Punctuality based calibration of railway capacity models

The developing railway operational situation repeatedly increases the need for consolidated models for capacity calculation, strictly related to the level of punctuality achievable at the different levels of traffic density. This paper, presents a study, of an extended set of carrying capacity calculation methods, based on stochastic algorithms as well as simulation models. The aim is to provide a contribution in the key phase of calibration. In this phase, the aim is to; acquire relevance of both the level of perturbation accepted in the normal operation and the infrastructural and technological equipment of the concerned lines and nodes. Therefore, the research development has been organised in: 1) detailed analysis and tests of case studies for the commonly used calibration approaches; 2) investigation of selected case studies in terms of punctuality, by collecting an extended set of data on daily operation traffic including deviations between scheduled and real timetables; 3) application of calibration procedures to consolidated capacity models to achieve reference results; 4) systematic comparison between expected (according to classical calibration methods) and real (according to collected data on daily operation) punctuality levels; 5) proposal and test of adjustments to the classical calibration procedures; 6) identification and qualification in terms of potential benefits of the proposed calibration approach to increase representativeness, generality and adherence of results to the daily operation implications. The analyses have been supported by the use of the web portal railway capacity assessment tool RC-k, developed within the author’s research group.