An effective adaptive customization framework for small manufacturing plants using extreme gradient boosting-XGBoost and random forest ensemble learning algorithms in an Industry 4.0 environment

Abstract

The prevailing competitive manufacturing industry calls for continuous customer satisfaction for business sustainability. With the emergence of the Industry 4.0 paradigm, product customization, which gives customers the means to personalized products to meet their needs, has become a strategy to increase companies’ value. High-tech manufacturing firms are already diving deep into Industry 4.0 standards adopting innovative strategies to outstand themselves in the market, while small manufacturing plants are slow in embracing the digital transformation. The high cost involved in acquiring indispensable resources and the lack of expertise are some of the obstacles low-tech businesses face in endorsing this new paradigm. Inspired by the customization challenges of a small manufacturing plant, our main research contribution is to develop an effective adaptive customization platform that encodes the customization data history of a small manufacturing plant, from a static database, into a dynamic machine learning model to produce personalized products for their customers accurately. Our research improves customers’ experience by reducing the customization system’s complexity consisting of inputting several parameters to obtain personalized products to a single entry. The back-end of the platform uses powerful machine learning (ML) algorithms like extreme gradient boosting (XGBoost) and Random Forest (RF) ensemble learning to match a single customer input to the desired customized product category. Our research experiments convey insights, such as the best scenarios to use XGBoost over RF algorithms for regression problems with non-linear data. The excellent experimental results achieved on both machine learning models show the merits of this customization platform.