A Mathematical modeling of project risk response according to primary, secondary, and residual risks under conditions of uncertainty using the Tabu search algorithm

Today, the uncertainty in the estimated time and cost of industrial projects is considered as an important challenge in the science of project management. If risk management is done regularly to identify potential problems and find their solution, it will easily complement other processes such as organizing, planning, budgeting, and cost control. In this regard, one of the most important and effective solutions to this problem is risk analysis (primary, secondary, and residual). In this research, an optimization model has been proposed to select actions to respond to risk for all three primary, secondary and residual risks. This research is quantitative. In building the model, the objective function is to minimize the total risk costs and the costs of reducing the time constraints applied to the relationship between two activities. Then, by determining a suitable reasonable time for the whole project and solving the model, an optimal set of actions to respond to the risks is determined. The basic innovation of this research, which does not cause the selection of a predetermined strategy, is the two limitations that examine the two dimensions of time and cost in response to primary and secondary risk. The results indicate that the initial risk costs have decreased. Also, by responding to the primary risk, secondary risks were created, which imposed a cost on the system, but this cost was reduced by assigning secondary strategies, as well as the optimal cost of activity failure with the sensitivity analysis that was done, the maximum amount of time that the project can end It was equal to 78 days and more than that makes the cost of failure of activities to be zero. Also, in this research, the genetic meta-heuristic algorithm and the Particle swarm algorithm were used to solve the problem in high dimensions, and the results showed that there is no difference in the results of these two algorithms.