AI, Machine Learning and IoT are rapidly changing the face of worker safety. These fast-evolving technologies might not as yet guarantee a turnkey solution, but have no doubt: they can and will contribute significantly. Here’s our quick rundown of the do’s, don’ts and often unexpected considerations when embracing AI in worker safety and your overall safety structure.
When it benefits productivity and the bottom line, industries generally proved themselves noticeably open to adopting new and fast-evolving technologies. Machine Learning and IoT being some of the more common recent examples. A similar acceptance of these and other innovative technologies is now transforming worker and workplace safety. Confirming companies increasingly acknowledge that safety ultimately benefits said productivity and bottom line.
Avoiding the don’ts
So how do AI & co assist in reducing accidents and injury rates to (almost) zero? And what about the voices arguing these same technologies might have adverse effects on worker safety and productivity? An often-mentioned example is that increased monitoring can lead to the so-called ‘Big Brother’-perception, micro-management, augmented stress and anxiety. It could also lead to a technology-centric, overstated and ultimately counterproductive belief in the authority of robotics, algorithms, predictive systems etc.
These arguments are most certainly not without merit. If only to recognize that these new technologies are not a guaranteed one size fits all solution for every occupational health and safety problem. Also, we should bear in mind that technologies should never be considered in isolation. It’s the way they are implemented that creates positive and, luckily to a much lesser extent, negative conditions.
Embracing the do’s
The possibilities for workplace progress, growth and safety with the integration of AI applications and tools are many and various. New technologies give companies the capability to collect an ever-increasing amount and diversity of data. The challenge is to process and analyze this data, whether historical or in real-time. Thereby transforming it into accessible and actionable insights.
1. Reducing of human error
An undisputable majority of accidents and injuries in the workplace are directly or indirectly caused by human error. Often a relatively small mistake or momentary lapse of judgment can have grave consequences for oneself or others and can cause injuries or worse, as well as costly damages in material and downtime. AI is far less prone to mistakes and so eliminates an important cause of safety issues.
2. Employee monitoring
Yes, we know how those two words sound. Still, when done right, AI in worker safety can help in strengthening the benefits of employee monitoring while strongly diminishing the disadvantages. Consider the implications for training, lone worker support, evacuation management, zone management, permit-to-work, etc. .
3. Lifesaving decision making
AI can analyze data and predict dangerous situations almost instantly. This allows for a potentially lifesaving increase in alert- and response time. With or without a human checkpoint in the safety loop.
4. Automation curbs danger
With AI being able to perform increasingly complex procedures and tasks, a lot of the more hazardous jobs get taken out of human hands. AI is also progressively capable of handling repetitive jobs that previously required a human eye or intervention. Jobs that, because of their monotony, led to mistakes and safety incidents because of boredom, overconfidence or simple fatigue.
5. Freeing up labor
Ideally, when AI-driven technology takes on dangerous or repetitive (parts of) jobs, it frees up workers for tasks that humans are better equipped for. For example, those involving creativity, esthetics, empathy, supervision, … . Also: AI doesn’t necessarily take away a human task. Often the combination of humans with AI support can achieve significantly higher qualitative and quantitative results in efficiency, productivity and safety.
6. Preventive insights
Current technologies allow for the collection of vast amounts of data. AI and Machine Learning can decidedly assist in transforming this often-underused data into new insights and actionable information. An example: by tracking unreported near-miss incidents between forklifts and on-foot workers AI can highlight previously unknown hazardous zones without – and before – there ever being an actual accident. AI can point out weak spots, identify trends and provide you with a better view/understanding of your entire production process.
