چكيده به لاتين
Railway geometry is constantly subject to change due to repetitive dynamic loads and unstable environmental conditions. Changes in the geometry of the line mainly affect the movement of the train, in order to solve the geometrical changes of the line and to fix the defects caused by these changes, the need to carry out maintenance operations is considered inevitable. The repair process includes many things, among which sublimation is very important. This task, which is done by the tamping machine, takes a major part of the costs, and if a plan is included and this operation is performed only in the places that need tamping, the costs can be significantly reduced and the quality of the road is optimal. upgraded In the railway system of Iran, according to the current operating method, developing a proportional model can reduce a huge part of the related costs (such as the cost of tamping, the cost of preparing the taming machine, the cost of blocking the line, and the fixed cost, etc.) and these costs are spent on other such as railway development. The problems related to the optimization of maintenance planning are very large and are classified as NP-hard problems, and it is not possible to provide a definitive optimal answer in a limited time. Meta-heuristic methods, due to their efficiency and flexibility in solving large and complex problems, provide solutions close to the definitive optimal answer to solve problems and provide the right answer in a reasonable time, which is a very suitable tool for solving maintenance issues. In this research, gray wolf meta-heuristic optimization methods and genetic algorithm have been used to optimize the maintenance cost, and to check the proposed model, geometric data of the line related to Salam-Mashhad axis have been used. By defining a model based on cost reduction and taking into account the constraints related to the recovery model after tamping, the maximum acceptable length balance and the useful life of the railway components are placed in the appropriate range and the costs are optimized. The results show that the gray wolf optimization algorithm has a reduction of about 6-7% in total maintenance costs. In a certain piece, the undercutting planning resulting from the gray wolf algorithm results in fewer undercutting times than the genetic algorithm. Also, the reduction of computing time has been observed in the gray wolf algorithm compared to the genetic algorithm.
