We have presented a two-level vehicle routing problem with time windows related to the logistic reorganisation of the hospital complex of Tours. The problem consists of determining vehicle routes and the size of the warehouse team within a large hospital, which leads to solving two interconnected vehicle routing problems. To solve this problem, we have proposed a genetic algorithm and a tabu search. The computational experiments based on real data sets prove that the best results are given by the genetic algorithm. These results will help hospital logistic centres in establishing a new organisation.