© Jakub Jirsák | Dreamstime.com
Dreamstime L 17143545 6025abd6c70c2

How to Make Better Decisions, According to Data

Feb. 12, 2021
Lean into the power of data analytics to make decisions in real time by harnessing the power of historical data.

It’s a widely cited claim that the average adult makes 35,000 decisions per day, or roughly 2,000 decisions per hour, or one decision every two seconds. That includes decisions about what to wear or what to eat for breakfast.

The outcomes of these decisions will most likely not have a large impact on that day, month or year. However, for safety professionals, sprinkled into those thousands of daily decisions are potentially impactful decisions that will not only affect themselves, but others. And the outcomes of those decisions could have life-altering repercussions.

Safety professionals lean heavily on personal experience and heuristics to make decisions with the goal of mitigating risk for employees and the company as a whole. However, experience and heuristics can sometimes fool the decision maker.

In the world of safety, decisions are generally classified as safe or unsafe. The details (data) surrounding outcomes are typically not observed, reported, measured or tracked. Therefore, relying on past experiences to make future decisions gives safety professionals a false sense of security, which can lead to misguided judgements and potentially unsafe outcomes.

Data Driven Decision-making

Data and analytics have historically received a bad rap. More than 96% of data collected and utilized today in the safety industry is historical frequency data. In many ways, historical frequency data is similar to playing roulette at a casino.

Knowing where the ball stopped on the roulette wheel for the last 20 spins isn’t an indicator of where you should place your bets for the next spin. Though it may seem like it, the probability of the ball landing on a red or black is the exact same.

Historical frequency data may make safety professionals feel like they have an advantage on the future but unfortunately, they do not. If observations exist on an angle grinder with the only tracked measure being the outcome—safe, 80 times—safety professionals have no data to leverage when a decision needs to be made. The key becomes what data needs to be tracked and measured to create insights that safety professionals can deploy in real time. 

Before safety professionals can make decisions, they must first answer two questions:

  1. Do I know what it will take to succeed?
  2. Can I predict the probability of an outcome?

That’s where a reimagined look at data and analytics comes into play.  

Developing Safety Metrics

The goal of collecting safety data and analytics should not be to abandon core safety management tactics that have been developed but rather to enhance them. Risk intelligence tools exist today for this exact purpose. Thanks to technology, historical frequency data can be transformed into predictive risk intelligence insights that safety professionals can leverage in their decision-making process.

Here are two examples of how safety professionals can use data to answer these fundamental safety questions.

Do I know what it will take to succeed?

In order to answer this question, safety professionals traditionally lean on training, operating procedures, job hazard analysis, job safety analysis and job safety plans. Safety professionals assume from these documents that employees understand how to succeed. However, instead of assuming comprehension, safety professionals should measure the execution of training, pre-job hazard analysis and the like with the goal of developing new safety metrics that would lead to advanced data that could actually impact decision-making.

Example 1:

A company has completed another working from heights training. The company’s safety team can pull all observations from previously recorded working from heights training and compare results. The company now has a benchmark to quantify the return on investment for the training. Safety professionals can also see the amount of risk that has been mitigated. They can also track the amount of risk absorbed along with what key additional risks and mitigators were impacting these measures.

Result:

Three months after the training, the company has found a 3% increase in risk mitigation in observations recorded on working from heights. The main impact of the training was a thorough explanation of transitioning anchorage points, which has decreased as a correction 82%.

Can I predict the probability of an outcome?

Safety professionals are tasked with making decisions as if they have a crystal ball. Usually, those decisions are based on previous experience, such as “Have we previously completed this task, in the same manner, and the result was safe?” Depending on any human to quantify an infinite amount of data in real time and then make a reasonable decision is difficult. However, with data analytics, this decision can now be made with evidence rather than intuition.

Example 2:

A crew of four employees will be removing a piece of a pipe that is in a pipe rack. To do so, they will be cutting the pipe with an angle grinder. Because of the pipe rack, the crew will need to remove the handle from the angle grinder.

The safety professional reviews the data for cutting material with angle grinder at the current location for and sees the probability of incident is 19.3%. Removing the handle increases the incident probability by 7%. Furthermore, removing the handle and not using 2-hand operation increases the probability of incidence by 24%. As the safety professional dives deeper into the data, removed handle and not using a 2-hand operation have been combined 87% of the time.

Result:

The safety professional has identified the incident probability could have a potentially dramatic increase due to the working conditions. Prior to work beginning, the safety professional meets with each crewmember to personally discuss the need for 2-hand operation with the handle being removed. The handle being removed makes the operation inherently more dangerous, but because of the safety professional’s pre-task intervention, 12 observations were logged, all of which returning 2-hand operation present in the crews mitigation. The crew is able to accomplish the task safely.

Safety professionals experience will always be important. Safety management techniques will always be important. Risk intelligence is the tool to optimize safety performance and make safety professionals better equipped to make decisions in real time by using historical data to prevent future injury.

Kevin Miranda is the vice president and director of operations for MAC Intelligence and chief of operations at MAC Safety Consultants, Inc. He earned his bachelor’s degree in economics from Geneva College, a master's degree in health and safety management from the University of Alabama and his MBA from Cornell University.

Sponsored Recommendations

Managing Subcontractor Risks: Ensuring Compliance and Mitigating Disruptions in Complex Supply Chains

Sept. 26, 2024
Learn how to manage subcontractor risks and ensure compliance in complex supply chains. Explore best practices for risk mitigation, communication, and accountability.

Navigating ESG Risk in Your Supply Chain

Sept. 26, 2024
Discover the role of ESG in supply chains, from reducing carbon footprints to complying with new regulations and enhancing long-term business value.

Understanding ESG Risks in the Supply Chain

Sept. 26, 2024
Understand the critical role of ESG in supply chains, the risks for hiring companies, and the competitive edge suppliers gain by prioritizing sustainability.

Best Practices for Managing Subcontractor Risk

Sept. 26, 2024
Discover how to effectively manage subcontractor risk with unified strategies, enhanced oversight, and clear communication for consistent safety and compliance.

Voice your opinion!

To join the conversation, and become an exclusive member of EHS Today, create an account today!