Direct transmission in widespread wireless sensor networks, where the cluster heads (CHs) and the base station (BS) are far from each other, is considered a critical factor because of its influence on network efficiency in terms of power consumption and lifetime. This paper focuses on the discovery of an optimal multi-hop path between a source (CH) and a destination (BS) to reduce power consumption, which shall maximize network lifetimes, by proposing a new Optimal Multi-hop Path Finding Method (OMPFM). A genetic algorithm is utilized in the proposed method to find an optimal path by proposing a new fitness function. Moreover, two pre-processes are proposed to select the CHs and increase the efficiency of the genetic algorithm in terms of the execution time and the quality of the chromosomes. The evaluation of the proposed method is conducted in MATLAB simulator and compared with other related methods. Experimental results show that the proposed method is better than LEACH, GCA, EAERP, GAECH and HiTSeC by 35%, 34%, 26%,19% and 50%, respectively, in terms of the first node die metric, and by 100%, 99%, 87%, 78% and 50%, respectively, in terms of the last node die metric.