Abstract
With the dynamic arrival of production orders and unforeseen changes in shop-floor conditions within a production system, production scheduling presents a challenge for manufacturing firms to ensure production demands are met with high productivity and low operating cost. Before a production schedule is generated to process the incoming production orders, production planning is performed. Given the large number of input parameters involved in production planning, it poses the challenge on how to systematically and accurately predict and evaluate the production performance. Hence, it is important to understand the interactions of the input parameters between production planning and scheduling. This is to ensure that the production planning and scheduling are coordinated and can be performed to achieve the optimal production performance such as minimizing cost effectively and efficiently. Digital twin presents an opportunity to mirror the real-time production status and analyze the input parameters affecting the production performance in smart manufacturing. In this paper, using the capabilities of real-time synchronization of production data in digital twin, we propose an approach to develop a surrogate model to predict the production performance using input parameters from a production plan. Multivariate adaptive regression spline (MARS) is applied to construct the surrogate model based on three categories of input parameters, such as current production system load, machine-based and product-based parameters. The effectiveness of the proposed MARS model is demonstrated using an industrial case study of a wafer fabrication production based on random sampling of varying numbers of training data set.