Forensic, General & Medical
Expert Witnesses

A Troubling Process Industry Trend

     By Samuel Woinsky Consulting Petroleum, Petrochemical and Chemical Industry Expert Witness

PhoneCall Samuel Woinsky at (281) 419-0325

This trend has increased risks and caused process plant explosions.
04th July 2009

A Troubling Trend: When I started working as a chemical engineer in the 1950s, the world was a very different place. Large petroleum and chemical companies had their own maintenance and R&D departments. Typically scheduled maintenance shutdowns were spaced a year apart. Companies weren’t risk adverse relative to building plants based on their own R&D. If you were a good worker at an operating company (not so at an engineering construction firm), you had a job for life, if you wanted it. Contrast that with today’s situation where scheduled maintenance is typically done on a 5 year cycle using contract personnel who don’t know the plant as well as company employees used to. R&D departments are gone altogether or a shadow of their former selves. Now, companies typically look at building plants based on new technology created by others but with a “not me first” attitude. In the early 1980s, downsizing started and now jobs are only secure when business is good. The loss of older long-term employees resulted in a loss of corporate memory (what works and what doesn’t) and an erosion of corporate culture.

Some Results of This Trend: In October 1989 an explosion at a Phillips Petroleum plastics plant in Pasadena, TX killed 23. The blame was attributed to primary contractor or owner failure to carry out their responsibilities at the multi-employer worksite.

In July 1990 an explosion at an Arco Chemical plant in Channelview, TX killed 17 and injured 5. This explosion was caused by contract maintenance workers.

In October 1995 an explosion at a Rouseville, PA Pennzoil Refinery’s tank storage area killed 5 and caused extensive damage to the plant. The ignition source for this explosion was welding during maintenance being conducted on a tank’s stairway. The tanks were nearly 60 years old and were “grandfathered” so they didn’t have new construction’s typical pressure relief devices or roof seams designed to fail first in an explosion. This allowed tanks to fail at their bottom seams releasing flammable hydrocarbon liquid in addition to vapor. The problem was compounded further by the fact that many of Pennzoil’s tanks were not individually diked. This allowed burning liquid to quickly spread over a wide area.

In March 2005 an explosion at an isomerization unit in a BP refinery in Texas City, TX killed 15 and injured more than 170. This explosion was caused by a number of failures and might have been avoided or mitigated if just one less failure had taken place. The operators were starting up the raffinate splitter and started feeding it highly flammable hydrocarbon liquid for more than a day without beginning discharge of the product. Liquid was overfilling the splitter but instruments that would have warned the operators were inoperable since open maintenance orders hadn’t been completed. The splitter’s safety relief valve eventually opened but was tied to a blow-down drum feeding an atmospheric vent stack instead of the safer knock-out tank and flare system that is typically used for new construction. Vapor and liquid came out of the vent stack and engulfed a pick-up truck owned by a contractor. The owner started up the truck in an attempt to move it out of harm’s way providing the ignition source for the explosion. People in a trailer located too close to the unit were killed or injured. A panel led by former US Secretary of State James Baker III investigated the safety culture and management systems at BP North America. The principle finding was that the metrics, incentives and management system at BP focused on measuring and managing occupational safety while ignoring process safety (i.e. design for safety, hazard analysis, material verification, equipment maintenance, process upset reporting, etc.).

What’s Driving This Trend: The driver is economic analysis that doesn’t account for intangible costs. Downsizing cut payrolls with no accounting for the value of lost corporate memory and erosion of the corporate culture (admittedly things difficult to put a value on). Increasing the time between scheduled maintenances increases operating hours per year and annual production without using scarce capital, but with a higher risk of accidents. Outsourcing maintenance saves on labor costs between scheduled maintenances, but at an even greater risk of accidents. Since these two trends have been taking place simultaneously, one should be able to get some idea of the combined intangible cost by looking at the frequency and magnitude of plant accidents vs time from 1950 to today. With regard to specific petroleum refinery/petrochemical plant accidents, one has to recognize that prior to the run-up of gasoline prices, petroleum refineries had very small profit margins. Staff had been cut to the bone and capital for projects was scarce. The only large capital projects being done were to meet mandated environmental requirements, to increase capacity and increase the value of the product mix. Coming from that type of culture, it isn’t surprising that over the period from 1991 through 2002, first Amoco and then the 1999 merged BP/Amoco passed 6 times on upgrading to a flare system. One of those times the cost was only $150,000 vs the $2.125 billion BP set aside to settle claims from the 2005 explosion (see BP’s 2007 annual report). Clearly BP was penny wise and pound foolish. They could have looked back at the Phillips, Pennzoil and Arco explosions to get some idea of what the intangible liability cost might be. Also, there are intangible public relation costs (lost sales, difficulty finding sites for new plants, difficulty getting plant additions approved). However, the largest intangible cost is the liability cost and that is the easiest one to put a price tag on.

Conclusion: Those that don’t learn from history are bound to repeat the mistakes of the past. Hopefully, corporations will realize the importance of including intangible costs in their economic analysis methods.

P.S. BP’s 1st quarter profits were $0.617 billion in 2000 (5 quarters after BP merged with Amoco starting at the beginning of 1999), $4.361 billion in 2007 (the year it set aside $2.125 billion for its 2005 explosion), $6.588 billion in 2008 and $2.387 billion in 2009. BP purchased Solarex (a solar energy company) in 1999 and started a rebranding campaign in 2000 as a “green” company (with the aim of raising retail fossil fuel earnings by over 10 % per year). By the end of 2002 BP had spent about $0.952 billion on the rebranding marketing campaign ($0.607 billion) and solar investments ($0.345 billion). BP solar started showing a profit in 2007.

ABOUT THE AUTHOR: Samuel Woinsky
Samuel Woinsky is Proprietor of Samuel Woinsky Consulting. Expertise: Chemical Engineering, Cost Analysis and Economic Evaluation, Licensing and Technology Transfer, Technical Sales Assistance, Cost Reduction Analysis, Energy and Water Conservation, Waste Minimization, Alternate Fuels, Process Troubleshooting and Debottlenecking, Process Engineering, Technology Development and Commercialization, Economic Analysis of New Technologies and R&D Programs, Expert Witness

In these difficult times I will be focusing on the following.

1. Technical sales assistance to help you increase your sales

2. Cost reduction analysis to help you reduce your operating costs (with little or no capital cost)

3. Assistance to your lawyers to help you win your case (studies, reports, depositions, expert testimony)

Copyright Samuel Woinsky Consulting

More information about Samuel Woinsky Consulting

While every effort has been made to ensure the accuracy of this publication, it is not intended to provide legal advice as individual situations will differ and should be discussed with an expert and/or lawyer.
For specific technical or legal advice on the information provided and related topics, please contact the author.

Find an Expert Witness