Optimizing a fuzzy inventory model using multithreading and artificial neural network under inflation

Abstract

The application of neural network is spread-out throughout the field of optimization. Within neural network diagram, the optimization with multithreading has gained a lot of popularity. This proposed research introduces the effect of neural network with multithreading in a basic inventory model under uncertain environment. This study develops multi-items in a multi-period inventory model under the effects of time value of money, uncertain demands and inflation. Fuzzy models are based on more realistic approach on data. Traditionally, for the sake of simplicity most of the time researchers considers demand as constant. But in reality, it is very difficult to determine the exact amount of future demand. Therefore, results, generated by constant terms, estimate of the total cost but may or may not be same as the value observed in reality. Therefore, in order to obtain a more realistic idea about the cost, this paper discusses about an uncertain demand and it is applied to a bi-objective multi-product inventory model. The total inventory cost is minimized with respect to minimum space required for storage for finding an optimal balanced range by using concepts of multithreading and artificial neural networks. Multithreading is used to generate multiple threads, where each thread is a possible solution. Each of these threads acts as a node in a neural network and then after a comparison procedure the final result set is generated with best fits. Some numerical examples are given to illustrate the results. The comparative study with other existing algorithm is conducted. Results show that this model gives reduced cost with reduced time.