Designing bus line plans for realistic cases - the Utrecht case study
Introduction
High-quality public transport systems are fundamental for the normal functioning of cities. They provide many benefits, but they also represent large costs, so it is very important to design and operate them efficiently. In research and practice, the problem of designing bus networks is commonly divided into five stages, namely: route network design, frequency setting, timetabling, fleet assignment and crew assignment (Ceder & Wilson, 1986). Given the complexity of these problems, they have been commonly solved sequentially. However, the current developments in computer capacity and solution techniques allow to combine some of these stages to solve them in an integrated manner. For example, one can solve the transit network design problem, also known as line planning, deciding at the same time the frequencies of the lines, giving as a result the Transit Network Design and Frequency Setting Problem (TNDFSP) (López-Ramos, 2014).
Despite the increasing interest in the TNDFSP, the problem is still very difficult to solve and therefore it is mostly only considered for small theoretical examples. There is also scarce literature on case studies and practical applications of optimization techniques for the TNDFSP. Indeed, in practice the line planning process is still mostly an iterative manual process, that relies on the experience of the transit planners and their knowledge of the areas under study. The lack of application of optimization techniques is in part due to the many assumptions that are done in theory and the long computing times required to solve large instances. A common misunderstanding when discussing the TNDFSP is that the computation time is not a pressing issue since bus lines are typically only modified every several years. However, solving the TNDFSP is only one step in a long and iterative planning process, so efficient algorithms that take a few hours to solve large instances are required for practical application.
The objective of this study is to bridge this gap between theory and practice and to propose an alternative bus line plan for the area of Utrecht, The Netherlands, using optimization techniques that take into account the service frequencies. In order to generate a meaningful solution for practical purposes, for Utrecht and more generally, several modifications are required to the models typically used in scientific studies, in order to get a realistic representation of reality. Furthermore, the network used to represent the city is much larger than the instances commonly solved in literature. By addressing these challenges, we are able to design a more efficient bus line plan for the city of Utrecht.
The contributions of this paper are:
- •
To show the benefits of decision support when designing a realistic line plan in a large-scale case study.
- •
To provide a detailed explanation on the construction of a realistic TNDFSP instance from real data.
- •
The development of an appropriate solution approach, a memetic algorithm (MA) in this case, that is able to address constraints from practice, such as a subset of terminal nodes, discrete frequencies, allowing circular lines and using a flexible number of lines.
- •
The proposition of new metrics to compare and evaluate different line plans for an instance from practice.
- •
The data for the case study is made available allowing future researchers to use this as a benchmark instance.
The rest of the paper is structured as follows: Section 2 presents an overview of the related literature. Section 3 describes the case study and how a realistic instance of the TNDFSP was generated. Section 4 describes the algorithm used to solve the problem and to generate alternative line plans. In Section 5, the computational results are presented and discussed. Finally, conclusions and future work are presented in Section 6.
Section snippets
Literature review
The Transit Network Design and Frequency Setting Problem (TNDFSP) has been widely studied in the past, as is reported in several comprehensive reviews (Farahani et al., 2013, Guihaire and Hao, 2008, Iliopoulou et al., 2019, Kepaptsoglou and Karlaftis, 2009, López-Ramos, 2014). The basic problem is defining the sequence of stops to be served by each of the lines that conform the transit network, and to assign a service frequency to each line. When the passenger travel time is considered
Case study
The aim of this study is to propose an alternative bus network for the city of Utrecht (The Netherlands) and to illustrate how large TNDFSPs can be addressed in practice. The study area considered includes the municipality of Utrecht and its immediate surroundings, with a population of approximately 640.000 inhabitants. The transit network of the city is composed by a bus network, one tram line and some regional train lines. For this study only the bus network is considered. However, although
Solution approach
The solution approach used in this paper is an extension of the Memetic Algorithm (MA) proposed in (Duran-Micco et al., 2020), where its capability to efficiently solve benchmark instances of the TNDFSP was demonstrated. For a thorough description of the algorithm and its steps readers are referred to that paper. The main structure and the modifications made for this study are discussed here. These modifications are related to the inclusion of circular lines, using only discrete frequencies and
Computational results
This section presents the results obtained by applying the model and solution technique described above to the instance of Utrecht. The first set of experiments intends to validate the proposed model as a realistic representation of the problem. Later, the way to compare different line plans is discussed, in order to compare the current bus network with two selected alternative solutions generated by our approach. Finally, several experiments are performed to visualize the impact of several
Conclusions and future work
This paper presents a method to design better line plans for realistic cases. This results in alternative bus line plans for the city of Utrecht, The Netherlands. The problem is modelled as a Transit Network Design and Frequency Setting Problem (TNDFSP) integrating a number of challenges from practice. The resulting problem is solved using a bi-objective memetic algorithm, that minimizes the average passenger travel time and the fleet size. The method to generate the TNDFSP instance from real
CRediT authorship contribution statement
Javier Durán-Micco: Conceptualization, Formal analysis, Funding acquisition, Investigation, Methodology, Software, Validation, Writing - original draft, Writing - review & editing. Marcel Kooten Niekerk: Conceptualization, Data curation, Methodology, Validation, Writing - review & editing. Pieter Vansteenwegen: Conceptualization, Funding acquisition, Methodology, Validation, Resources, Supervision, Writing - review & editing.
Declaration of Competing Interest
The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.
Acknowledgments
We thank the bus operator in Utrecht, Qbuzz, for their commitment during the research, providing the data and valuable comments about the results.
Funding
This work was supported by CONICYT (Chilean National Commission for Scientific and Technological Research) “Becas Chile” 72180156; FWO (Research Foundation Flanders) project G.0853.16N.
References (44)
- et al.
Solving urban transit route design problem using selection hyper-heuristics
European Journal of Operational Research
(2019) - et al.
Efficient transit network design and frequencies setting multi-objective optimization by alternating objective genetic algorithm
Transportation Research Part B: Methodological
(2015) - et al.
Transit-network design methodology for actual-size road networks
Transportation Research Part B: Methodological
(2011) - et al.
Transit network design with allocation of green vehicles: A genetic algorithm approach
Transportation Research Part C: Emerging Technologies
(2009) - et al.
A differential evolution for simultaneous transit network design and frequency setting problem
Expert Systems with Applications
(2018) - et al.
Modeling horizontal and vertical equity in the public transport design problem: A case study
Transportation Research Part A: Policy and Practice
(2019) - et al.
An adaptive neighborhood search metaheuristic for the integrated railway rapid transit network design and line planning problem
Computers & Operations Research
(2017) - et al.
Bus network design
Transportation Research Part B: Methodological
(1986) - et al.
Transit network design: A procedure and an application to a large urban area
Transportation Research Part C: Emerging Technologies
(2012) - et al.
Considering emissions in the transit network design and frequency setting problem with a heterogeneous fleet
European Journal of Operational Research
(2020)
A review of urban transportation network design problems
European Journal of Operational Research
Line planning with user-optimal route choice
European Journal of Operational Research
Transit network design and scheduling: A global review
Transportation Research Part A: Policy and Practice
Transit OD matrix estimation using smartcard data: Recent developments and future research challenges
Transportation Research Part C: Emerging Technologies
A heuristic aided Stochastic Beam Search algorithm for solving the transit network design problem
Swarm and Evolutionary Computation
A multi-objective meta-heuristic approach for transit network design and frequency setting problem in a bus transit system
Computers & Industrial Engineering
Bus transit network design with uncertainties on the basis of a metro network: A two-step model framework
Transportation Research Part B: Methodological
Designing a large-scale public transport network using agent-based microsimulation
Transportation Research Part A: Policy and Practice
A route set construction algorithm for the transit network design problem
Computers & Operations Research
A simultaneous transit network design and frequency setting: Computing with bees
Expert Systems with Applications
Complete hierarchical multi-objective genetic algorithm for transit network design problem
Expert Systems with Applications
Sustainable urban transit network design
Transportation Research Part A: Policy and Practice
Cited by (5)
Speeding up the Passenger Assignment Problem in transit planning by the Direct Link Network representation
2024, Computers and Operations ResearchDesign of mixed fixed-flexible bus public transport networks by tracking the paths of on-demand vehicles
2024, Transportation Research Part C: Emerging TechnologiesTransit network design considering link capacities
2022, Transport PolicyCitation Excerpt :A lower number of lines allows higher frequencies and therefore reducing the waiting times with the same fleet size. More details about these comparisons between the algorithm results and the real bus network in Utrecht are provided in (Durán-Micco et al., 2022). Despite the good results and the improvements made to the model compared to previous ones, there are still some aspects that limit the direct application of the algorithm in practice.
Line Planning for Different Demand Periods
2023, Operations Research Forum
- 1
Present address: MOBI Research Center, Vrije Universiteit Brussel (VUB), Pleinlaan 2, Brussels 1050, Belgium.