Oren Nahum, Bar-Ilan University
In recent years, due to environmental concerns, there is an increasing need to develop alternative solutions to traditional energy sources (e.g., fossil fuels). Since transportation is a major fossil fuel consumer, development of electric vehicles (EVs), especially electrical buses, reduces fossil fuels uses, and, therefore, provide a better living environment. The aim of the work is the development of a system-wide wireless charging stations optimal allocation model. The main advantages of wireless charging are the need for a much smaller battery, and the contactless charging, both static and dynamic (in-motion). Unlike previous works that dealt with the allocation of wireless charging stations along a single route, or for a given network, the suggested model is a multi-objective model that selects the location for the charging stations while minimizing the costs (charging stations installation and batteries), maximize the number of routes that can be operated by wireless charging buses, and maximizes the environmental impact. The problem is formulated as a multi-objective non-linear optimization model. An efficient genetic algorithm is introduced for solving the problem. A test case is used to demonstrate the model, so the decision maker is provided with a solution set from which the best fit solution can be selected considering costs, the number of routes and environmental impact
Flexible Bus Dispatching System with Modular and Fully Automated Bus Units
Igor Dakic, Eidgenossische Technische Hochschule ZurichShow Abstract View Presentation
Kaidi Yang, Stanford University
Monica Menendez, New York University, Abu Dhabi
Joseph Chow, New York University Tandon School of Engineering
This study proposes a novel flexible bus dispatching system where a fleet of fully automated modular bus units, together with conventional buses, serves the passenger demand. These modular bus units can either operate individually or combined forming larger units with a higher passenger capacity. This provides enormous flexibility to manage the service frequencies and vehicle allocation, reducing thereby the operational costs and improving passenger mobility. In this paper, we develop an optimization model to determine the optimal composition of bus units and the optimal service frequency at which the buses (both conventional and modular) are dispatched across each bus line. We explicitly account for the dynamics of traffic congestion at the network level based on the recently proposed three-dimensional macroscopic fundamental diagram (3D-MFD). Results show improvements in the total system cost by adapting the number of combined modular bus units and their dispatched frequencies to the evolution of both the car and the passenger demand. To the best of our knowledge, this is the first application of the 3D-MFD and modular bus units for the frequency setting problem in the domain of bus operations.
Investigating the Effects of User Behavior and Trip Length Patterns on the Optimal Bus Network Design
Igor Dakic, Eidgenossische Technische Hochschule ZurichShow Abstract
Ludovic Leclercq, Université Gustave Eiffel
Monica Menendez, New York University, Abu Dhabi
Public transport systems are often seen as a key component of a sustainable urban development. They are regarded as a public service that should provide mobility accessible to all citizens in an urban area. Among the different aspects of public transport (in particular the bus system), that one can consider to be of influence for the user perception of the level of service, its network structure and operating regime stand out, i.e. how the bus lines that compose the system are arranged atop the street network and how the service frequency is adjusted to meet urban mobility patterns. This problem is known as the bus network design problem, and it has been the subject of several studies. Most scientific literature has determined the optimal bus network design and the service frequency based on one particular trip length pattern and type of user behavior. Potential effects of other types of user behavior and trip lengths patterns on the user and/or operator cost function have not been investigated whatsoever. In this study we aim to close this gap by providing a general framework, capable of accounting for any trip length pattern and two types of user behavior. The proposed framework is used to investigate how these two features affect the optimal network configuration per cardinal direction. Numerical analysis shows that both, the user behavior and the trip length patterns, have significant effects on the operator and the user cost function.
The Influence of Walkability Around Feeder Bus Stops on Rapid-Transit Station Boardings: The Case of Los Angeles Multimodal Transit System
Luis Ramos-Santiago, Clemson UniversityShow Abstract View Presentation
Numerous studies have focused on the potential influence of land-use and built-environment features (LU.BE) as key determinants or mediators of rapid-transit patronage. Some find statistically significant associations, yet their effects are relatively weak as compared to socio-economic, transit network connectivity, regional accessibility, and transit service level factors. Most, if not all recent LU.BE / transit-patronage studies, focus their attention on stations’ pedestrian service areas yet ignore areas surrounding feeder bus stops. Large, more dispersed, poly-centric urban agglomerations in the United States (e.g. Los Angeles, Miami, and Atlanta) register higher bus access share (>30%) as compared to older and more center-focused cities (e.g. Chicago, Boston, Philadelphia). As many cities in the United States continue to grow and decentralize, bus access to rapid-transit stations could become more relevant to multimodal transit effectiveness and efficiency. This study examined the Los Angeles metropolitan rapid-transit system to understand how LU.BE attributes around feeder bus-stops might affect boardings at rapid-transit stations. A multilevel generalized structural equation model (GSEM) is implemented and fitted with survey-based and GIS-based socio-economic, transit service quality, and proprietary LU.BE data in explaining the number of person-trips from feeder bus-stops to rapid-transit stations. Results strongly suggest that LU.BE attributes around feeder bus-stops, captured as a multi-dimensional walkability index (WalkScore®), as well as distance-based subcomponents of the same index, register statistically significant associations with higher boarding at feeder bus-stops. However, their impact relative to the number of jobs, automobile availability, license tenure, and particularly feeder route service quality is small.
Where to First Electrify Bus Transit Routes: A Case Study for Austin, Texas
Jugal Amodwala, University of Texas, AustinShow Abstract
Kara Kockelman, University of Texas, Austin
Based on a holistic literature review, battery electric buses (BEBs) are the best alternative to harmful diesel buses that a majority of cities use. Hybrid buses are often touted as the stepping stone from diesel to electric, but given the 12-to-18-year lifespan of a public transit bus and the level of maturity that electric bus technology has reached, hybrids are no longer needed. While hydrogen fuel cell buses have the benefit of long range and low net emissions, that technology remains prohibitively expensive and unreliable for long term usage. When compared to on-route BEB charging, overnight or depot-based BEV charging is more feasible and straightforward to implement, resulting in more U.S. Grants to subsidize higher initial costs plus legal agreements that reduce risk for transit agencies transitioning to BEB systems. The City of Austin, Texas’ transit agency, Capital Metro, has announced a rough guideline as to how the city will implement overnight BEBs. Out of the three route types (MetroBus, MetroExpress, and MetroRapid) currently offered in Austin, this study finds that MetroExpress routes for a BEB pilot program to be most reasonable. Metro Routes 10, 982, and 801 were analyzed using GTFS and manually collected GPS data to illuminate how to determine which routes are most cost-effective to electrify. Due to MetroExpress routes having fewer stops and shorter lengths, it is evaluated here as a good option for initial Austin-area BEB implementation, and all seven MetroExpress routes were analyzed.
Preferences for Autonomous Public Transit Vehicles: Insights from a Survey and a Public Engagement Event in Singapore
Samuel Chng, Singapore University of Technology and DesignShow Abstract
Sabreena Anowar, University of Missouri, Columbia
Lynette Cheah, Singapore University of Technology and Design
One of the key challenges to Autonomous Vehicle (AV) adoption and realization of the benefits is how the general public perceives the technology and how willing or unwilling they are to accept it. However, research focusing on understanding user’s perception and acceptance is still incipient, despite the recent proliferation of literature on autonomous mobility. In our study, we aim to address the gap in the literature using Singapore as our case study. Instead of relying on a single source of data, we draw and compare results from two different data sources collected using different methodologies (a survey and a public engagement event) to explore preferences about the impending introduction of AVs in public transport. Top three implementation preferences found both in the survey and debate samples include: 1) having clearer clarification of liability when an accident occurs, 2) implementing public education campaign on how the autonomous technology works and what the possibilities and limitations are, and 3) having AVs tested on the roads by the authorities first before making available to the public. We also found that trust in AVs exist; this trust can be further strengthened through addressing the concerns and uncertainty surrounding AV technology as identified in the paper.
A Bi-Level Optimization Model for Exclusive Bus Lane Design to Improve Passenger Travel Time Reliability
Weibin Kou, Beijing Jiaotong UniversityShow Abstract
Xumei Chen, Beijing Jiaotong University
Lei Yu, Texas Southern University
This paper proposes a reliable and practical method to design exclusive bus lanes (EBL) in order to improve the passenger travel time reliability. First, the quantitive indicator of passenger travel time relibaility composed of the reliability preference parameter and flucation parameter is determined. Then, a reliability-based bi-level optimizaiton model of EBL is established. The upper model considers the tradeoff among bus and private car passengers travel time cost as well as EBL construction cost. The lower model quantifies the impact of travel time reliability on passenger travel choices and analyzes dynamic changes of transit network passenger distribution under the background of EBL. A two-stage solution algorithm on the basis of feedback-loop iteration is designed. This paper conducts the statistical modeling and fitting analysis by actual bus trajectory data, which is from the express buses lines operating on EBL of the Third Ring Road in Beijing during peak hours. Calculation results have shown that the unit expectation and standard deviations of travel time are 1.59 min/km and 1.15 min/km, respectively. The reliability cost is to reflect the improvement effect of transit passenger travel time reliability. The results indicate that the passenger travel time reliability for express lines can be improved by 8.2% and 5.5% with the proposed EBL optimization scheme in the numerical test and case study, respectively. The optimization results also show the indirect effects on improvement of the travel time reliability of branch bus lines due to the shifting of passenger distribution among express lines and branch lines.
Optimal Design for Hub-and-Spoke Bus Transit Network with Heterogeneous Demand
Lijun Yu, South China University of TechnologyShow Abstract
Yizhou Zhu, South China University of Technology
Xue Li, South China University of Technology
Yipeng Peng, HNTB Corporation
In this paper a multi-objective nonlinear mixed integer optimization model for Hub-and-Spoke bus transit route network design problem is being proposed, aiming at the minimum total weight cost of users and suppliers considering the high heterogeneous passengers’ travel demand and operating constraints. The decision variables are routes service frequencies. To solve this model, an efficient meta-heuristic method based on improved cuckoo algorithm is proposed. This method has been implemented as a computer program and applied to a Hub-and-Spoke bus transit route network optimal design problem. The results of the numerical analysis show that by selecting the service frequency of all potential candidate routes, using this improved algorithm will yield a near-optimal bus transit route network. On the other hand, by maintaining the existing efficient bus routes, it provides a better solution for the redesign of the network with high heterogeneous travel demands. By comparing the results and the CPU time of the three algorithms, this paper demonstrated that improved CS is reliable and more efficient than GA and CS. The general analytic framework proposed in this paper provides an effective tool for the optimal design of a large-scale hub-and-spoke bus transit route network in practice. Keywords : Bus transit network design; Hub-and-spoke network; Heterogeneous demand; CS Algorithm; Frequency
Improving Transfer Time Optimization Modeling for Public Transit Systems
Zahra Ansarilari, University of TorontoShow Abstract View Presentation
Mahmood Mahmoodi Nesheli, University of Toronto
Merve Bodur, University of Toronto
Amer Shalaby, University of Toronto
Siva Srikukenthiran, University of Toronto
In transfer-based transit networks, timetables should be fully synchronized to avoid long transfer waiting times which might discourage people from using public transit. Transfer time optimization has been investigated for decades with the objective of either minimizing total transfer waiting times or maximizing the total number of successful transfers. There is an essential need to compare the results of both types of approaches at the network level as well as for individual transfer nodes under various headway combinations and demand distributions across the network. Additionally, most existing models make the strong assumption of unlimited vehicle capacity available for transferring passengers to board, which might lead to unreliable results if bus services are operating at demand levels near capacity. This study aims at comparing different modelling approaches and providing useful insights into how agencies can design their services to improve transfers. Four different modelling approaches are tested and analyzed through a numerical example of a network consisting of three nodes with different headway values. The results show that the consideration of demand in the objective function as well as vehicle capacity constraints have notable effects on model outcomes. Also, it was found that transfer node characteristics, such as its location, demand distribution, or headway combination constitute a major factor that agencies should take into account in the choice of the most appropriate transfer optimization model.
Reliable Schedule Coordination for a Bus Transit Corridor: A Shanghai Case
Xiongfei Lai, Tongji UniversityShow Abstract
Jing Teng, Tongji University
Paul Schonfeld, University of Maryland, College Park
Lu Ling, Purdue University
Providing convenient transit services at reasonable cost is important for transit agencies. Timed transfers that schedule vehicles from various routes to arrive at some transfer stations simultaneously (or nearly so) can significantly reduce wait times in transit networks, while stochastic passenger flows and complex operating environments may reduce this improvement. Although transit priority methods have been applied in some high-density cities, operating delays may cause priority failures. This paper proposes a reliable schedule coordination method for a bus transit corridor, which analyzes link travel time, passenger loading delay and priority signal intersection delay. It minimizes the total cost at different passenger flow rates, whether or not transit priority is provided. The data accuracy and result validity are improved with automatically collected data from multiple bus routes in a corridor. The Yanan Road transit corridor in Shanghai is used as a case study. The results show that the proposed method can reduce the total cost at average passenger flows. Even at extremely large flows the proposed method is reliable since it reduces passenger loading delay and signal priority intersection delay.
Transit Route and Frequency Design for Environmental Sustainability
Mei Zhu, Southeast UniversityShow Abstract
Zhu Mei, 东南大学
Li Wenquan, 东南大学
Chen Qian, 东南大学
This paper proposes a bi-level transit network design problem where the transit routes and frequency settings are determined simultaneously with the purpose of reducing energy consumption. The upper level problem is formulated as a mixed integer linear program with the objective of minimizing the carbon emission and constraints of reasonable travel cost and operation cost. The lower level problem is a transit assignment problem with capacity constraints. A genetic algorithm is developed to solve the bi-level transit network design problem. An instance of Changxing county in China was introduced to demonstrate the performance of this genetic algorithm to this bi-level formulation. The result shows that total carbon emission has been decreased significantly.