Feds Want Stricter Oversight of California Oil Refineries . . . by Oil Companies

Friday, December 20, 2013

A U.S. regulatory agency has responded to the devastating oil refinery fire in Richmond last year by declaring a national “refinery safety problem” and proposing a new, more “rigorous” way of preventing dangerous problems.

The U.S. Chemical Safety Board (CSB) issued a draft report this week that recommends the oil refinery industry shift from the conventional regulatory oversight system to one called “safety case,” an internationally popular concept being adopted in a range of industries with safety-critical systems.

It is a fan-favorite of industries that want a more streamlined regulatory system, but it has its critics among those who think it doesn’t work and is window-dressing for just letting companies regulate themselves. Although “safety case” means different things to different people in different industries, it commonly emphasizes that regulation be based on setting goals with stated objectives, instead of having prescribed measures imposed with an ongoing inspection regime. 

A plan is devised, executed and monitored by the company, and regulators check the plan. Great Britain and the U.S. both became interested in the concept when Ronald Reagan was president, but Britain has been more aggressive about using it. 

The CSB report was triggered by an incident at the Chevron Bay Area refinery, where a corroded pipe gave way, releasing gas that exploded in a fireball. The city was engulfed in choking black smoke and potentially dangerous sulfuric acid and nitrogen dioxide fumes that sent 15,000 people to the hospital.

Reuters quoted CSB Chairman Rafael Moure-Eraso as sounding liked he blamed the fire on regulatory requirements. “Refinery safety rules need to focus on driving down risk to the lowest practicable level, rather than completing required paperwork.”

If only there hadn’t been so much paperwork, Chevron wouldn’t have been handed 25 citations and fined $963,200 by the state after the California Division of Occupational Safety and Health (Cal/OSHA) said the company had heard repeated warnings, many of them internally, about “pervasive violations.”

Twenty-three of the 25 violations were considered “serious” by the state agency, including Chevron’s failure to institute emergency procedures to shut down the unit, the ordering of workers into the “hazardous incident zone” without proper equipment, and failing to recognize the “potential for catastrophic release of ignitable diesel fuel.”

CSB Western region head Don Holmstrom told the San Francisco Chronicle that “safety case” could have prevented the accident last year because the oil company would have had to identify all hazards and replace corroding pipe.

Not everyone might agree with that.

A white paper (pdf) by Nancy Leveson, a professor of aeronautics and astronautics at MIT, argued “there is no real evidence that one type of regulation is better than another.” The evidence of its success anywhere is anecdotal and untestable.

Leveson warned that safety cases don’t work after a design is completed (like on an existing refinery); they need to be created during design. Otherwise, “the effort will not improve safety and becomes, as apparently has happened in the past, simply paper exercises to get a system certified.”

Leveson cited the military’s experience with “safety case” and the 2006 crash of a British Nimrod XV230 that killed 16 servicemen in Afghanistan. A review of the crash concluded that “safety case” procedures “led to a culture of ‘paper safety’ at the expense of real safety.” The report complained that, “Safety cases were intended to be an aid to thinking about risk but they have become an end in themselves.”

The review said large amounts of money were spent on things that did not improve safety and concluded, “Such safety cases tend also to give the answer that the customer or designer wants, i.e. that the platform is safe.”

Leveson took issue with the very concept of defining a safe system that borrows mathematical precepts applied to closed theoretical systems. Engineered systems, like refineries, are open, so the safety analysis is being applied “where there is no complete mathematical theory to base arguments and guarantee completeness.”

“The problem simply devolves to defining what is ‘acceptable’ and to whom: to the producer of the system who is paying the cost of making it safe or to the potential victim?”

The U.S. Chemical Board is just an advisory agency with no powers to enforce its will. A California state interagency working group recommended in July that Governor Jerry Brown adopt policies similar to those recommended by the board, according to Reuters.

–Ken Broder


To Learn More:

U.S. Safety Board Urges New Rules for California Refineries (by Erwin Seba, Reuters)

Feds Want Tougher Refinery Oversight in California (Associated Press)

Dramatic Safety Overhaul Sought after Chevron Fire (by Jaxon Van Derbeken, San Francisco Chronicle)

Federal Report Calls for Sweeping Reforms in Aftermath of Chevron Richmond Fire (by Robert Rogers, Contra Costa Times)

Richmond Protests and Sues Chevron on Anniversary of Devastating Refinery Fire (by Ken Broder, AllGov California)

White Paper on the Use of Safety Cases in Certification and Regulation (by Nancy Leveson, Aeronautics and Astronautics/Engineering Systems at Massachusetts Institute of Technology) (pdf)

Leave a comment