DUE DATE ASSIGNMENT USING NEURAL NETWORKS FOR STANDARD PRODUCTS IN SMALL BATCH AND MULTI ASSORTMENT MAKE-TO-ORDER COMPANY

The purpose of the article is to investigate artificial neural network ANN use in due-date assignment DDA for standard products in the real-life small batch and multi assortment make-to-order production company. The research was conducted with the historical data and resulted in the ANN aided DDA encoded into the company ERP system. The paper contains a comparison between different due-date assignment methods. The methods include company's own procedure, multi variable linear regression and multilayer feedforward neural network. The details of quantitative and qualitative input variables, an output variable, the neural network structure and its training are presented in the paper. DDA approach incorporating material deliveries uncertainty was applied to both the linear regression and the neural network methods. DDA using neural networks proved to outperform the other analysed methods.