By using tdwi.org website you agree to our use of cookies as described in our cookie policy. Learn More

TDWI Upside - Where Data Means Business

How Predictive Analytics Can Improve Manufacturing Safety

These five use cases illustrate how predictive analytics is being used to keep workers safe.

Worker safety should always be a top priority for businesses. In the past, safety programs have relied heavily on manual inspection, opinion, and attitudes on the shop floor. Traditional metrics (such as incident rate and lost-time incident rate) only track issues after the fact.

For Further Reading:

The Age of Instrumentation Enables Better Business Outcomes

Predictive Analytics and Employee Retention: A Winning Combination

IoT and the ML Connection

As with other areas in the manufacturing sector, safety professionals are beginning to use data analytics to perform a more objective evaluation of their environment. By harnessing the power of data and predictive analytics, businesses can begin to track safety and reduce risk using leading rather than lagging indicators.

Let’s look at five examples of how predictive analytics is being used in enterprises to improve worker safety.

Remote Monitoring

Often, collecting machine data for predictive analytics involves remote sensors. A side benefit of sensors is that equipment can be checked without workers being near it, which is particularly useful in hazardous locations. A technician does not need to be exposed to risks unnecessarily.

An additional benefit is equipment uptime. When equipment breaks down unexpectedly, maintenance workers must rush to repair it. This can lead to unsafe situations. Was the lock out/tag out procedure followed properly? Was the technician hurrying to fix the issue and did he or she cut corners? Remote monitoring combined with predictive maintenance means that most breakdowns can be foreseen, which reduces these risks dramatically. Technicians can take the time to ensure equipment is operating correctly and safely and that proper procedures are followed.

Scenario Modeling

After an incident occurs, it is essential to determine the root cause quickly. Your analysis helps you prevent future similar incidents. Software can aid in this analysis by running potential scenarios and quantifying risk factors. A case study in Industrial Hygiene and Safety News pointed out that:

Predictive analytics allowed for immediate processing of data through What-If scenarios. ... 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 in Chemical Reactions

In chemical manufacturing, production of some products involves dangerous reactions. These reactions need to be carefully controlled to avoid hazards, and many manufacturers are using predictive analytics to do so.

When a deviation in normal condition occurs, a smart control system will react and automatically adjust system parameters to mitigate risk. These systems can react faster than humans and point out anomalies that are not obvious. Historical analysis of events can be completed faster with software, saving time and effort.

Natural Disaster Modeling

You may think that worrying about natural disasters is a bit extreme in the context of this article, but they should be part of a company’s health and safety program. The effects of natural disasters can be devastating. For example, in 2017 Hurricane Harvey caused a huge disruption to industry in Texas. It caused an estimated $85 billion in damages; 82 people died. Many plants shut down or reduced their output.

How can analytics help? Past events have created large data sets that can be used to predict future occurrences. For instance, potential flooding areas can be mapped accurately. When an enterprise recognizes it is at a higher risk for a flood, it may change decisions about its infrastructure and operating practices.

Worker Monitoring Through Wearables

Some businesses have begun to implement tech wearables, such as glasses with displays, hard hat sensors, or location detectors. Sensors can be used to study patterns of movement. When a person is fatigued, there are certain physiological signs that can be detected using technology. For example, a specially designed hat can alert the wearer if there are signs of drowsiness predicting sleep on the shift (a dangerous situation).

Other hazards can be anticipated, keeping workers safe. Through tracking a worker’s temperature, heart rate, and activity level, potential heat exhaustion can be predicted. Just as heat hazards can be mitigated with predictive software, so, too, can the risks of working in the cold.

A Final Word

These digital strategies have started to make their way into hazardous manufacturing environments. Companies will continue to use predictive analytics in many ways. Because safety is a core element in manufacturing, it will likely receive more attention in the IoT and predictive analytics space. As these kinds of tools gain in use, the workplace will become safer than ever. Reducing risk affects the bottom line, and it is simply the right thing to do.

About the Author

Bryan Christiansen is the founder and CEO at Limble CMMS. Limble is a modern, easy-to-use mobile CMMS software solution that takes the stress and chaos out of maintenance by helping managers organize, automate, and streamline their maintenance operations. You can contact the author via email or on social media.


TDWI Membership

Accelerate Your Projects,
and Your Career

TDWI Members have access to exclusive research reports, publications, communities and training.

Individual, Student, & Team memberships available.