Now more than ever, the safety of employees is a top priority for businesses. In manufacturing facilities, there are often a lot of moving parts, people and departments, so safety is even more critical to ensure operations run smoothly. In July's Back to Business with AI virtual event, David McCall of Tyson Foods shared how AI has impacted them in a positive way and how safety has remained top priority in their manufacturing plants.
You can view McCall's complete session in the video above or catch a recap below.
While, symptomatic workers of COVID-19 are a heavy concern for any workplace, busy manufacturing companies like Tyson Foods are implementing the necessary AI-related technologies to help monitor temperatures, proper PPE and social distancing protocols among employees. Tyson has implemented computer vision systems, machine learning (ML) and thermal imaging/cameras to help make this happen. As employees enter a plant and clock in, for example, thermal cameras measures their temperature and integrates with thermal imaging systems to alert management should an elevated temp be detected. Additionally, computer vision systems can monitor other necessary protocols, such as mask wearing and distance, are followed throughout the plant.
Tyson manufacturing also uses geofencing to ensure workplace safety. Geofencing is essentially the practice of using GPS or radio frequency to define a geographic boundary. Once the virtual barrier is established, triggers can be set to alert the proper officials should issues occur within the defined boundary.
In dynamic geofencing, a vision system is mounted to a moving vehicle to view different perspectives. This vision system, with the power of AI, continuously searches for scenarios of concern as it moves around the facility and alerts management of issues. As an example, if someone crosses the plane of a geofence surrounding a hazardous area, an alert can be sent out to turnoff dangerous equipment in that area to keep the employee from getting hurt.
Like dynamic geofencing, static geofencing uses a vision system to detect hazardous scenarios, though in this case, the cameras are static in one place. In manufacturing, equipment is typically located in the same area, so this static system would be mounted in one spot overlooking the entire area. This could even be utilized along an assembly line to watch for material or items that might collide.
With the use of IoT sensors and ML technology, manufacturing plants can monitor equipment that may be deteriorating, which could lead to dangerous situations for plant workers. By building models off of incoming data, this technology can be essentially "trained" to understand what good equipment looks like and what equipment defects look like, in real-time. Anomalies around electric equipment, for example, may include amperage pole or temperature changes that indicate there is a problem that needs to be addressed immediately.