In this paper we present a new alternative for the routing problem in Wireless Mesh Networks (WMN) taking into account the quality of service (QoS). The actual problem includes multiple objectives with conflicts among them. Classical approximations optimize a single objective or QoS parameter, however they do not not take into account the conflicting nature of these parameters leading to suboptimal solutions. We take a new approach trying to improve the routing solutions and propose the use of a multi-objective evolutionary algorithms (MOEA), specifically the NSGA II algorithm. A mathematical model is introduced for this problem, which includes QoS parameters such as bandwidth, packet loss rates, delay and power consumption.