Time-Cost-Quality Trade Off in PERT Networks Using Neural Network and Evolutionary Algorithms

Time, cost and quality are important goals of any project. Todays, the demand of project stakeholders to reduce total project costs has increased simultaneously time reduction and augment in quality of the project. This leads researchers to develop models that add quality factor to previous models of cost-time balance. In this paper, a time-cost-quality trade -off model with three objective functions includes minimizing project termination, minimizing total project cost and maximizing total quality of activities in a multi-modal PERT network was investigated. After presenting the appropriate mathematical model, a decision was made on the basis of a test plan of possible levels for each variable. Then, using the simulation process, random values of decision variables and response variables were implemented at each time, and we developed a neural network model using artificial neural networks. To solve this model, since the problem is in the NP-hard category, two multi-objective meta-heuristic algorithms NSGA-II and MOPSO were used. To determine the performance of the proposed model, the problem was tested on a small, medium and large-scale PERT network. The parameters of these two meta-heuristic algorithms were adjusted by Taguchi method and the results were based on the parameters set showed that the NSGA-II algorithm performs better than the MOPSO algorithm.