An adequate placement of vehicles in depots is essential for them to leave it smoothly for their service trips. This work presents the methodology adopted in an automated placement software used daily by one of the largest Swiss public transport companies. The problem has several hierarchical objectives and many constraints. The most important one is to place as many vehicles as possible on parking lanes. The second hierarchical objective is to maximize the number of free parking lanes, while grouping the vehicles of same type on contiguous lanes. The third objective is to spread the leaving of the vehicles while respecting timetable and topological constraints. The proposed methodology is to decompose the problem into independent sub-problems treating one objective at a time. Since the solution of a sub-problem may over-constrain the sub-problem at a further stage, several sub-optimal solutions are kept to increase the probability that one of them leads to a feasible solution for a subsequent stage.