The Future of Machine Learning in EHS and ESG
Why machine learning? Why now?“Machine learning” is a general name given to any technology-based approach that uses methods that “learn” or adapt, and in doing so, help the user solve problems using data. We communicate to machine learning software, aka a “model,” to train it to communicate with us, and turn data into actionable insights. In that sense, machine learning is a form of artificial intelligence (AI). We’re hearing more about machine learning today, but it’s nothing new – IBM employee and gaming pioneer Arthur Samuel coined the term in 1959.
But what is new is the ability to harness the benefits of machine learning at minimal cost and maximal analytical power. This isn’t your grandparents’ machine learning. Modern technology makes it possible for EHS professionals to use machine learning at scale, with minimal disruptions to business and much less time getting the technology, and the workers using it, informed and up to speed.
Use Case Example: ErgonomicsThat’s all well and good, but how would an EHS or ESG professional use machine learning? There are many ways, but for now we’ll focus on the example of industrial ergonomics management.
Ergonomics, of course, is the process of fitting the work environment to the worker. A central task in ergonomics is identifying and controlling potential for musculoskeletal disorders (MSDs), which are injuries to muscles, tendons, ligaments, joints and nerves from overexertion, repetitive motions or awkward motions. Common ergonomics assessment challenges include determining if you’re evaluating the right job tasks, and whether you’re accurately measuring body angles and force exertions.
In the past, many technological tools an EHS professional used added extra layers of complexity to the task, such as sensors and cameras that themselves required constant calibration and needed optimal conditions such as perfect lighting to deliver useful information. Even if the EHS professional used then-available machine learning tools to analyze the collected data, she/he often needed to spend months of work getting the machine learning model “trained” to provide useful information.
Things are different today. Modern AI-driven ergonomics software can run from a simple mobile phone used to record movements of employees as they perform job tasks. You can get the model up and running within days, and without major interruptions to work. Instead of having to identify where all the force is applied in tasks, the software, relying on embedded ergonomics expertise, provides you with useful information about ergonomics risks based on just a few key inputs, like direction of motion, in addition to analysis of employee physical motions.
Of course, one of the most important factors is how you use the information you get. Software can help there, too, by identifying job tasks with the highest ergonomics risks, so you can prioritize. From there, the software helps identify root causes from expert-built lists for the specific MSD risks, which in turn helps you identify effective control measures. In short, you don’t need to be an expert anymore to manage ergonomics like one, because expertise is baked into the machine learning model.
Ergonomics is only one of the areas of EHS where machine learning can help. For example, machine learning can help EHS professionals sift through large quantities of injury and illness records to identify trends and patterns and generate lessons learned so they can more effectively manage prevention.
Benefits of Machine Learning for EHS and ESG ProfessionalsWe can see that the direct benefits of machine learning are the collection of better information and faster insights, which will enable better decision making to help make workplaces safer and more sustainable. There are also many benefits beyond this, including:
- Improved engagement. One of the most persistent challenges for EHS managers is that they feel they must “go it alone,” and shoulder all the responsibility for managing EHS. But modern machine learning-based software can not only cut down on time involved to manage key safety tasks, but also help share responsibility for safety. In our ergonomics example, employees participate in identifying their own safety risks via the motion capture exercise, and then can more easily share their insights about root causes and feasible corrective actions during follow-up. They’ll see how their participation matters and be more likely to participate in your EHS programs going forward. Higher levels of employee engagement are the foundation of a mature management system and give you the support needed to shift from traditional compliance-focused EHS management to Environmental, Social and Governance (ESG) maturity
- Reduced turnover and lower costs. Machine learning makes it easier for employees to participate in and improve safety. They feel safer working for you and feel that they are making a difference. Their awareness of the processes for identifying and controlling workplace risks will reduce their stress and the psychosocial risks that come with it, including lack of job control, low levels of influence, and underuse of skills. Involved and informed employees are committed, feel respected, cared for and want to continue working for you. That’s an especially important consideration during times of high employee turnover, like the present, where lost productivity and increased costs in onboarding take a toll on business growth
- Continuous improvement and relevance to stakeholders. Using AI technologies such as machine learning cuts down on time spent on maintenance or tactical work and frees up time to focus on higher-value, more proactive approaches, such as ESG initiatives. Increasingly, both internal and external stakeholders expect higher levels of management performance and a focus that moves beyond regulatory compliance. Machine learning tools give you the agility and insight to get there
About the Author
Phil Molé is an EHS & Sustainability Expert for VelocityEHS. Phil speaks at numerous conferences and organizations, and often presents webinars on EHS management topics.
He previously worked as a Global EHS Coordinator for a large manufacturing company, where he developed and facilitated trainings on a variety of subjects and managed the company’s ISO/OHSAS certifications.
Phil has also worked as an environmental and safety regulatory consultant for approximately 13 years.
Phil's professional accreditations and memberships include: National Institute of Occupational Safety and Health (NIOSH) Traineeship from 1995 to 1997, American Society of Safety Professionals (ASSP), National Association of Environmental Managers (NAEM), OSHA 30-hour training (2012) and ISO/OHSAS Internal Auditor training (2012).
Phil has received a Bachelor of Science degree in Chemistry from DePaul University in Chicago, Illinois, and a Master’s in Public Health (MPH) in Environmental Health Sciences from UIC in Chicago, Illinois.