Optimization of the urban line network using a mathematical programming approach

Abstract

The paper deals with the line-planning problem related to urban public transport.Given the transportation network in a city, the origin-destination matrix of travel demands, and the fl eet of available vehicles, the goal is to design the routes and frequencies of lines.The proposed solution method is a combination of the exact mathematical programming algorithm and a trip assignment procedure.The solution process consists of three stages: (i) initialization, (ii) designing the line network and setting the initial frequencies of selected lines, (iii) solution improvement.In the fi rst stage, an initial set of feasible lines is proposed.In the second stage, an optimal subset of candidate lines is selected and initial frequencies of lines are computed by solving a mathematical programming model of the line-planning problem.The problem is formulated as a multiple criteria optimization problem, where the criteria refl ect travelers' demand for a high quality service, operator's interest in an effective service, and the environmental impact of the vehicles.The solution of this problem specifi es the number of vehicles of the given mode and type operating on the lines.The lines which are not assigned a vehicle will not operate.The assigned number of vehicles determines the frequency of a given line.At the same time, the solution specifi es optimal passengers' routes in the line network.The third stage consists of an iterative process, which computes new line frequencies with regard to in-vehicle and waiting times, transfers, and passengers' behavior in a situation when they have multiple travel alternatives.The approach has been verifi ed using real transportation data of a middle-sized city in the Slovak Republic.The paper presents the results of the case study.

Cite this work

No public copy is stored for this record, so only the metadata is cited.

Stored only when both papers are in the corpus — 4 citations exist upstream.

Papers that cite the same references — the relational fallback while embeddings are still backfilling.