Automatisierte Tourenplanung in der Transportlogistik

In the future the rising transport costs will become more important in transport logistics. To keep environmental pollution and transport costs as low as possible, the vehicles of the company for distributing its products must be used extremely efficiently. To achieve this, a targeted automated route planning can be used in many cases of the company's internal disposition. In this thesis we consider the so-called Vehicle Routing Problem, a combinatorial optimization problem. Customers have to be supplied with products using delivery vehicles starting at one or more depots. This process has to be organized such that the resulting transportation costs are as low as possible. It has to be decided, which transport orders will be allocated to particular vehicles, and in which order the customers within a tour will be served. In this work it is shown by which mathematical algorithms this problem can be solved approximately. To do this, a JAVA based software was developed, which also supplies a comprehensive visualization of the problem instances. Furthermore, the origin of the map data is discussed, which are used for solving the VRPs. Concerning these data we have to work with a huge amount of informations, because the distances, velocities restrictions, one-way restrictions and driving bans have to be considered. Finally, VRPs are solved by various methods and the obtained solutions are compared with respect to their quality and runnning time.