Algorithm based Transport Management System to optimize daily tour planning of trucks



Following a realignment of the company, the existing software was to be transformed into a fully comprehensive transport management system. The focus was placed on the daily route planning of trucks of medium-sized companies. Especially in such companies, route planning is often a complicated manual process with pen and paper.


The planning of tours is usually based on many different data such as distances, weather, free trucks, loading time windows, waiting times and many more. Depending on the complexity, the process can be very demanding on a human. Our approach was to support this cognitively demanding task with an algorithm. For this purpose, we created an intuitive user interface that not only displays the results of the algorithm, but also supports the user's manual post-processing.


As part of a small design team, I was responsible for creating and maintaining a design system, as well as the complete design cycle from research over ideation to support the developers for the implementation. I also had the opportunity to assist in product management to build the overall product vision and help implement changes in the team.


I've learned that an algorithm is a pretty good sales argument. In real usage its super hard to fulfill promises because there exist a lot of soft factors which are influencing the results. When results of algorithms are not satisfying you need to help the user to fix or adjust the results. For users that's a pretty hard tasks, because the algorithm may doesn't think in his patterns. It will be hard to fix issues which are not based on you own thought patterns. Besides that it was awesome to support in product management. So I had my first practical experiences in supporting a product owner.


Christian Xu, Raoul Korzuschek, Oliver Roscher, Corinna Riederer, Julian Thümer and the development team of Cargonexx