Application of Neuro-Fuzzy Logic on the Newsvendor Inventory Model
Alezander Mikhail O. Galindo, Jullian Dominic D. Ducut, Elmer P. Dadios, Ira C. Valenzuela, Robert Kerwin C. Billones · 2021 IEEE 13th International Conference on Humanoid, Nanotechnology, Information Technology, Communication and Control, Environment, and Management (HNICEM) · 2021
Challenges arise in inventory management and control when there is uncertainty involved. In the newspaper inventory model where demand levels are stochastic, ensuring the optimal quantity of inventory for each time period is crucial in optimizing profit. In this paper, neuro-fuzzy logic is utilized to evaluate this optimal quantity of inventory through the use of the software Matlab. The input variables are purchase cost, selling price, salvage value, mean demand, and standard deviation demand while the output variable is quantity. Iterations of the experiment were done with the objective of minimizing error and maximizing correlation values. The best iteration of the experiment achieved a training error value of 0.0012437 with a coefficient determination value of ~1 for both the training data and testing data.