3 Ways Big Data Factors Into the EHS Conversation

The potential of Big Data has factored into so many conversations of late. Between boosting profitability through enhanced analytical capabilities, to improving asset performance, to achieving superior operational intelligence, we’ve talked to the ends of the earth about the possible value Big Data presents.

Follow @PaulLeavoy

However, not a lot of ink has been spilled on Big Data’s potential impacts on environment, health and safety (EHS) performance. And yet, when we look at the fact that so much of what drives improved health and safety performance is analytics-driven, it’s clear Big Data can play a big role on this front.

With that, let’s look at the three key ways Big Data will begin to enter the EHS conversation.

1. Asset performance and EHS performance will become even more intrinsically linked

Yes, the linkages between Big Data and asset management have already been noted. But in many ways we have seen this manifest from an efficiency standpoint. For example, Big Data presents opportunities to help us know when our assets require maintenance, when they are at risk of failure, or when they are presenting undue energy impacts and costs.

But, as my colleague Dan Miklovic often states, “healthy assets are the foundation of a healthy business,” and this extends right up to health and safety management, where some of the greatest impacts on our people and shared environment take shape. We’ve often underscored the relationship between asset performance and health and safety consequences. For example, when an asset underperforms, that situation can lead to, in some cases, a gas leak or an explosion. That leak can result in the release of noxious gasses, and the explosion can often lead to compromised worker health and safety.

In the case of Pacific Gas and Electric (PG&E), this kind of situation came to a head with the San Bruno rupture of 2010. The utility company operates over 90,000 miles of pipeline in northern California, and if any inch of that pipeline is impacted by a rupture or dig in, the result can be an explosion or fire. In the San Bruno situation, this was the case, and eight people were killed with many more injured and dozens of homes destroyed.

“We didn’t have enough insight into what that data was telling us, in order to pinpoint that rupture to take action fast enough. It took almost 90 minutes to isolate the gas and shut it in,” noted Mel Christopher PG&E’s senior director of gas system operations in a recent video. “Our response was to turn data into intelligence so that can never happen again.” In the interim, PG&E has used OSIsoft’s PI system to gain actionable intelligence in order to ensure such a catastrophe does not occur again.

From an asset performance standpoint, we’ve often evaluated the health of an asset independently, without drawing many correlations between asset health and EHS performance. Further, we tend to manage and control these performance areas independently. But all these aspects are deeply related.

Big Data, leveraged and managed well, presents the opportunity for business to actually link asset performance and EHS performance effectively. It will not necessarily be an easy journey, but once we are able to effectively link asset and EHS performance data we will have established a more symbiotic and closed-loop system in terms of how we assess EHS performance.

2. Our capacity to proactively address EHS incidents will be more within reach

For years, we have been talking about the need to use ‘leading’ indicators as opposed to the ‘lagging’ indicators we have traditionally relied on. However, in practice, this has been hard to achieve.

Robust leading indicators, fueled by Big Data analytics, however, might position us to become more proactive about health and safety management.

Of course, as always, there’s a human aspect in all this. We can analyze data eternally, but we need timely, accurate data inputted into our systems if we are to achieve any value from that data at all. So with that, in order to get ahead of incidents, we need to ensure our people and processes are as effective as the data we are presenting, and that we are developing processes that respond to that data, and grooming our people accordingly.

For example, instead of simply measuring and analyzing lost time accidents and incidents after they happen, Big Data, coupled with improved observer behavior-based intelligence, will enable manufacturers to get in front of accidents and incidents before they occur.

3. The ‘Holy Grail’ of zero incidents might become achievable

There’s been a lot of talk and very little walk over the zero-incident ‘fable,’ as some say, in the EHS community in recent years. And personally, I believe some of it is true. Human error and human negligence, however much it can be controlled through cultural initiatives, will always remain a factor, just as process and technological failures will always present weaknesses.

However, in light of this, while the zero-incident notion seems like an unachievable dream, Big Data, well integrated with EHS and other enterprise management programs, puts this elusive dream more within reach. Cisco’s recent campaign on ‘firsts and lasts’ can be applied to the health and safety situation. It may be aspirational in scope, but the notion of achieving the ‘last safety incident,’ for example could be the centerpiece of any health and safety program. It may seem impossible to achieve, but if Tennyson’s notion that ‘a man’s reach ought to exceed his grasp, else what’s a heaven for?’ applies, then it stands to reason our health and safety efforts ought to strive for perfection. And Big Data can help us get closer to that ideal.

Firstly, we’ll be better equipped to analyze asset health and, accordingly, correlate the performance of assets with other operational EHS consequences. Secondly, on-the-field observations will be easier to capture (aided by the rise of mobile tools, wearable tech, and improved sensors for monitoring equipment and environmental factors) and trend, thereby helping us to get to the root of many EHS failures. Thirdly, these robust analytics will feed into alert and notification-based signals to help us proactively mitigate (as opposed to reactively respond to) adverse events.  

Sounds great, right? Well, we’re not there yet. While there has been a lot of progress in terms of linking assets and building the right tools to improve our understanding of EHS analytics, there is work to be done in terms of connecting all of these elements and building out a system that effectively aggregates and connects all of this once-dislocated data.

What do you think about the future of Big Data in EHS? Feel free to share your thoughts in the comments section below.