Keywords: Texas refinery, explosion, failures, report, BP,
  • Print
Latest Comments

There are no comments for this news.

See comments for this news

Related Discussion

There are no related discussions.

Start new discussion

Go to discussion forum home

US report lists BP safety failures from top to bottom

News | HSW
01.03.2007

Patchy accident reporting, rushed training and a focus on individual safety - such as reducing slips and trips - at the expense of major hazards all contributed to process safety failures that led to the 2005 explosion at BP Amoco's Texas City refinery, which killed 15 people and injured 180 (see below). These are the findings of the independent panel headed by the former US secretary of state James Baker, convened by BP after the explosion at the request of the US Chemical Safety Board (CSB).

The panel's report (available at www.safetyreviewpanel.com/cmtdoc/Panel%20Report%20-%20January%202007.pdf) is a catalogue of failures in BP's process safety management and procedures from the bottom to the top. It criticises lack of leadership by senior management, poor hazard analysis, excessive overtime working, "initiative overload" and a lack of communication between managers and workers at the US operations.

The panel researched safety procedures and conditions at BP's five US refineries: Carson, California; Cherry Point, Washington; Texas City, Texas; Toledo, Ohio; and Whiting, Indiana. It reviewed documentation, interviewed managers and surveyed the plants' workforces.

It says its findings are aimed at BP but urges other companies to take them on board and review their process safety mechanisms and work cultures.


Failed inspection

One of the clearest safety failings the panel finds is that staff did not keep to maintenance and inspection schedules. It points to various BP audits, including one of the Carson plant in 2005 which found that 134 pressure-relief valves and 46 pressure vessels were past their due dates for inspection, and a random sampling of 10 turbines which found only one had a record of being tested for over-speed tripping.

The panel makes a special point about rupture discs (a deliberate failure mechanism intended to relieve excessive pressure in a system, preventing damage elsewhere). Where the discs are used in front of relief valves they need regular checking to see that they have not been compromised by small leaks which equalise pressure either side, making them ineffective. The panel's technical consultants checked refinery inspection logs and found many instances of discs recorded as compromised without timely follow-up action.

At four out of the five refineries, around 20% of the engineers disagreed with the statement that BP regularly maintained safety-related devices such as interlocks, and as many as 40% of operators said the same at one site.

Hazard analysis was inconsistent from site to site. At Toledo and Whiting, the process hazard analysis programmes for highlighting design and process safety weaknesses did not take account of process failures other than during normal operation (at start-up, for instance).

The panel found accident reporting systems were under-used for various reasons, including "fear of being blamed or retaliated against - or that no corrective action would be taken so reporting would not be of any value".

Focusing on training and development, the panel finds BP had no set safety qualifications or competences for its refinery plant managers (the most senior line managers) and did not define the level of process safety knowledge or competence required of executive management or line management above the refinery level. It relied on prior experience and on-the-job learning rather than specific process safety training for new managers.


A little knowledge

Conversely, below management level, the panel found inadequate training again, but highlighted the lack of mentoring and drills where workers are asked to go through their responses to a hypothetical emergency, plus patchy refresher training in process safety. At most sites, training regimes relied too much on computerised modules for safety awareness.

The panel interviewed a sample of new recruits who expressed concern that they were trained by inexperienced supervisors (a view backed up by training coordinators). There was no evidence that supervisors needed familiarity with the equipment in their units before promotion to that grade. Operators complained training was rushed and significant minorities of non-managerial staff at many levels felt they lacked training in hazard identification, control or reporting.

Workers were burdened with too many management programmes on issues such as climate change, leading to "initiative overload" and distracting them from process safety. They also felt pressured to work excessive levels of overtime, with operators averaging up to 30% on top of contracted hours at some sites, increasing safety risks.

The lack of a safety culture at the Texas City plant was highlighted in BP's own report on the incident which admitted work conditions had eroded to the point that safety procedures were not followed, managers failed to support proper safe operations and workers felt "disempowered" to push for improvements because of a lack of leadership from managers.

But the Baker panel says the lack of grip goes right to the highest levels of the organisation. "BP does not have a designated high-ranking leader for process safety dedicated to its refining business," says the report.

The panel's finding that BP prioritised occupational health and accident reduction initiatives over more critical hazards echoes the preliminary findings of the CSB's own inquiry. Launching its interim findings in October, CSB chair Carolyn W Merritt said BP's global management had responded to problems at the refineries at Texas City before March 2005 with a variety of safety measures: "However, the focus of many of these initiatives was on improving procedural compliance and reducing occupational injury rates, while catastrophic safety risks remained."

But while the Baker panel's report says clearly that BP's training and inspection regimes suffered from under-investment and short staffing, it fights shy of direct accusations of cost cutting as a root cause of the safety failures that led to the Texas City incident.

The CSB is not expected to follow the panel's lead in its final report due next month. In October, Merritt said BP "implemented a 25% cut on fixed costs from 1998 to 2000 that adversely impacted maintenance expenditures and infrastructure at the refinery."

BP has said it will act on the Baker panel's findings.


The Texas City incident

At 1.20pm on 23 March 2005, there was a sudden, geyser-like release of flammable hydrocarbon liquid and vapour from an atmospheric vent stack at the isomerisation unit at BP's Texas City refinery. The release created a flammable vapour cloud that was ignited by an unknown source, causing as many as five explosions, which killed or injured workers in temporary trailers between 30m and 45m from the vent stack. The Chemical Safety Board's investigation found that alarms that should have warned operators of abnormal conditions in the isomerisation unit did not go off.

Before the incident, BP had been placed on the US Occupational Health and Safety Administration's (OSHA) watch list for another fatal explosion at the refinery the previous September, in which two pipe-fitters died. OSHA fined the oil giant million (£10.67 million) for the March 2005 incident, the regulator's biggest ever penalty, and has not ruled out a criminal prosecution.

  Your Comments No Comments
Related Comments & Analysis

Leeds Hospitals' burning issue

Published:06.03.2008

Portable IT equipment and combustible materials

Published:09.06.2008

Competent for a fire risk assessment

Published:01.08.2006

Fire and emergencies

Published:01.10.2005

Escape routes

Published:01.05.2005

See all comment & analysis

Latest Discussion

The latest discussion & debate from the healthandsafetyprofessional.co.uk discussion forums...

The HSE and road safety

On page 3 of the December issue of the IIRSM newsletter was an article headed "HSE Could be more involved". ...

John Hill | Nov 20 2008 10:14AM

HSE - Work Related Road Deaths

Does the HSE have the correct view on Work Related Road Death ... quote : 'work related road traffic accidents ...

John Fitton | Oct 09 2008 10:46AM

Looking for a cancer community!!!

Hello everybody, I am a college student and I am here for some help from you guys. I live with ...

Herry smith | Oct 07 2008 07:08AM

Start new discussion

Go to discussion forum home