详情

Anv�ndning av Samgodsprognos i Bangods

项目名称：	Anv�ndning av Samgodsprognos i Bangods
摘要：	The freigth railway forecast from the Samgods model (the Swedish national freight transport model) is based on the development per product group in the model. These numbers are mapped to the freight train usage for the base year at the individual train levels, the reason being that a direct usage of the detailed Samgods model results are judged too uncertain. By using the actual base year traffic volumes more reasonable results are obtained at disaggregated levels. However, the technique with the mapping from rather aggregated growth numbers in Samgods to rather detailed train traffic in Bangods (the specific national railway freight model) lead to some problems. One of them is that the geographic distribution of the growth pattern vanishes, which can result in inconsistent and unrealistic results for parts of the railway system. Another problem is that the technique is not automized and it therefore requires unnecessarily long handling times and leads to high risk for mistakes. Thus there are both quality and user handling reasons for investigating the possibilities for improving the current method and/or modify the method in Samgods so that Bangods can be abandoned. One way of solving the problem is to investigate the possibilities for integrating statistics in Bangods and other input data with Samgods forecasts, and to carry out the forecasts for Bangods with some type of pivot-point method. A second option for solving the problem is to investigate the possibilities for changing the basic freight train handling in Samgods, and instead implement some form of time table handling. This would encompass a more dramatic change of the model's structure, in particular the railway capacity management module, and therefore a pre-study of its pros and cons would be suitable. The purpose of this project is to describe and evaluate the current method used with respect to objectives and requirements associated with forecasts from Trafikverket, and then to suggest a number of methods that may be used. In a second phase a suitable method will be suggested for translating the forecast from Samgods into Bangods, in such a manner that the spatial distribution across product groups is maintained. This involves a description of today's Bangods, objectives and requirement on forecast/policy-analyses, construction of examples regarding how the base year situation is presented, and how forecast/policy-analyses are related to the base year. Given this a number of proposals regarding how the Bangods requirements may be met by changes and adjustments of Samgods should be constructed. The purpose is to investigate a number of suggestions, and to choose the best suggestion for a possible implementation in the Samgods model. Ex 1. Conversion to handle an aggregate timetable for freight trains in Samgods. The idea here is to define a time table based handling of freight train, in which the transport supply is defined by the frequencies of the various freight train time tables (frequencies and routes). The method would require a definition of freight rail lines in the LOS-matrices, in combination with variables for defining the yearly usage frequencies of the lines. The frequencies should be aligned with the current RCM-model (Railway Capacity Management) in such a manner that all transport are covered, and all capacities are satisfied. Ex 2. Set up a freight train plan satisfying today's RCM-solution. The idea is to define a set of freight train lines covering the transport demand as presented in the RCM-solution. This is a rather straight forward (linear) set covering problem, but also comprising the actual link capacity flow constraints to be satisfied. The approach is quite similar to the one in Ex 1, but here we have not predefined any line in the LOS-data. Ex 3. OD-matrix for railway. From Samgods we can derive OD-matrices for different train transports describing the regional distribution pattern per product group. Through a mapping at suitable regional levels of today's freight train pattern to the base year situation, it ought to be possible to combine the model results for the base and the forecast years with the mapping to the actual base year transport distribution into a pivot-point method.
资金：	0.00
资助组织：	Trafikverket, Sweden
项目负责人：	Edwards, Henrik
执行机构：	Sweco AB, Sweden
开始时间：	20160511
预计完成日期：	20161224
实际结束时间：	0