Resource-Constrained Project Scheduling Problem Based On Brain Storming Optimization Algorithm

Resource-constrained project scheduling problem(RCPSP)is to allocate resources and to schedule the project activities reasonably under the condition of satisfying the precedence and the resource constraints,generating the optimal scheduling plan.This is of great significance for predicting the project cycle and guiding the smooth implementation of the project.Thus,the problem research in this area has important theoretical value and practical significance.This paper mainly studies the brain storm optimization(BSO)for solving RCPSP.Firstly,take the advantages of good global search ability and robustness of the brain storming optimization algorithm,brainstorming optimization algorithm is proposed to solve RCPSP based on the optimization framework of the algorithm to solve practical problems.The algorithm sets the upper search space and the lower communication discussion space.The search space generates a solution space according to the problem mathematical model of the optimization problem.The communication discussion space uses the "convergence" and"divergence" mechanisms to group the search space individuals to discuss.Inter-group discussion and intra-group discussion control the global search and local search capabilities of the algorithm separately.Multiple standard test functions were selected for experimental simulation,and compared with other algorithms commonly used to solve RCPSP.The comparison results show that the convergence performance of the BSO algorithm is overall better.Next,this paper use the BSO algorithm to solve RCPSP.The algorithm use random priority coding.The scheduling sequence generated based on the neighbor matrix that can show the precedence of project activities.The algorithm uses preempt-based mode resources assignment and the serial decoding method to generate an effective project scheduling plan.The standard PSPLIB example of classical RCPSP was used for numerical simulation,which verified the stability and effectiveness of the algorithm.Then,the adaptive selection strategy is introduced to improve the BSO algorithm for the defects,such as poor precision and slow convergence speed.The algorithm change the probability of individual selection between inter-group discussion and intra-group discussion by adjusting the size of the inertia weight to achieve the purpose of enhancing the global search performance at the early stage and the local fine search performance at the later stage,avoiding effectively trap in local optimum.The brain storming optimization algorithm based on adaptive selection strategy(AD-BSO)is proposed.The improved algorithm is applied to solve multi-mode resource-constrained project scheduling problems(MRCPSP).Two-dimension coding is used,and the various operations are defined.The effectiveness of the algorithm is verified by simulation.Lastly,this paper further studies the robust resource-constrained project scheduling problems(RRCPSP).With the proposition the robust measure index,the optimal model with robustness is developed,and according to optimal goals,the algorithm for the model is presented with the hierarchical optimization principle.The simulation results of standard examples show that the AD-B SO algorithm has better convergence performance and high efficiency.It is an effective algorithm for solving complex project scheduling problems.