Newsletter Sign-up

Occupational hazards abound at oil refineries, where workers can be injured or even killed if conditions are unsafe.

BP, one of the oil industry giants, responded to industry safety concerns by keeping track of an array of statistics in an effort to reduce the number of on-the-job injuries. Those statistics showed, for instance, that at BP the most frequent cause of workforce fatalities was driving-related incidents.

The very next year, however, BP suffered a massive and fatal accident at a Texas refinery that had nothing to do with driving. Fifteen people died and more than 170 were injured in an explosion and fire at BP's Texas City refinery on March 23, 2005. It was one of the most serious workplace disasters in the United States in nearly two decades.

It didn't stop there.

Four months later, on July 28, 2005, a hydrogen fire that had the potential to cause additional deaths and injuries broke out at the Texas City refinery, resulting in a Level 3 community alert.A Level 3 alert, which is the second highest emergency classification under Texas City refinery plant procedures, applies when an incident is not under control and protective action might be necessary for the surrounding or off-site area.

And then on August 10, 2005, the refinery experienced another Level 3 incident involving the gas oil hydrotreater. This incident resulted in a community order to stay indoors.

What happened?

Although BP had achieved significant improvement in its personal safety performance, those statistics didn't tell the whole story, according to a January 2007 report of the BP U.S. Refineries Independent Safety Review Panel.

Worse, they created a false sense of confidence about the company's overall safety performance.

The 11-member panel was formed by BP at the recommendation of the U.S. Chemical Safety Board and was headed by former Secretary of State James A. Baker III.

The panel conducted a detailed review of the company's corporate safety culture, safety management systems and corporate safety oversight at its U.S. refineries and issued a report that spans more than 300 pages.

The panel's report found that, in addition to putting too much emphasis on personal injury statistics, BP also failed to develop a corporate safety culture that put a priority on process safety. The company lacked effective process safety leadership and lacked a common, unifying process safety culture.

Steve Arendt, vice president of organizational performance assurance with Houston-based risk management services provider ABS Consulting, worked on the Baker panel report and has met with about 50 companies in the last nine months to help evaluate their corporate safety culture.

He says that, over time, many companies become blind to their own shortfalls.

"It's not so much that people are ill intentioned, but with the busyness of business, it gets off their radar screen," Arendt says. "They lose situational awareness, their risk nerves get calloused over a little bit," he says.

A critical review of their process safety management practices can act as "a bit of sandpaper to sandpaper the calluses off their nerves," he says.

"We can have blind spots, and sometimes you have to have somebody from the outside to take a look at you to help show you the blind spots," he says. "Sometimes companies refuse to look in the mirror. Sometimes they act as though it's a carnival mirror. In other words, it makes them look fat and they don't think they are fat," he says.

One of BP's blind spots was its reliance on occupational illness and injury rates, which limited its perception of process risk within the U.S. refineries.

Personal safety statistics, unlike process safety statistics, are widely tracked, benchmarked and reported, according to the report. BP tracked metrics such as fatalities, days away from work, case frequencies and recordable injury rates. But these metrics bear little relation to the state of process safety, the report said.

"BP interpreted improving injury statistics, which are widely tracked in the industry, as an indication of acceptable process safety performance at its U.S. refineries," the panel said in its report.

"BP's reliance on this data and its inadequate process safety understanding created a false sense of confidence that BP was properly addressing process safety risks at these refineries," the report said.

The panel also found that:

- BP did not provide effective process safety leadership in making certain its management and U.S. refining workforce understood what was expected of them regarding process safety performance.

- Process safety leadership appeared to have suffered as a result of high turnover of refinery plant managers.

- At some of the refineries, BP had not established a positive, trusting and open environment with effective lines of communication between management and the workforce.

- BP had not always ensured that it identified and provided the resources required for strong process safety performance, including both financial and human resources.

- BP also did not effectively incorporate process safety considerations into management decision-making that affected U.S. refineries.

- Workers held the widespread view that profit came before safety.

In the report, the panel repeatedly emphasized the difference between personal safety and process safety.

Personal or occupational safety hazards give rise to incidents--such as slips, falls and vehicle accidents--that primarily affect one individual worker for each occurrence, according to the report. Process safety hazards, on the other hand, can give rise to major accidents involving the release of potentially dangerous materials, fires and explosions.

These major incidents may not happen very often, but when they do happen, they can have widespread repercussions for the company and the surrounding community as well.

These incidents can result in injuries and fatalities, as well as widespread property and environmental damage. A major incident also could lead to long and costly litigation, the loss of reputation and significant financial losses.

BP has set aside $1.6 billion to compensate victims and was hit with a $21.3 million fine by the Occupational Safety and Health Administration regarding the accident.

LESSONS PROCESSED

To reduce the risk of a major loss, companies in the chemical, pharmaceutical, and oil and gas industries should practice process safety management.

Process safety management got its start in the mid-1980s in response to the infamous Bhopal disaster in India. In December 1984, a Union Carbide subsidiary's pesticide plant released 40 tons of methyl isocyanate gas, immediately killing nearly 3,000 people and ultimately causing at least 15,000 to 22,000 total deaths. It is frequently cited as one of the world's worst industrial disasters.

Although process safety management has been around for about two decades, there is still room for improvement, especially in the oil refinery industry, according to some experts.

In June, OSHA announced it was launching a national program to determine if refineries were complying with its process safety management standard. As part of this program, OSHA says it will inspect 81 refineries over the next two years.

OSHA said that the program was prompted by the large number of incidents involving catastrophic releases of highly hazardous chemicals at refineries.

Since the process safety management standard was promulgated by OSHA in 1992, "no other industry sector has had as many fatal or catastrophic incidents related to the release of (highly hazardous chemicals) as the petroleum refining industry," OSHA said in a directive.

In its report, the Baker panel said it was under no illusion that the deficiencies in process safety culture, management or corporate oversight identified in its report were limited to BP.

The panel said it believes all companies in the refining, chemical and other process industries should give serious consideration to its recommendations.

"You want your top management to really determine (process safety management) is something that is important and set priorities, and the good programs and good resources flow from there," says Scott Berger, director of the Center for Chemical Process Safety.

"Even if companies have always been champions of process safety, they should go look again and just make sure that they haven't perhaps assumed that, since no accident has happened recently, that everything is fine," Berger says.

Many companies know they need to practice process safety management and understand that it can help reduce the risk of a serious incident, says Dave Seaman, vice president and engineering manager at FM Global's chemical operations.

"Most of (our customers) are at the point at which they realize that it makes sense to practice PSM and to have a culture at a plant from the operating level to the plant manager--that it makes sense from a business standpoint," Seaman says.

"If you practice PSM, not only will you have fewer incidents, you will have fewer releases, your name will be in the press a lot less," he says. "To do that, it means the reliability of your operations has to improve; so it is good for business," he says.

Even so, many companies still need to do a better job on process safety, Arendt says. In response to the Baker panel report, many companies are now taking a hard look at their process safety management practices, he says. As part of this self-examination, a number of companies have asked Arendt to visit their facilities and perform evaluations.

"A lot of companies are trying to figure out how to assimilate these lessons," he says.

Although most companies are familiar with process safety management, their performance can stagnate after awhile, Arendt says.

To get a better understanding of their process safety hazards, some companies are now developing their own metrics for measuring process safety. Traditional safety metrics, such as the number of accidents and injuries, are lagging indicators.

Those indicators provide information about injuries from incidents that have already happened. They are not all that useful, however, when it comes to large-scale incidents.

"What we are trying to do is prevent a rare event from occurring," Arendt says. "You can't measure the rate at which something happens that almost never happens," he says.

Leading indicators, therefore, help to provide information about how well a process safety management system is performing. Some of these leading indicators attempt to measure things such as the:

- incident rate

- near-miss rate

- rate at which there are unsafe acts or a failure to follow procedures

- rate at which a company fails to fix problems they know about

- rate at which the company fails to identify problems--in other words, an inspection backlog

- failure to adequately assess the risk

The chemical and refining industries have found religion, for now, Arendt says. But he's not sure how long it will last.

"Everyone right now is serious about it," Arendt says. "Over time, the urgency and the religion that companies get can decay away," he says. "It will be interesting to see what happens in two or three years."

BP, for its part, issued a statement in January saying it will implement the recommendations made by the panel as part of the company's continuing effort to improve its safety culture and to strengthen and standardize process safety management at BP's five U.S. refineries.

BP has already undertaken a number of measures, including dedicating significant resources and personnel, to improve the process safety performance at the BP refineries. The company also has said it will develop plans for additional action at its U.S. refineries and for applying lessons learned elsewhere.

As the Texas City incident shows, however, it is easy for companies to become complacent and to believe that they are doing a good job on safety just because their occupational injury rates are improving.

"If there's a lesson there, it's that you can't rest, you have to always be vigilant," Berger says.

PATRICIA VOWINKEL lives in New Jersey.

December 1, 2007

Copyright 2007© LRP Publications