Application of the Artificial Neural Network with Multithreading Within an Inventory Model Under Uncertainty and Inflation

The solutions to real-life problems are challenging to find out in the exact form as the dimensions of the problems are significant. A multi-period multi-product inventory model is tested in this study through an artificial neural network for experiencing an uncertain environment. Whenever obstacles present in every period, a neural network with multithreading is one of the optimization procedures to find the best optimal solution with an inflation and time value of money. A fuzzy approach is used here to deal with the uncertainty, and the total cost of the system is calculated using the bi-objective constraint objective function. The first objective is to find the minimum cost of the system with the optimum space, which is the second objective. The solutions of the mathematical model are obtained by generating multiple threads that every thread is a possible solution. The numerical experiment chooses the best fit from the multiple solutions. An illustrative comparative study with the existing methodologies is provided. Results show that the proposed approach is the best for cost optimization and time minimization. It is found that the neural network with multithreading converges over other algorithms and the results obtain within less than 30% time of the existing algorithm.