The French national railway company, SNCF, is responsible for the maintenance of its rolling stock. The most heavy maintenance operations, and thus longest ones (several weeks), are carried out in ten different dedicated workshops. Knowing rolling stock arrival dates in the maintenance workshop, we propose a MILP model to schedule maintenance operations respecting industrial constraints. Each rolling stock unit is considered as a project and operations as activities requiring a certain number of resources to be processed. Thus, we deal with the multi-skill resource-constrained multi-project scheduling problem (MSRCMPSP), as resources have multiple skills and several rolling stock units are maintained simultaneously. We focus on minimizing the sum of weighted tardiness of the projects and the sum of their weighted duration.