Calibration of Vilnius public transport model

Abstract

The application of information technologies preconditions the optimisation of performance of transport systems: improvement of the performance quality, safety and efficiency of the overall system, increase in capacity, reduction of the trip duration without high financial investment into construction of the new technical infrastructure. The transport modelling is the only economical and sufficiently reliable way to carry out a forward assessment of the impact of the innovations to be applied on the overall system. The network of Vilnius public transport was built up on the data of 2002, and the year 2002 saw the last comprehensive surveys. Building up the PT network the data was taken from the VIDAS database, created in 2002 while drafting the special plan for Vilnius transport infrastructure (tram) development. The morning rush hour, when the passenger flows are maximum, was chosen for the modelling. Calibration of Vilnius PT network was carried out after selection of three possible methods: TSys‐based, Headway‐based and Timetable‐based. In the timetable‐based model Logit, Kirchhoff, BoxCox and Lochse distribution factors are inter‐changed. Analysis of all coefficients received when modelling allows a conclusion that further modelling of the development of Vilnius public transport network should be based on Timetable‐based model choosing Kirchhoff or BocCox distribution laws, whereof conformity to the basic averages of coefficients of the 2002 survey is respectively 0,82 and 0,81. This would facilitate adopting solutions to the development of the public transport systems and would increase their reliability. The calculated coefficients revealed that TSys‐based methods were mostly removed from reality, and the average coefficient of failure to conform to the data of the 2002 survey is 0,24. This method did not give any data about the load on the stops, although the number of trips modelled with the help of this method was most proximate to the survey data, i.e. 0,69 %.

First Published Online: 27 Oct 2010

Keyword : Transport system, public transport, modelling, calibration of model