During the implementation, construction projects usually encounter situation that considerably affects the project scheduling and cost. This study aims at using an improved version of the harmony search algorithm (HSA) to schedule resource constrained construction projects. This model is formulated as a global optimization problem. It will determine the duration of each activity to minimize the project total cost. The algorithm tries to find the best duration for each activity so that it leads to the total consumption of the corresponding resource. This may cause some activities to start simultaneously. The improvements have been made to increase the convergence rate and to lower the cost and shorten duration of the project. A numerical example has been proposed to evaluate the efficacy of the algorithm. This algorithm also addresses issues such as multi-resource, resource combination, and resource limit. Two scenarios have been considered for the test problem. The former scenario shows the project scheduled using the minimum duration list and the latter scenario schedules the project using the optimization algorithm. A comparison between the two scenarios shows the effectiveness of the proposed algorithm in decreasing the total cost and duration of the projects.