I read a discussion response to a post I put up on a LinkedIn risk management group.  The response said something to the effect of, "disasters are going to happen, nothing we can do other than cope with the impact".

Yes disasters do occur. The question is should they have? Were all of the prescribed procedures followed?

I call on the risk management world to give consideration to the fact that there were risks that could have been mitigated in each of the fatal industrial incidents that have occurred in the last several years. Investigations have shown that backlogs of maintenance, budget cuts to critical programs, and hasty decisions that overstepped procedural checks and balances were among numerous causative factors that led up to these fatal disasters. 

So who’s to blame?

The respondent also said, we would all expect those who are culpable of being negligent to be dealt with. So, in everyday terms, I could defer maintenance on my car and cut spending in its upkeep. Yet if I choose to get behind the wheel and my brakes fail, while driving through a school crossing and kill several people, there would be no question that I was guilty of reckless endangerment. I did not choose to hurt or kill anyone but my inaction and neglect were part of the disaster I caused.  What if it was a school bus that careened uncontrollably through the cross walk?  If it were determined that the county school bus fleet manager had been negligent in managing the brake testing, inspection, and repair for the bus fleet, there would be a public outcry and he/she would be indicted. 

Now the world of reliability engineering has developed numerous work processes that are focused on mitigating risk. Hazop, safety instrumented systems, RBI and numerous others exist to mitigate risk. Meridium works with companies who use our APM Software to PREDICT future failures of equipment so that they can make outage decisions in order to make preventive repairs and avoid disaster and death. According to Marsh, up to 20% of an organization's risk profile is determined by their ability to demonstrate in a timely and straight forward manner that they are managing their physical assets well and are mitigating risk. In other words those companies who can prove the number of “saves” they’ve made have a more favorable risk profile in the eyes of the underwriter.

The investigation in the Deepwater Horizon fatality showed that there was a backlog of over 390 maintenance items totaling more than 3,000 man-hours of maintenance that was overdue. A recent CNN news story reported allegations that PG&E used money that was allocated for pipeline inspection for executive compensation.

So I challenge the idea that “these things are going to happen… you can’t mitigate all of the risks completely”. No you can't but in each of these cases someone’s decision overstepped the checks and balances of established reliability best practices. 

I wonder every day where the next disaster is going to strike. I even have a few ideas. I'd be interested to know if people feel uncomfortable going to work knowing there is a real risk that something disastrous might happen?

OSHA Deputy Assistant Secretary Jordan Barab leveled a very direct judgment against petroleum and petrochemical processing companies at the NPRA conference last week.  He did not mince words.

"Your workers are dying on the job and it has to stop"

"Stop boasting about your safety records when you're literally putting out fires.  You're undermining your credibility"

He called on industry leaders to learn from their own mistakes and those of others.  He further called on them to share best practices and and success stories.  New metrics were also a topic of discussion since current metrics are not preventing death and disaster.  Mr. Barab concluded by urging companies to promote a real and lasting cultural transformation in the area of PSM. 

Can improving metrics and scorecards help avert disasters?  There is a need to systematize Process Safety Management from the initial HAZOP to the decommissioning of the plant.  What I have come across quite often in my daily conversations with reliability professionals is that too many of the efforts that impact Process Safety Management are run independently.  Risk based inspections are being done by one set of engineers.  Safety instrumented systems management is done by yet another group.  Inspection management is being carried out by a group that is not connected to either of these. 

One of the more troubling findings Mr. Barab refers to was that findings and concerns are not being communicated across corporations, throughout industry, or even within different units within the same refinery. 

The problem doesn't seem to be limited to the process industry.  This seems to appear in the medical industry, education, government, and heaven forbid, homeland security and intelligence.

 

There are 16 different US federal agencies that perform intelligence analytics. In the weeks and months prior to 9/11, parts of the system were blinking red. But the disparate red blinking lights did not coalesce and form a comprehensive warning that may have averted the loss of life and catastrophic events of that day. Even though parts of the system were blinking red no one was able to connect the dots between all of the available pieces of intelligence. 

(Enterprise 2.0: New Collaborative Tools for Your Organization's Toughest Challenges, Andrew McAfee)

This is a poignant example of what disconnected information and data can result in. If they had been able to connect the dots they may have been able to perceive the nature of the threat more clearly and prevent the attacks from occurring. 

With all due respect and sensitivity I use this example because it is hard to ignore. It demonstrates, through an unforgettable catastrophe, what occurs in the asset-intensive industry in terms of asset performance. Are there parts of the system that are blinking red in one department that are not signaling trouble to other departments?