Lead time prediction in a flow-shop environment with analytical and machine learning approaches

Abstract

Manufacturing lead time (LT) is often among the most important corporate performance indicators that companies wish to minimize in order to meet the customer expectations, by delivering the right products in the shortest possible time. Most production planning and scheduling methods rely on LTs, therefore, the efficiency of these methods is crucially affected by the accuracy of LT prediction. However, achieving high accuracy is often complicated, due to the complexity of the processes and high variety of products. In the paper, analytical and machine learning prediction techniques are analyzed and compared, focusing on a real flow-shop environment exposed to frequent changes and uncertainties resulted by the changing customer order stream. The digital data twin of the processes is applied to accurately predict the manufacturing LT of jobs, keeping the prediction models up-to-date via online connection with the manufacturing execution system, and frequent retraining of the models.