Detection and Classification of Human Activity for Emergency Response in Smart Factory Shop Floor

Factory shop floor workers are exposed to threats and accidents due their encounters with tools, equipment, toxic materials. There cases of occupational resulting in injuries precipitating lawsuits, which on the other hand affect company’s operational cost. To ensure safety within floor, there is a need for proactive activity monitoring. Such activities include detection falling objects, abnormal vibration, movement humans an acceptable area factory floor. Breathing sensor-based monitoring smart can also be implemented. This human activity, especially where isolation no available emergency assistance. Internet Things (IoT), Industrial (IIoT), machine learning (ML) have enabled so many possibilities this area. In study, we present simple test-bed, made up vibration sensor, breathing Light Detection Ranging (LIDAR) sensor. These sensors were used gather normal data at factory. We developed dataset based possible real-life situations it about 10,000 points. The was split ratio 75:25 training testing model. investigated performance different ML algorithms, including support vector (SVM), linear regression, naive Bayes (NB), K-nearest neighbor (KNN), convolutional neural network (CNN). From our experiments, CNN model outperformed algorithms accuracy 99.45%, 99.78%,100%, 100%, respectively, movement, breathing, distance. successfully assist research community field.