The objective of this paper is to present a GA-based optimal model for resource leveling problem, which overcomes the drawbacks of traditional resource leveling models. Based on the problem characteristics, the code scheme, genetic operators and algorithm structure of the proposed model are designed. By adopting several improved techniques, the GA-based model can determines the optimal solution to multiple resources leveling problems for a construction project. A case example is presented to demonstrate the performance of the GA-based model against heuristic methods