Theproblemofplanningtheactivitiesofafleetofautonomous mobile robots in the context of performing a production plan is tackled in this paper. Three kinds of tasks are considered: the ones related to supplying workstations with the components or tools that are used in the operations, the ones related to the evacuation of empty containers or garbage collecting, and the latter ones that aim at moving semi-finished production from workstations to others. This paper shows that the plan- ning of these activities can be modeled homogeneously as a particu- lar pickup-and-delivery problem with time windows. To solve the latter problem, we propose a greedy heuristic as well as a mixed-integer linear programming approach. An illustration of the interest in the approach is provided in a production context, demonstrating its validity and high- lighting its advantages and limitations. A benchmark of the literature is also considered to show how our approach performs in the context of a job-shop problem with transportation constraints.