Article: Data fusion for improved circularity through higher quality of prediction and increased reliability of inspection Journal: International Journal of Sustainable Manufacturing (IJSM) 2022 Vol.5 No.2/3/4 pp.164 - 199 Abstract: In order to meet customer requirements and regulations, such as low carbon footprint, companies are implementing AI-enhanced applications in production. However, AI is often used in stand-alone applications and lacks integration into the overall life cycle of products. To address this gap, this article proposes a framework for improving circularity through data fusion methods in product inspection. Data fusion combines multiple sources of data, such as sensor and business data, to improve machine-based predictions. The framework analyses AI applications, prediction during inspection, and data fusion methods, and addresses challenges in integrating business data into predictions. It demonstrates how data fusion improves prediction quality and stability in inspection. The framework is applied and evaluated in a case study from the automotive sector, showing an increase in good-quality predictions based on sensor data, leading to improved resource efficiency and circularity. This framework can be applied to any sector seeking sustainable manufacturing (SM). Inderscience Publishers - linking academia, business and industry through research

Title: Data fusion for improved circularity through higher quality of prediction and increased reliability of inspection

Authors: Robert Schimanek; Pinar Bilge; Franz Dietrich

Addresses: Chair of Handling and Assembly Technology Research, Institute of Machine Tools and Factory Management, TU Berlin, Pascalstraße 8-9, Berlin 10587, Germany ' Chair of Handling and Assembly Technology Research, Institute of Machine Tools and Factory Management, TU Berlin, Pascalstraße 8-9, Berlin 10587, Germany ' Chair of Handling and Assembly Technology Research, Institute of Machine Tools and Factory Management, TU Berlin, Pascalstraße 8-9, Berlin 10587, Germany

Abstract: In order to meet customer requirements and regulations, such as low carbon footprint, companies are implementing AI-enhanced applications in production. However, AI is often used in stand-alone applications and lacks integration into the overall life cycle of products. To address this gap, this article proposes a framework for improving circularity through data fusion methods in product inspection. Data fusion combines multiple sources of data, such as sensor and business data, to improve machine-based predictions. The framework analyses AI applications, prediction during inspection, and data fusion methods, and addresses challenges in integrating business data into predictions. It demonstrates how data fusion improves prediction quality and stability in inspection. The framework is applied and evaluated in a case study from the automotive sector, showing an increase in good-quality predictions based on sensor data, leading to improved resource efficiency and circularity. This framework can be applied to any sector seeking sustainable manufacturing (SM).

Keywords: data fusion; inspection; artificial intelligence; AI; remanufacturing; circular economy.

DOI: 10.1504/IJSM.2022.134562

International Journal of Sustainable Manufacturing, 2022 Vol.5 No.2/3/4, pp.164 - 199

Received: 30 Dec 2021
Received in revised form: 31 Jan 2023
Accepted: 22 Mar 2023
Published online: 27 Oct 2023 *

Title: Data fusion for improved circularity through higher quality of prediction and increased reliability of inspection

Keep up-to-date