We study the Hub Location-Routing Problem (HLRP) aiming at the design of an efficient freight transportation network for LTL (less-than-truck) transport between many origins (suppliers) and destinations (clients). Such a network relies on the location of consolidation hubs, the organization of routings for the collection/distribution of freight from suppliers to hubs and from hubs to clients, as well as direct shipment of consolidated freight between hubs. We focus on the Capacitated Single Allocation Hub Location-Routing Problem (CSAHLRP) in the case of distinct collection and delivery processes. We propose mixed integer linear programming (MILP) model and a Memetic Algorithm (MA) to solve the problem for minimizing the total cost of the network. Then we extend the model into a bi-objective model for minimizing both the total cost and CO2 emissions of transport. A modified memetic algorithm (MA) combined with a fast elitist non-dominated sorting genetic algorithm (NSGAII) is developed to capture the trade-off between minimizing total cost and CO2 emissions and exhibit approximations of the Pareto front. At last, a two step procedure is proposed to solve the single-objective HLRP based on a hub location problem (HLP) and two distinct vehicle routing problems for suppliers and clients allocated to each hub by the first step. Our single objective MILP model is decomposed accordingly and our MA is adapted to solve the HLRP following these two steps. A data base of instances of different sizes and characteristics has been developed in order to conduct extensive experiments for solving all these problems using the different solution techniques and validate our approaches.