Abstract: The Resource-Constrained Project Scheduling Problem (RCPSP) is a NP-hard problem in information engineering. The activities of a project have to be scheduled for satisfying all the precedence and resource constraints. We presented a heuristic algorithm (EDAS) to deal with this problem which employed an estimation of distribution algorithm (known as EDA) and improved the local search capacity with a simplex search. In this algorithm, the EDA firstly searched the solution space and generated activity lists to provide the initial population; then, the EDA selected the sample solutions to build a probability distribution model. The new individual was generated by sampling this model. The simplex search was used to enhance the local search capacity of the EDA. Compared with state-of-the-art algorithms available in the literature, we showed the effectiveness of this approach empirically on the standard benchmark problems of size J60 and J120 from PSPLIB.