Estimating Lead Time Using Machine Learning Algorithms: A Case Study by a Textile Company

The inclusion of artificial intelligence in both our daily and business life has created a new alternative to meet many different needs. In this study, the production time of a textile product was estimated by using the data generated from a textile company serving in the field of ready-made clothing. After determining the factors affecting the production time within the company, this paper aim to develop a decision support system with the model that gives the best results among a support vector regression and three ensemble learning methods (Random Forest, Decision Tree Regressor, and Bagging) based on an end-goal to predict the production time. The results show that the bagging and random forest yielded highest R2≥ 0.84 and with a minimal predictive error when compared with other approaches. A demo was prepared for the decision support system that can be used to predict the production time of new orders using the random forest model by developing an interface for users.