In the practice of scheduling of construction projects, there is a great variety of methods and procedures that need to be selected at each construction process during project. Accordingly, it is important to consider the different modes that may be selected for an activity in the scheduling of construction projects. In this study, first, we mathematically formulate the resource-constrained project scheduling problem with multiple modes while minimizing the total project time as the objective function. Following, we propose a new random key-based genetic algorithm approach which includes the mode reduction procedures to solve this NP-hard optimization problem. Finally, in order to evaluate the performance of our method, we are scheduled in the close future to implement the proposed approach on some standard project instances as the computational experiment and analyze these experimental results comparing with the bi-population-based genetic algorithm by Peteghem and Vanhoucke .