THE BAKERRISK BULLETIN

THE BAKERRISK BULLETIN The Newsletter of Baker Engineering and Risk Consultants, Inc. Volume 15 Issue 1 January 2009 MAINTAINING SAFETY DURING A RECESSION by Quentin A. Baker, President With an uncertain economic climate facing all of us this year, we My safety message for oil and chemical are reminded that maintaining safety practices during a recession will provide safety benefits that can far outweigh the companies is clear: even during economic consequences of an incident. Unfortunately, there is often not a downturns, spending for needed process convenient cost argument for safety benefits because good safety measures must be maintained. results the absence of incidents does not provide a measurable benefit. Moreover, cutbacks that directly affect John S. Bresland, CSB Chairman major risks may have a delayed effect, possibly on a time scale of years. In a news release* issued December 22, 2008, the chairman of the Chemical Safety Board, John S. Bresland, noted that corporate spending decisions can directly impact plant safety when budgets are cut for preventive maintenance, training, and staffing in efforts by companies to economize their operations. Chairman Bresland also mentioned that downturns and recessions can actually be a good time to take care of deferred maintenance because there s less financial impact from temporarily shutting down a process during periods when sales are depressed. Of course, the business reality may limit activities that can be conducted during a recession. Nonetheless, safety should continue to remain a top priority for all of us, regardless of economic conditions. *Editor s Note: to view the entire news release, visit the CSB website (www.chemsafety.gov) INDUSTRY AND REGULATORY TRENDS IN FACILITY SITING by Quentin A. Baker, President Several trends have emerged in industry and regulatory practices with regard to the siting of buildings at chemical processing facilities that present potential explosion, flammable, or toxic hazards. These trends are a result of a number of factors including: Several major accidents that resulted in fatalities with increased visibility to the public, Recommendations from the U.S. Chemical Safety and Hazards Investigation Board regarding siting of portable buildings, Publication of American Petroleum Institute (API) Recommended Practices RP753 for portable buildings, Management of Hazards Associated with Location of Process Plant Portable Buildings, Occupational Health and Safety Administration s (OSHA) National Emphasis Program (NEP) audits, and Pending revision of API Recommended Practice RP752 for permanent buildings, Management of Hazards Associated with Location of Process Plant Buildings. In this Issue Industry Trends in Facility Siting... 1 Natural Hazards Risks and Decision Making... 2 2009 High Pressure JIP... 4 Risk Management / QRA... 6 (continued on page 5) www.bakerrisk.com Page 1

NATURAL HAZARDS RISKS AND DECISION MAKING by Charles A. Pacella, Principal Consultant The most powerful, and often most unpredictable forces on earth are driven by nature, and the hazards they pose can be devastating to businesses in every industry. Many companies still may not fully understand the potential impact to their business and fail to address the risks associated with natural hazards. Historically, according to the U.S. Geological Survey, one in four businesses impacted by a natural disaster do not recover, and insurance coverage for those losses is typically less than 20% of the total loss. E arthquakes, hurricanes, tornadoes, hail, and flood can all have a profound affect on a business. Still, many companies do not fully understand the risks or how they can be managed. A company s ability to manage the risks associated with natural hazards can make the difference between success in achieving strategic objectives and business failure. Although it is impossible to predict natural disasters, companies can minimize their impact by first analyzing and modeling how an event at a single facility can ripple through the entire supply chain, or how a single event can impact multiple components or locations. Once the risks and their potential impacts are qualified and quantified, it becomes possible to develop a sound business plan and process for mitigating such exposures and applying the appropriate operational, financial, and strategic risk management strategies. Prior to the 1980 s, only actuarial techniques were available to estimate potential losses. These were based on historic frequency and intensity data. Unfortunately, such data were limited and covered only a relatively brief period of time. The resulting single point loss estimates were of limited and questionable use. In short, they did not provide a good basis for decision making. Advancements in computer technology over the last two decades have allowed development of sophisticated Probabilistic Risk Assessment (PRA) models. These models are an integration of expert knowledge, experience, and technology from a wide range of engineering, geophysical sciences, and mathematics. They allow for hundreds of thousands of simulations or scenarios where results populate a virtual or stochastic universe with natural hazard event frequencies and severities. Perhaps the single most important output of these PRA models is the Exceedance Probability (EP) curve. Such a curve puts financial exposure into a context that financial and risk managers can readily understand and use. (continued on following page) NEW CHICAGO OFFICE BakerRisk is pleased to announce our continued growth with the addition of a new office in Chicago. The office is conveniently located northwest of downtown in the Citicorp Plaza Building, east of O Hare Airport on the Blue line. BakerRisk Chicago Office 8430 West Bryn Mawr, Suite 720 Chicago, IL 60631 Tel. (708) 667-5550 www.bakerrisk.com Page 2

(continued from previous page) For example, the typical EP curve shown in Figure 1 gives the relationship between the loss and the probability of exceeding that loss for a given portfolio. The same curve can be flipped as shown in Figure 2 to show the relationship between the loss levels and their mean return periods. (Mean return period is the average time interval between reoccurrence of the same magnitude of event). Additionally, this same curve can be expanded to show longer or shorter return periods, as shown in Figure 3. Annual Exceedance Probability 20% 18% 16% 14% 12% 10% 8% 6% 4% 2% 0% $0 $50 $100 $150 $200 $250 $300 $350 $400 Projected Loss Level (USD$ million) Figure 1: Typical Exceedance Probability Curve These types of output and others are invaluable input to management decision making. In addition to the big picture corporate vulnerability assessments, these results are instrumental in: Estimating property damage potential How large is your loss likely to be, and what are the chances of incurring such a loss? Estimating business interruption loss potential Where are your supply chain and operations most vulnerable and how long would key facilities and operations be affected? Projected Loss Level (USD$ million) $400 $350 $300 $250 $200 $150 $100 $50 $0 0 10,000 20,000 30,000 40,000 50,000 60,000 70,000 80,000 90,000 100,000 Mean Return Period (years) Figure 2: Exceedance Probability Curve showing relationship between loss levels and their mean return periods Evaluating cost effectiveness of risk management strategies Does the probability of loss warrant additional coverage, operational or production changes, engineering redesigns, or new financial risk transfer strategies? Developing emergency response plans What is your strategy if a facility is impacted by a natural hazard event? Developing business continuity response plans What must you consider if operations are impacted, or if more than one facility is subject to more than one loss event? Projected Loss Level (USD$ million) $400 $350 $300 $250 $200 $150 $100 $50 $0 0 100 200 300 400 500 600 700 800 900 1,000 Mean Return Period (years) Figure 3: Typical Exceedance Probability Curve showing shorter return periods Gathering the right information is critical for conducting these analyses. Through a detailed analysis, risks and their impacts can be better quantified and qualified, providing essential components for making effective business decisions. Although we cannot prevent natural disasters from occurring, gaining understanding of the exposures to your business can help establish strategies to manage these risks. www.bakerrisk.com Page 3

HIGH PRESSURE TESTING SAFETY RESEARCH 2009 JOINT INDUSTRY PROGRAM by Jeff Baker, Business Development BakerRisk s Joint Industry Program (JIP) for High Pressure Testing Safety Research begins its fourth year in 2009. The JIP evaluates pressure-testing hazards that are common to oilfield equipment companies. The following three studies are planned for the 2009 program year: 1. Testing of barriers with energy absorbing material sandwiched between steel plates, 2. A survey of best practices for pressure testing procedures, and 3. The construction of a water tank for underwater high pressure gas testing. The JIP helps interested companies better understand and predict hazards in their pressure test facilities, learn how to effectively improve testing safety, and mitigate test fragment hazards with effective protective structures. The focus is on safety issues common to the industry, with the results, benefits, and costs shared among all Participants. Participants of the 2008 JIP were: Baker Hughes, Cameron, FMC Technologies, Halliburton, Schlumberger, Smith International, Weatherford, WellDynamics, and the Wood Pressure Control Group. The 2006 and 2007 test programs focused on hydrostatic testing, including a range of tool fittings, test pressures (up to 30,000 psi), test temperatures (up to 450 ºF), and test barriers and shields (including various barriers and shields of interest to the Participants). The 2008 test program focused on similar test conditions with high pressure gas testing. Test results were documented in a technical report, including Hydrostatic Tool Failure Launching a 13-lb Bull Plug comparisons to current test cell design methodologies and calculations. Many of the Participants use the JIP data (high speed test videos and photos) in training programs for managers and technicians involved in pressure testing. Most pressure testing personnel have never seen a test failure under pressure conditions and do not appreciate the potential severe consequences of such an event. Safety awareness has improved significantly after participation in training programs using the test videos. An annual JIP meeting is held to review the JIP testing results in which the JIP Participants and BakerRisk determine the testing parameters and conditions that will be run for the following year. Based upon the results of the 2006-2008 Programs, and the requirements for pressure testing (including increases in the number of tests, testing locations, test pressures and temperatures), the interest in this JIP continues to grow. www.bakerrisk.com Page 4

(continued from page 1 Industry Trends) Release Conditions for Consequence Based Analysis There is a growing movement to consider larger release sizes than were common practice when siting studies were first performed over a decade ago for facilities using a consequence-based approach. A consequence-based approach uses damage or injury as the end point, irrespective of the probability that the postulated event will occur. For example, API RP753 states that Consequence analysis should be based on major release scenarios, considering incidents and their outcomes that have or could have occurred in similar process units in the industry (i.e., scenarios with the most severe consequence). Unfortunately, no current or pending U.S. Government or industry criteria exists regarding the size of release that should be evaluated, leaving owners/operators to select their own criteria. We suggest reviewing release scenarios in light of RP753 and the pending revision of RP752 when it is released. Also, companies may want to consider a risk-based approach in which a wide range of scenarios are evaluated with the appropriate incident frequency assigned to each. Occupancy Criteria Screening of buildings from a siting evaluation has become more stringent. API RP753 adopted a functional definition of occupancy and disallowed screening based on the example occupancy criteria (man-hours) in the current edition of API RP752. According to RP753, if a portable building is intended for occupancy, it is considered occupied regardless of occupancy level. In addition, OSHA issued an NEP audit guideline disallowing the API RP752 example occupancy criteria. Many companies adopted standards based on that criteria, and now face the prospect that OSHA considers buildings occupied that had previously been screened out as unoccupied. Identification of essential personnel and the location of non-essential personnel is another recent trend. RP753 introduced the concept of essential personnel in an industry recommended practice. Non-essential personnel are not allowed in portable buildings in a zone that is close to process units. A guiding principle of RP753 is: Locating personnel away from process areas (is) consistent with safe and effective operations. The essential personnel approach is one means of enforcing this guiding principle. OSHA Citations A review of recent OSHA citations illustrates the agency s position on facility siting, including: Failure to perform a siting study that addressed all occupied buildings, Failure to mitigate risks identified in existing siting studies, and Failure to address all of the hazards associated with fire, explosion, and toxic releases. The OSHA NEP audits have been limited to refineries, but OSHA plans to expand this program to other processing facilities in the near future, including facilities that process combustible solids. OSHA inspectors have demonstrated high interest in buildings located inside of process units and questioned continued occupancy of conventional buildings that are not specifically blast hardened in these locations. The trend is clearly towards less tolerance of occupied conventional buildings inside process units by both government agencies and industry groups. Pending Changes API RP752 is currently in revision and the 3rd edition will be published in mid-2009. BakerRisk is participating on the API Task Force. The 3rd edition will be a major revision, and chemical processing companies should be prepared to obtain and review it as soon as it is released. In light of these trends, we encourage facility owner/operators to review their facility siting studies and evaluate whether they meet current and anticipated industry guidelines and practices. We recommend that companies perform a self-audit prior to an OSHA NEP audit. Specific issues to review include: scenario selection, occupancy screening, siting of portable buildings, location of non-essential personnel and conventional buildings inside process units. For facilities with combustible dust hazards, audit for compliance with NFPA 654 is also recommended. www.bakerrisk.com Page 5

RISK MANAGEMENT - QUANTITATIVE RISK ANALYSIS (QRA) AND THE POWER OF OBJECTIVE RISK COMPARISON by Jatin N. Shah, Sr. Principal Consultant Successful organizations understand that a goal of zero risk is not possible. Understanding the degree of risk one is exposed to and managing it to a tolerable level is a critical component of successful risk management strategy. There are a variety of qualitative and quantitative methods that companies can utilize to identify and assess risk exposures associated with their activities. You want a valve that doesn t leak and you try everything possible to develop one. But the real world provides you with a leaky valve. You have to determine how much leaking you can tolerate. Arthur Rudolph (1906-1996) Project Director for the Saturn V rocket program While the vast majority of risk issues can be addressed utilizing qualitative techniques, some risk exposures require additional clarification utilizing quantitative techniques. For these types of exposures, BakerRisk utilizes Quantitative Risk Assessment (QRA) studies to quantify the impact and likelihood of potential risk exposures to provide insight and help answer typical questions from clients such as: We have already adopted many risk reduction technologies; do we need to do more? If so, what? We operate this high risk unit at multiple locations; how do the risks compare? Where do I start first? We have four possible project configurations that we can adopt; how do the risks compare? Our facility has many occupied buildings; which ones need to be relocated? Which ones can be upgraded? Which ones are OK as is? In a QRA study, statistical databases are used to determine release frequencies based upon the type of equipment utilized in the process being evaluated, which allows the analyst to use a range of release sizes instead of having to assume a credible release size. Event trees are utilized to capture conditional probabilities (success failure of various mitigation measures, potential ignition of flammable release, outcome of ignition fire or explosion) and explore possible outcomes associated with the accidental release. Results are typically presented as geographic risk contours and in terms of societal risk (FN curves, Societal Risk Index). An example of offsite risk exposures from process units at five different facilities is shown in the figure at left. The cumulative frequency of having N or more impacts is depicted for each site from all potential scenarios within the unit for each site. Even though each location utilizes the same process technology, the site-specific differences are clearly evident, allowing one to prioritize the risk reduction efforts. Example FN comparison at five site locations (continued on following page) www.bakerrisk.com Page 6

(continued from previous page) The chart below shows onsite risk comparison by nine equipment groups at one site. Here, the information can be used to identify equipment groups with the higher risk contribution and target appropriate risk reduction efforts to those individual items. In this particular case, efforts are directed at addressing risks associated with equipment group 1, 3, 4 and 5. Example risk comparison by nine equipment groups at one site location Though the QRA study produces quantitative numerical values of risk expressed as lines on a chart or as index values, it provides our clients with the ability to objectively compare risks of various options, locations, or operations. The ability to specifically target equipment groups that contribute the most to risk exposure improves the efficiency of the risk management options and allows companies to produce highly effective, defensible risk management strategies. BAKERRISK IN THE NEWS In 2008, BakerRisk was named to the Inc.5000, the Inc. magazine list of the fastest growing private companies in America. Inc.5000 is the most comprehensive survey of fast-growing companies assembled by any organization. We were ranked #2,872 overall for 2008 based on corporate growth. In addition, we made the list of Top 50 Engineering Companies, ranked 44th. BakerRisk was also honored by the Center for New Ventures & Entrepreneurship at the Mays Business School at Texas A&M University. We were named to the Aggie 100 for 2008, which identifies the 100 fastest growing businesses in the world that are owned or run by Texas A&M alumni. BakerRisk ranked 74th. We are honored to have received these awards and pleased that the efforts of our team have been recognized by these organizations. www.bakerrisk.com Page 7

B AKER ENGINEERING AND RISK C ONSULTANTS, INC. About BakerRisk Baker Engineering and Risk Consultants, Inc. is one of the world s leading explosion analysis, structural design, and risk engineering companies. BakerRisk provides comprehensive consulting, engineering, laboratory and range testing services to government agencies and private companies who are involved with dangerous, highly hazardous, reactive, or explosive materials. Our Mission: To provide integrated engineering, research, and risk assessment to aid our clients in managing hazards associated with explosive, flammable, reactive and toxic materials. Headquarters San Antonio, TX (210) 824-5960 Chicago Office (708) 667-5550 Houston Office (281) 822-3100 Los Angeles Office (424) 218-3800 Washington DC Office (202) 496-1295 Canada Office (289) 288-0100 UK Office (+44) 1244 893 178 Blast Effects & Explosion Testing Dynamic Structural Analysis & Design Risk Engineering Process Safety Incident Investigations Reactive Chemicals Testing and Materials Engineering BAKERRISK CONFERENCE SCHEDULE Look for BakerRisk at these upcoming events: MARCH 26-27, 2009 International Concrete Repair Institute (ICRI) Carolinas Chapter Spring Conference Southport Durham, North Carolina www.icricarolinas.org/spring_conference.htm APRIL 14-17, 2009 Defense Research Institute (DRI) Product Liability Conference San Diego, California www.dri.org/open/seminardetail.aspx?eventcode=20090200 APRIL 26-30, 2009 American Institute of Chemical Engineers (AIChE) Spring Nat l Meeting & 5th Global Congress on Process Safety Tampa, Florida www.aiche.org/conferences/springmeeting/index.aspx MAY 11-15, 2009 International Symposium on the Interaction of the Effects of Munitions with Structures (ISIEMS) Bruhl, Germany www.streitkraefteamt.bundeswehr.de/portal/a/ska/isiems MAY 12-13, 2009 National Petrochemical and Refiners Association (NPRA) National Safety Conference and Exhibition Grapevine, Texas http://www.npradc.org/forms/meeting/meetingformpublic/view?id=18139000002b8 For subscription inquiries, address changes or additional information about BakerRisk, please send your request to BakerRisk_News@BakerRisk.com www.bakerrisk.com Page 8