Predictive analytics: A powerful new tool
Identify at-risk workers & teams for pilot projects
Most companies have data that can be used to prioritize and analyze employees, processes or workspaces at elevated risk. But currently, most organizations must first compile and export their data from multiple environmental, health, safety, and quality (EHSQ) and human capital management (HCM) platforms and then analyze it outside of these applications using tools, such as Microsoft Excel, Cognos, or Tableau.
Big Data benefits
Advancements in data science dramatically improve the information available to EHSQ professionals. Data science is making ‘Big Data’ applications, such as predictive analytics, mainstream. Predictive analytics allows you to add a future-looking element to your data that was not previously possible. Predictive analytics finds patterns in data to predict a future state by using those patterns. Although predictive models are not the crystal ball that gives the exact time and place that an event will occur, they can find patterns associated with elevated risk.
And of course, the more you elevate your risk of injury or incident, the more often it will happen. Most predictive models are created using machine learning, which can account for the complex relationships among variables that no human could ever program. They can also automatically adapt and improve themselves through feedback from their use, which supports the constantly evolving nature of your EHSQ programs.
Identify at-risk workers – but don’t punish
EHSQ programs and policies are designed to support and ensure the overall health and wellbeing of a workforce. They are not constructed to identify and punish individuals. This is the same with predictive analytics. For example, the models can be designed to identify workers at most risk of getting injured over a specified period – accounting for variables like shift, tenure, job position group, training, and more. To maintain employee privacy and ensure a fair workplace, software companies must remove all personally identifiable information (PII) from a set of data prior to its use.
Predictive analytics is a tool that allows EHSQ professionals to do a better job – providing a wealth of insight and variables that allows them to design more effective workplace safety processes and policies. It is not designed to isolate, identify and punish employees.
Do you have a smart speaker at home? A Google Home, Amazon Echo, or Apple HomePod device? If so (or even if not), these devices are designed to recognize your device and know what you’re saying without having to train it. For example, you can ask Google to call your mom and it will. It will even call your spouse’s mom if your spouse is the one asking. Apple calls this functionality an “Intelligent Assistant,” and that is a good example of what predictive analytics can do – it’s an assistant that can intelligently point out areas of interest surrounding worker safety.
Identify pilot groups for new programs
An organization can leverage data to gain deeper insight to assist in the planning of process and policy development. This data can be used to identify pilot groups to test new safety and wellness programs. For example, when partnering with a large manufacturer, we observed a spike in the anticipated risk of injury as a result of what may have been worker or position complacency. This allowed the EHSQ professionals in the organization to further investigate why. Was it due to repetitive strain, the need for workplace job change, or something else?
In this case, predictive analytics allowed for immediate processing of data through What-If scenarios within the EHSQ platform. The manufacturer was able to choose multiple variables and gain immediate insight. They were easily able to quantify and display their findings to gain support and budget to pilot projects.
Predictive analytics for workplace safety will continue to evolve as models and databases expand to include:
- human capital management data (workforce management, scheduling and learning management data)
- peer data
- 3rd party data such as weather information
- data from other connected devices (badging systems, telemetry solutions, heat and/or fatigue monitors, and more).
As the sources and volumes of data increase, the ability to have a 360-degree view of worker safety will too.
The more we can improve predictable processes and policies the greater positive impact we can have on employee safety, operational uptime, and profitability. Putting predictive analytics directly in the hands of the EHSQ professional is like giving them a superpower.