Resource leveling is a commonly used planning technique to avoid extraordinary demands or excessive fluctuations in labor and plant resources required for a construction project, which could otherwise lead to a drop in productivity or an increase in production cost. In performing resource leveling, many planners or managers would adopt standard heuristic approaches to obtain an acceptable ... [Show full abstract] solution. This is because mathematical methods are only considered suitable for small to medium networks due to the combinatorial non-deterministic nature of the problem. The leveling of multiple resources is also dominated by the chosen heuristic methods, e.g. whether by leveling multiple resources in series or through combined resource leveling. Although heuristic approaches are easy to understand, they are problem-dependent. Hence, it is difficult to guarantee that an optimal solution can be achieved. This paper proposes a new Genetic Algorithms (GAs) enabled multiobjective technique for optimizing the multi-resource leveling problem. Adaptive weights are introduced so that each resource is assigned with a certain priority. This could effectively avoid the dominance of only one resource through the optimization process, as the adaptive weights can `learn' from the last generation and guide the genetic algorithms to balance the search pressure among different resources.