OPTIMIZING HUMAN FACTORS AND SAFETY MANAGEMENT IN SMART WAREHOUSES USING A FUSION OF INTERNET OF THINGS (IoT) AND COMPUTER VISION

Abstract The warehousing business, especially in the U.S. is experiencing serious problems which involve labour shortage, expensive running and the safety of the workers. Although the full automation of warehouses is the future, it is expensive in terms of conversion and cannot be implemented anywhere at the moment. In turn, human-robot collaborative warehouses have become one of the workable solutions in order to enhance the operational efficiency and safety. The given paper introduces an innovative method combining the Internet of Things (IoT) technology of sensing data with computer vision (CV) to streamline human and safety management in smart warehouses. The IoT sensors will gather real-time information about physiological and behavioural variables by providing the workers with some kind of wearable (smart wristbands or vests), which will capture data (real-time) about things such as heart rate, body temperature and movement patterns. Together with computer vision, the cameras are widely installed across the warehouse to observe the posture and behaviour of the workers and detect possible workplace risks, including the incorrect lifts and unsafe work with automated vehicles. Combining IoT and CV data and then performing machine learning model processing will allow predictive insights of worker fatigue and risk of injury, which can be used by the manager to act in the efficient operation of the workflow and responding to the safety risks proactively. The paper has pointed out the transformation to being able to replace the human being to being able to assist the human being, with the increasing role of digital sensing and data-driven decision-making process in improving safety and productivity of the warehousing activity. The offered solution provides a cost-effective and scalable approach to optimization of human-robot collaboration and improving labour welfare, as well as general performance of the warehouse.