Table of Contents
By Phil Molé, MPH
Welcome to the second blog in our ongoing series on AI and EHS! In our first post, we’d provided a general overview of the topic to introduce it to EHS pros curious to learn more about the ways AI can support their life’s work of keeping people safe.
In this latest installment, we’ll take a closer look at machine learning (ML) and highlight the key reasons that EHS professionals should pay attention to it.
What is Machine Learning?
ML is the method of training algorithms to learn from datasets and provide useful feedback to improve performance, without being explicitly programmed to do so.
An AI model starts with some initial parameters, and as it processes more data, it adjusts those parameters to make better predictions. It’s an iterative process, so it’s always comparing new data to what would have been predicted by its model and then refining the model to improve its predictions.
A key idea here is that ML is one way that AI models improve their performance over time based on new information, which is where the “learning” part of the name comes from. And because it process and analyze new data much faster than humans do, it can learn much faster than humans do.
How Does Machine Learning Help EHS Professionals?
The simplest way to explain how ML helps is to say that it enables EHS professionals to conduct analyses they need to do anyway, only much faster and more accurately.
Potential for Severe Injuries and Fatalities (PSIF) Risks
For example, take the issue of identification of PSIF risks. As mentioned in our earlier blog post, rates of severe injuries and fatalities (SIFs) have remained stubbornly flat over the last 20 years, even as rates of less serious injuries have declined. We wouldn’t see that trend if we’d collectively done a good job assessing risks and addressing them before major accidents could happen.
That result didn’t happen because EHS professionals have been laying down on the job – far from it. EHS professionals understand the assignment, which is to maintain good records of workplace incidents and then use those records to identify risks and trends, and ultimately use that information to prevent more injuries. In other words, the EHS professional is using the dataset to learn, and identify connections between incidents and their root causes, and about the risks of injuries lurking within the data. The problem is, even a very smart person can only learn so much, so quickly. Human beings also have many factors that interfere with their learning, whether those are competing EHS tasks still undone, or personal issues weighing on their mind while they’re working.
An AI/ML model has no such problems. It can process much larger incident datasets than people can much more quickly, without the anxieties or distractions that limit human efforts to learn from datasets for past incidents. Therefore, it can find PSIF risks that are hidden in the details for less serious incidents, like near misses/close calls.
Are there other examples of how ML can supplement the expertise of EHS professionals? There are so many in fact, that we’ll be devoting many future installments of this AI and EHS blog series to discussing them. For now, here are a couple more.
Root Cause Analysis
EHS professionals all know that sometimes, even though they’ve done their best to assess and control risks, a workplace injury happens anyway. The next step at that point is to conduct a root cause analysis (RCA), to identify the underlying systemic reasons why the accident happened, so you can understand how to prevent it from happening again. There are two main problems that often derail this process.
The first is subjectivity, which is to say that any individual has preset ideas about what causes injuries. That’s why so many people conducting investigations fall back on the old umbrella category of “human error,” which at best is vague, and at worst implies that employees are t blame for their own injuries.
The second is what we might call “choice paralysis,” which is especially likely in some cases where software is used to conduct RCA. For example, someone using incident management software without AI/ML enhancements might find herself confronted by a menu of over 100 “causes” to choose from, which can be overwhelming. The end result of that process may be a kind of shoulder shrugging or guessing, and selection of a cause or causes that don’t precisely fit the situation.
Incident management software with ML-driven RCA guidance bypasses these common issues. Because the ML is driven by data and iteratively improves as the dataset grows, it not only is making objective evaluations but also gets better over time. Also, since it presents the user with limited, curated root cause choices, it avoids the “choice paralysis” that often undermine RCA.
Corrective Actions
The next step after you’ve done RCA is to develop and track a corrective action – i.e., something you’re going to do to address the underlying reasons why the incident happened and keep it from recurring. Of course, you need to address every individual root cause you identify through a corresponding corrective action, and the corollary is that the effectiveness of your corrective action selection depends on the effectiveness of your RCA.
It follows that one of the things that stops EHS pros from doing good corrective actions is that their RCA was lacking, and you’ve learned already that AI/ML can help there. Another reason that corrective action selection is often unsatisfactory goes back to the “learning” and data processing issue. The selection of a corrective action that suitably addresses a root cause depends on a knowledge base – you have to know which actions historically prove most effective countermeasures for specific risks, and the knowledge base of any individual person is going to be limited. AI/ML, on the other hand, can absorb and analyze massive amounts of data and objectively determine which corrective actions will work best in a particular instance.
ML Offers EHS Professionals a Generational Opportunity
Putting all of this together, you can see that AI/ML help EHS professionals level up by leveling up their ability to learn from data. You have the support you need to make fast, accurate assessments about risks and corrective actions that help you prioritize your actions. In other words, you’ll have the means to stop struggling to react to risks after something bad happens and finally get out three steps ahead of risks.
“AI and machine learning offer a generational opportunity to change the trajectory of workplace safety,” said Dr. Julia Penfield, PhD, VP of Research & Machine Learning at Velocity. “We don’t see AI as a replacement for human judgment but as a powerful ally. One that helps professionals see risk patterns sooner, act faster, and prevent the preventable.”
Ready to Learn More About EHS and Machine Learning?
Stay tuned for future installments of our AI and EHS blog series, where you’ll learn more about AI, specific use cases for AI in EHS management, considerations when evaluating EHS software and vendors, and more!
In the meantime, you can visit our AI Glossary & Learning Hub to continue learning on your own. There, you’ll find a curated list of resources covering various aspects of AI and EHS, as well as definitions of common terms.
We also invite you to download and read our new white paper, “Why EHS Professionals Can’t Afford to Ignore AI.” You’ll get a deep dive into all of the reasons why EHS pros like you have a generational opportunity to use AI to pivot from a reactive safety management approach to a proactive approach that reduces injury rates and fosters a positive safety culture. From there, you can also get some guidance on what to look for when selecting an EHS software vendor with AI capabilities with our new AI vendor evaluation checklist.
Let VelocityEHS Help!
If you’re ready to jump to the part about how Velocity can help, we’re standing by to talk!
We’d love to tell you more about how our VelocityEHS Accelerate ® Platform powered by Velocity AI and its interactive assistant Vélo can help you to end the cycle of struggling to keep up and start the cycle of staying a few steps ahead of hazards and risks.
We’re looking forward to talking about our PSIF Insights, our machine-language powered chemical ingredient indexing, our AI enablement supporting better root cause analysis, better incident descriptions and JSA job task descriptions, better controls selection in ergonomics assessments and JSAs, and auto-processing of contractor documents, just to name a few.
In fact, why not see for yourself how we can help? Get in touch today to set up a meeting so you can see our software in action.