eprintid: 714
rev_number: 8
eprint_status: archive
userid: 17
dir: disk0/00/00/07/14
datestamp: 2017-08-15 22:36:51
lastmod: 2017-08-15 22:36:51
status_changed: 2017-08-15 22:36:51
type: report
metadata_visibility: show
creators_name: Gavranovic, H.
creators_name: Stojančević, T.
creators_name: Kresoja, M.
creators_name: Charalambides, M.
creators_name: Miidla, P.
creators_name: Lynott, G.
creators_name: Mallinson, A.
creators_name: Kyriakides, I.
corp_creators: Andreas Athanasiades
title: Optimizing Bus Routes in Nicosia
ispublished: pub
subjects: transport
studygroups: esgi125
companyname: OSEL (Transportation Organization of Nicosia District), Cyprus
full_text_status: public
abstract: In this report the conclusions by the team of experts that took the ”Trans- portation Organization of the Nicosia District (OSEL)” challenge are provided. The challenge was to identify ways to improve efficiency of the bus network and increase the utilization of the network by the public. A thorough analysis of the various factors that affect bus route planning is provided. Moreover, a demonstration of a simplified route planning problem is described in order to motivate further work on this topic. Recommendations are provided to the company on the way to move forward towards solving the problem of creating a bus network with increased efficiency and grater appeal to the public. Specific recommendations include the collection of a larger amounts of data that can be used to generate models used in simulation analysis. Data include demographic data on bus usage and bus usage preferences by the public. In addition, data is required on bus travel times, walking distance to the nearest bus stop by the commuter, and traffic data.
problem_statement: Nicosia, the capital of the Republic of Cyprus, is the only divided capital in Europe. This, combined with the narrow streets towards the centre of the city, a lack of data and a lack of popularity of public transport, makes it very difficult to create and run a successful bus network. The Transportation Organization of the Nicosia District (OSEL) hopes that if the bus network were more efficient it would be more appealing to the public by covering more population while reducing commute time.

There have been many academic papers published on the transit route network de- sign problem due to its relevancy. According to [1] the problem is usually approached by using practical guidelines and ad hoc procedures, analytical optimization models or meta-heuristic methods. If the most basic approach is used and the transport route network is designed using ad hoc procedures and practical guidelines then there is no guarantee that the resulting network is not far from optimal. Equally analytical optimization models are only suitable for idealized situations. This is be- cause the problem is NP hard, combinatorially complex and can be non-linear. On the other hand meta-heuristic approaches are able to find locally optimal solutions.

According to both [1] and [2] the heuristic approaches used to tackle the transit route design network problem vary according to the specific nature of the case at hand as most transit route design problems involve maximising or minimising some form of cost function that encapsulates the specific constraints, requirements and wishes of the case at hand. Common algorithms/methodologies include the genetic algorithm, local search, simulated annealing, random search, the nearest first search method and the tabu search method. This means that most approaches are highly dependent on the problem solvers local knowledge and intuition. Common problems that occur involve over simplification and incorrect assumptions. However when a meta-heuristic model works well the real life benefits can be great. For example [3] tells the story of a small town in the North of England with a population of 103,000 which recorded an estimated loss of 35,000 in 1965-6. Here the current bus network had essentially remained the same for the last thirty years but the main industrial and residential locations had changed. In fact most of the service frequencies were based on the ferry services which were carrying less than half the number of people than thirty years previously. It was estimated by the authors of [3] that their new bus network could make annual savings of about 36,500.
In this work we propose a general framework for improving both the efficiency of public transport and its appeal to the commuters. The framework includes an op- timization problem with constraints that takes into account bus network resources, distribution of population to be served and user preferences. The formulation of the framework has yielded the need for acquiring more data from the bus network such as accurate bus travel times, distances between bus stops etc, and public needs to be served such as population distribution and demographics and commuter preferences. In addition, an optimization methods is applied based on available population dis- tribution data to produce bus routes alternative to the ones currently used. The method applied has the goal to maximize people transfer rate in order to both satisfy serving more people and reducing travel times.
date: 2017-07-31
date_type: completed
related_url_url: https://www.esgi-cy.org/problems/
citation:   Gavranovic, H. and Stojančević, T. and Kresoja, M. and Charalambides, M. and Miidla, P. and Lynott, G. and Mallinson, A. and Kyriakides, I.  (2017) Optimizing Bus Routes in Nicosia.  [Study Group Report]     
document_url: http://miis.maths.ox.ac.uk/miis/714/1/ESGI125-Problem_4_OSEL-report-optimizing-bus-routes.pdf