Colloquia

Summary

Every year the Dutch national railway company NS has to make decisions on the size of their fleet of trains for the coming years. The team responsible for these decisions collects all data necessary including the predicted demand of trains per year and the expected supply of trains available per year. These data are entered in large spreadsheet models by experts of the team to calculate the predicted gap between demand and supply. This is done in a deterministic calculation spreadsheet and a stochastic calculation spreadsheet. These spreadsheets are found to be outdated, hard to use, and prone to errors. Therefore, the goal of the research is to design a new tool that combines the functions of the outdated spreadsheet in a way that meets the requirements of the company.

The resulting tool, named the MPP Tool 2.0, is a tool developed in AnyLogic in which data on demand and supply can be filled. When the tool is executed the results show what the gap is between the demand and supply for each year and for each segment. Subsequently, the user is able to create a scenario by adjusting the base scenario via a User Interface. A user could add or subtract train units, or change maintenance percentages for a certain train type in a certain year. These scenarios can be tested and when found sufficient, from within the same tool, a Monte Carlo simulation can be executed. The results of the Monte Carlo simulation can be viewed in a Power BI dashboard in such a way that a risk analysis of the scenario can be done.

The tool is verified and validated by the end-users and experts by performing sensitivity tests, a statistical test, and conducting interactive workshops where the participants were able to operate the tool. All the requirements were found to be met. As a result, the research goal to design a simulation tool to support long-term strategic decisions on train fleet asset management is achieved. With the tool, NS can evaluate scenarios faster and easier which will lead to a more supported final decision on the long-term strategic asset planning.