Public transport network optimisation in PTV Visum using selection hyper-heuristics
- PDF / 1,715,972 Bytes
- 34 Pages / 439.37 x 666.142 pts Page_size
- 39 Downloads / 167 Views
Public transport network optimisation in PTV Visum using selection hyper‑heuristics Philipp Heyken Soares1 · Leena Ahmed2 · Yong Mao3 · Christine L Mumford2 Accepted: 18 August 2020 © The Author(s) 2020
Abstract Despite the progress in the field of automatic public transport route optimisation in recent years, there exists a clear gap between the development of optimisation algorithms and their applications in real-world planning processes. In this study, we bridge this gap by developing an interface between the urban transit routing problem (UTRP) and the professional transport modelling software PTV Visum. The interface manages the differences in data requirements between the two worlds of research and allows the optimisation of public transport lines in Visum network models. This is demonstrated with the application of selection hyper-heuristics on two network models representing real-world urban areas. The optimisation objectives include the passengers’ average travel time and operators’ costs. Furthermore, we show how our approach can be combined with a mode choice model to optimise the use of public transport in relation to other modes. This feature is applied in a special optimisation experiment to reduce the number of private vehicles on a selected set of links in the network. The results demonstrate the successful implementation of our interface and the applied optimisation methods for a multi-modal public transport network. Keywords Route optimisation · PTV Visum · Selection hyper-heuristics * Philipp Heyken Soares [email protected] Leena Ahmed [email protected] Yong Mao [email protected] Christine L Mumford [email protected] 1
Laboratory of Urban Sustainability, Nottingham University, Nottingham NG7 2RD, UK
2
School of Computer Science and Informatics, Cardiff University, Queen’s Buildings, The Parade, Cardiff, UK
3
School of Physics and Astronomy, Nottingham University, Nottingham NG7 2RD, UK
13
Vol.:(0123456789)
P. Heyken Soares et al.
1 Introduction Public transportation systems form an essential part of the infrastructure in urban areas. Their efficiency is vital to many stakeholders and, therefore, they require precise design and planning. The generation of efficient route networks for public transport systems is known in many studies as the urban transit routing problem (UTRP).1 This problem has been addressed in the literature by a considerable number of studies that vary in their application on different instances, objectives and solution methodologies. Despite intensive research into the UTRP, there is a vast gap between the often purely academic studies and the application of their findings in real-world planning processes. The reasons for the gap have not been thoroughly researched so far (Walter 2010). However, one possible explanation could be the differences in data requirements between the algorithms used in UTRP research and the commonly used planning tools. Planning agencies usually base their decisions on simulations made with professio
Data Loading...
