This paper treats of a generalization of the Traveling Salesman Problem (TSP) called Multi-commodity one-to-one Pickup-and-Delivery Traveling Salesman Problem (m-PDTSP) in which cities corresponds to customers providing or requiring known amounts of m different objects, and the vehicle has a given upper-limit capacity. Each object has exactly one origin and one destination, and the vehicle must visit each customer exactly once. This justifies the words "one-to-one" and "traveling salesman problem" in the name of the problem, respectively. We introduce a Mixer Integer Linear Programming model for the m-PDTSP, discuss decomposition techniques and describe some strategies to solve the problem based on a branchand-cut procedure. Preliminary computational experiments on randomly generated euclidian instances are shown.