From harmful substances and chronic occupational diseases to injuries and even fatalities, a manufacturing plant is, no doubt, one of the most dangerous areas for its employees' health and wellbeing.

According to the ILO’s estimations, 2.3 million women and men around the world succumb to work-related accidents or diseases every year; this corresponds to over 6000 deaths every single day. Worldwide, there are around 340 million occupational accidents and 160 million victims of work-related illnesses annually. The ILO updates these estimates at intervals, and the updates indicate an increase in accidents and poor health. [1]

Thanks to advanced solutions and smart AI algorithms, enterprises around the world are to reduce these horrific numbers and protect their employees from life-threatening events.

However, these are only some of the issues that business leaders should be aware of. They also need to manage the plant’s workflow to keep it up and running, including things like employee performance, and equipment and product condition, etc.

Here, high-quality AI-powered computer vision solutions are of great help at creating and maintaining a safe and comfortable work environment, optimizing streamlined workflows, and reducing costs.

Passing a medical examination

In the production industry, employees are often required to undergo routine medical check-ups before starting work. A doctor conducts a general examination — measures blood pressure, body temperature, checks if the person is under alcohol intoxication, etc.

The problem arises when we understand that the doctor can ‘forgive' minor deviations or ignore them simply because the doctor is a human being. So why not automate medical check-ups?

After the employee enters a special booth, cameras scan their face. This way the recognition system eliminates cheating during the check-up. The computer runs all the tests, records the result according to regulations, and sends it to the foreman. In addition to accuracy, an automated medical checkup takes only several minutes, which is six times faster than a regular exam with a doctor.

Irrational Use of Working Hours

How much time did they spend at their workplace? Where are they at the moment? Smoking again, aren’t they?

Factory workers tend to take breaks 30−40 minutes before the end of their shift. Irrational use of working time often results in decreased productivity and efficiency. Video analytics allow managers to keep track of employees' activities throughout the shift. Moreover, the presence of cameras works as a preventive mechanism to avoid violations in the future and to maintain efficient workflows.

Key-card Issues

Employees often forget their passcards, lending them to each other and creating the risk of the card falling into the wrong hands, sometimes into the hands of an intruder. An employee can easily leave their key card at home. Facial recognition eliminates this problem.

Some may claim that biometrics can be deceived or spoofed, for example, by showing a printed photo or a photo on a mobile phone to the camera. It depends on the quality of the system. NtechLab’s Liveness technology is capable of easily recognizing any such attempt by checking for a living person in the camera’s frame.
Liveness — Protection against spoofing reliably differentiates a real-life face from an image of a face and guarantees that there is a real person in front of a camera, thus significantly reducing the risks of fraud.

Reducing Human Labor

You can reduce the use of human resources when you can automate the process and thereby save money. Let’s take flaw detection, for eхample. Manufacturers often need to check products for defects and abnormalities manually. Video analytics combined with robotization allow you to automate these processes using computer vision. It is still a work in progress but just imagine how much manufacturing can benefit from this technology.

Special Requirements

A variety of incidents can lead to downtime. For instance, the lack of special uniforms in cleanrooms can lead to poor product quality. Here, the owner of the enterprise risks releasing a whole batch of defective products.

Video analytics helps prevent incidents related to both product safety and workers' health conditions. The system helps comply with specific requirements and can identify closed helmet visors, worn face masks, or protective eyewear.

Incident Investigation

Let’s say something has disappeared from storage over the last 7 days, but we do not know who the culprit is. Even with a camera and the correct viewing angle, the operator will have to manually review the entire video stream to identify unique visitors and talk to the suspects. In the best-case scenario, it would take at least a week to resolve the missing items.

With video analytics, the algorithm can immediately compile a list of unique visitors. It can then quickly determine who was in this area around the time of the incident.

Every enterprise has different business tasks. Some of them need to monitor container movements, others must be able to identify helmet use or one of many other things. One way or another, the solution for the manufacturer focuses on two factors: effectiveness and safety.

Management of access and employee movement in closed facilities and the monitoring of safety rules compliance:

  • Access to facilities through face recognition only or through two-step verification (face recognition + keycard or code).
  • Monitoring of trespassing in restricted zones.
  • Monitoring the presence of personal protective equipment (masks, etc).
  • Facility zoning: biometric control of employee access to secured areas.

How It Works and Where We Start

  • Step 1. Conduct an audit to understand the customer’s needs.
  • Step 2. Define individual objects and certain recognition scenarios.
  • Step 3. Train neural networks and create business logic.
  • Step 4. Deliver the final product.

When implementing video analytics at industrial facilities, all video streams must be gathered in a single place. Our system then simply connects to the centralized storage. That is, it takes a video stream and returns analyzed business requests.

You do not have to be a huge manufacturing corporation to enjoy the benefits of our video analytics technology, including access management, employee movement monitoring, and safety rules compliance. Once we collect the initial parameters, even if you don’t obtain the required equipment, we can help you out.

One of the largest oil and gas companies, as well as other manufacturers, already use NtechLab’s technology with 20 connected cameras and 500 people in their database.

[1] ilo.org/moscow/areas-of-work/occupational-safety-and-health/WCMS_249278/lang-en/index.htm