Sustainable preprocessing for AI repairability assessment
Aleksandr Bystrov and Ole Meyer and Fabian Kott and Karin Saemann and Achim Maat and Lisa Dawel
IFAC-PapersOnLine
The automotive industry increasingly embraces repair practices driven by sustainability trends shaped by government policies and consumer preferences. Artificial intelligence (AI) enhances repair processes through versatile applications like computer vision. However, its energy demands and carbon footprint have recently garnered attention. Image data preprocessing techniques applied during the data derivation stage offer solutions to reduce AI’s environmental impact. This paper presents novel preprocessing strategies implemented during the data collection phase to address challenges in applying computer vision within the automotive repair industry. The study balances the trade-of between deployment benefits and energy demands, particularly focusing on Fault Detection Models.
2025
article
2766-2771
CIRC-UITS
Circular Integration of independent Reverse supply Chains for the smart reUse of IndusTrially relevant Semiconductors
Circular Integration of independent Reverse supply Chains for the smart reUse of IndusTrially relevant Semiconductors
