---

The exponential growth of the flow of goods and passengers, fragility of certain products and the need for the optimization of transport costs impose on carriers to use more and more multimodal transport. In addition, the need for intermodal transport policy has been strongly driven by environmental concerns and to benefit from the combination of different modes of transport to cope with the increased economic competition. This research is mainly concerned with the Intermodal Terminal Location Problem introduced recently in scientific literature which consists to determine a set of potential sites to open and how to route requests to a set of customers through the network while minimizing the total cost of transportation. We begin by presenting a description of the problem. Then, we present a mathematical formulation of the problem and discuss the sense of its constraints. The objective function to minimize is the sum of road costs and railroad combined transportation costs. As the Intermodal Terminal Location Problemproblem is NP-hard, we propose an efficient real coded genetic algorithm for solving the problem. Our solutions are compared to CPLEX and also to the heuristics reported in the literature. Numerical results show that our approach outperforms the other approaches.

---

Due to the considerable growth in the worldwide container transportation, optimization of container terminal operations is becoming highly needed to rationalize the use of logistics resources. For this reason, we focus our study on the Quay Crane Scheduling Problem (QCSP), which is a core task of managing maritime container terminals. From this planning problem arise two decisions to be made: The first one concerns tasks assignment to quay crane and the second one consists of finding the handling sequence of tasks such that the turnaround time of cargo vessels is minimized. In this paper, we provide a mixed-integer programming (MIP) model that takes into account non-crossing constraints, safety margin constraints and precedence constraints. The QCSP has been shown NP-complete, thus, we used the Ant Colony Optimization (ACO), a probabilistic technique inspired from ants’ behaviour, to find a feasible solution of such problem. The results obtained from the computational experiments indicate that the proposed method is able to produce good results while reducing the computational time.