Abstract

As a key strategy for achieving a circular economy, remanufacturing involves bringing end-of-use (EoU) products or cores back to a ‘like new’ condition, providing more affordable and sustainable alternatives to new products. Despite the potential for substantial resources and energy savings, the industry faces operational challenges. These challenges arise from uncertainties surrounding core quality and functionality, return times, process variation required to meet product specifications, and the end-of-use (EoU) product values, as well as their new life expectancy after extended use as a ‘market product’. While remanufacturing holds immense promise, its full potential can only be realized through concerted efforts towards resolving the inherent complexities and obstacles that impede its operations. Machine learning (ML) and data-driven models emerge as transformative tools to mitigate numerous challenges encountered by manufacturing industry. Recently, the integration of cutting-edge technologies, such as sensor-based product data acquisition and storage, data analytics, machine health management, artificial intelligence (AI)-driven scheduling, and human–robot collaboration (HRC), in remanufacturing procedures has received significant attention from remanufacturers and the circular economy community. These advanced computational technologies help remanufacturers to implement flexible operation scheduling, enhance quality control, and streamline workflows for EoU products. This study embarks on a comprehensive review and in-depth analysis of state-of-the-art algorithms across various facets of remanufacturing processes and operations. Additionally, it identifies key challenges to advancing remanufacturing practices through data-driven and ML methods and uncovers research opportunities in synergy with smart manufacturing techniques. The study aims to offer guidelines for stakeholders and to reinforce the industry’s pivotal role in circular economy initiatives.