This article is based on a study of a container terminal of the Grand Port Maritime de Marseille (GPMM) in France. The goal is to optimize the allocation of the critical loading/discharging resources shared by vessels and inland transport modes (road, rail and waterways). After a short description of the context and a literature review, an integer linear programming model, based on a network flow representation of the problem, is ntroduced. This model being deterministic, a discrete event simulation model is proposed to validate the solutions determined by the mathematical programming model in a stochastic setting. Numerical experiments are reported and discussed for both models.