Transportation is often an important activity for companies, either it is their core or only supportive work. For all of these companies it is important to be able to fulfill their goal (to distribute goods, to visit all customers, etc…) at the lowest possible cost. Reducing the transportation cost can be done in different ways, e.g. finding shorter routes by reordering targets, reducing the number of vehicles used, or utilizing the capacity of vehicles better. Each transportation problem can have different conditions e.g. time windows, vehicles with different capacities, more depots from which the vehicles start their routes, etc. These tasks define a set of problems referred to as the Vehicle Routing Problems. Different subclasses of these problems target the various transportation tasks in industry. The practical importance of solving such problems makes this field researched in the past decades.
For companies with many vehicles and many targets, routing can be very difficult. Finding the optimal solution can be time consuming and has been successful only for instances less than 200 customers. However real-world problems can have much more customers. Therefore heuristics are used to find good enough solutions in acceptable time. In this thesis we examine the application of a heuristic approach, the Genetic Algorithm in the case of large VRPs.