Digitizing EHS inspections

Inspections, health and safety checks, and audits take a big part as the leading indicators for many organizations. Setting quantitative targets for the management, supervision, and EHS teams is very common across many industries. On the other hand, the quality of these activities is often being questioned:

• Are the companies getting the expected value from those activities?
• Is the time being used efficiently?
• How efficient is the output of those inspection activities?
• How do you measure the performance of the inspections and their efficiency?

I believe there are many more questions that you may ask or already have asked (or have been asked) before.

The time aspect is an important thing to consider for EHS inspections. You may hear data/figures like 4 hours/day per EHS supervisor, an inspection per 200 man-hours, etc.

What can be done to establish an effective inspection system that can help to monitor and control unsafe acts and conditions, generate robust improvement actions, and ultimately prevent accidents?

As technology is constantly evolving and improving; it maybe has the biggest potential to bring. Digitization of the inspections, automated reporting systems, and using artificial intelligence and computer vision are the first things coming to mind.

There are many advantages of utilizing technology for this routine part of Health & Safety. The first one is, of course; the time aspect – why spend 4 hours a day if artificial intelligence can be doing this for us by:

• Analyzing safety issues 24/7 with a high accuracy
• Generating objective results and outputs
• Notifying in real time
• Providing data that all your colleagues can access simultaneously

So, you don’t need to have a lot of file binders, keeping the completed checklists that mostly no one is looking at after the inspection is completed.

The other aspect of the digitalization of this process is about reporting unsafe acts/conditions and incidents. Companies are having many behavioral safety programs, observation card systems, etc. to encourage reporting. But we all know that it is impossible to report all the incidents, especially the unsafe acts and conditions. Let’s assume all of them are reported; how this process can be managed in terms of time, resources and efficiency? Even reporting an incident and following up on the corrective actions require days for the management and EHS teams. So again, if this can be done in the digital world of today, it makes perfect sense to utilize the technology, get a clear picture of your workplace in terms of unsafe acts and conditions, and even access a detailed analysis of those.

You may ask the question – what to do with this data? If you receive 200 reports a day, how can this be managed, even with the applicable technology? This is a great question to think about, and my answer to this would be prioritization which requires human involvement. For example, if you are receiving 100 hard hat unsafe act reports per day, it doesn’t make too much sense to focus on all the individual reports. But if you take corrective action, i.e., an EHS campaign, a safety talk, or even a simple poster and if this corrective action is reducing your hard hat reports to 40 per day, isn’t this a great value to reduce the risk of exposure or to improve the PPE compliance?

Let’s take another example, a high critical working at height unsafe act – here, the real value lies in the identification of this activity, which maybe you weren’t aware of before or even not assessed in your risk assessments. Focusing on this one unsafe act can help you a lot, as taking corrective action and preventing it from occurring again can help you to eliminate a serious workplace accident in the future.

In this example, you see how AI and CV can help you to detect unsafe acts and conditions. To be able to prevent the occurrence of this unsafe act which can result much worse, you need to understand the root causes. If it is approached as a robust incident investigation, you can extract much more value and come up with appropriate mitigations. One of the ways to operationalize this, you can include the questions, and root causes that you had gathered from the unsafe act. Some example questions are:

• Is there a job hazard analysis or safe work instruction for rigging the materials?
• Does the rigger have proper training? Does the training include safe loading/offloading of the materials?
• Is there proper access equipment nearby?
• Is there a process to report these types of hazards at the workplace? Do people have the right of stopping their activities when they encounter these types of unsafe conditions?
• Is the lifting gear certified and appropriate for the load?

So now, we don’t only have the image and video for the “working at height” related unsafe act; we also deep-dived into the possible causes and added these controls to our digital inspections which will be routinely controlled. Isn’t AI, computer vision and digitization of the inspections a perfect combination for the control of workplace hazards?

With the given examples and variety of the EHS scenarios like those above, it seems like we really need a helping hand from technology. Basically, to identify the unseen hazards, save time through reporting and inspections and take robust corrective actions to prevent them before any serious accident takes place.

And yes, AI and CV helped us a lot to identify unsafe acts and conditions. Now, Intenseye will be launching the new “Digital Inspection” module soon, where you can deep dive into the root causes of these types of unsafe acts and conditions and take corrective actions accordingly. This will certainly be an important element to close the loop and will help to make the workplaces much safer!