ABSTRACT
In this paper, the influence of the number and compilation of selected test sections on model validation results is assessed. Millions of model validation results of three different rail vehicles are calculated according to Validation Method 2 of EN 14363:2016 using hundreds of newly created test sections from the DynoTRAIN database. The presented investigation shows that the specified minimum of three test sections per test zone is a reasonable compromise between the effort and reliability of the validation, whereby it is recommended to use the same or similar number of test sections from all test zones. The evaluated validation method seems to be reliable. Once a model revealed to be not validated it is not worthwhile to select other test sections for validation expecting to change the overall validation result.
