Utilizing Untapped Human Operator Knowledge for Smarter and More Sustainable Manufacturing: Examples of Unconventional Sensor Data for the Human-in-the-loop Integration

Industrial activities represent about one-third of the world's annual energy consumption. A humancentered, smart factory environment can help to reduce energy demand and promote sustainable operations. While various cyber-physical systems have enabled the creation of new data value chains in manufacturing, most data flows are now automated; human cognitive skills still need to be better integrated into production processes. Despite existing digital support, there is still a lack of ways to engage workers that capture and enhance their cognitive abilities within socio-technical systems, which might also stem from incomplete data collection. This paper highlights the use of unconventional sensor data, including wrist acceleration, video data of operations, and electroencephalogram (EEG) measurements, to extract relevant information about human expertise in manual manufacturing operations. Our results demonstrate how unconventional human-sourced data can enrich cyber-physical systems and pave the way for more sustainable, human-centered manufacturing. The paper presents a feasibility study with illustrative examples rather than quantitatively proven sustainability improvements, explicitly framing the contribution as a proof-of-concept for human-in-the-loop integration. The aim is to inspire future research in this area. (c) 2026 Society of Manufacturing Engineers (SME). Published by Elsevier Ltd. All rights are reserved, including those for text and data mining, AI training, and similar technologies.