The goal of radiotherapy treatment planning (RTP) is to deliver a high dose to target volumes (TV) while causing minimal damage to the surrounding healthy tissues. Many approaches have been proposed to optimize the dose distribution mainly in TV. In this paper, we propose a novel RTP optimization approach based on the use of genetic algorithms (GAs) to determine the beam weights and orientation. GA is based on the Darwinian principles of the survival-of-the-fittest mechanism. Through genetic operators like cross over and mutation, the new generation composed of many chromosomes will be better than the old one. We proposed a heuristic operation called 'sudden death' to speed up convergence. We present an example demonstrating that the application of genetic algorithms to three-dimensional (3D) RTP is feasible.