Chapter 14. Case Histories and Lessons Learned

The learning objectives for this chapter are to:

  1. Discuss the consequences of inadequately addressing the 20 elements of risk-based process safety (RBPS).

  2. Describe other lessons learned from case histories.

The AICHE Center for Chemical Process Safety (CCPS) developed the risk-based process safety (RBPS) management system to assist chemical companies in implementing effective process safety management systems.1 The RBPS approach recognizes that all hazards and risks in an operation or facility are not equal and shows how resources can be effectively apportioned based on greater hazards and higher risks. This RBPS approach prioritizes efforts focusing on risk information and risk reduction. See Section 1-12 for more detailed information on RBPS.

1Center for Chemical Process Safety. Introduction to Process Safety for Undergraduates and Engineers (New York, NY: American Institute of Chemical Engineers, 2016).

This chapter will present several case histories based on the 20 RBPS elements. For each element, we present one or more case histories showing how an inadequacy in that RBPS element contributed to the incident, along with additional lessons learned from that incident. Although each case history is coupled with a specific RBPS element, all incidents always involve many of the 20 elements.

Learning from the past with case histories is important to prevent future incidents. As Santayana said, “Those who cannot learn from history are doomed to repeat it.”2

2“Learning from History: Famous Quotes”. 2017. http://www.age-of-the-sage.org/philosophy/history/learning_from_history.html.

14-1 Process Safety Culture

A good process safety culture is a positive environment where employees at all levels are committed to process safety. This starts at the highest levels of the organization and is shared by all. Process safety leaders nurture this culture.

Case History: Explosions at a Refinery Due to Inadequate Process Safety Culture

A very large oil refinery experienced a series of explosions in March 2005, leading to 15 fatalities and 180 injuries. The dollar losses were also significant: $1.6 billion to compensate victims, a fine of $87 million from the U.S. Occupational Safety and Health Administration (OSHA), and a $50 million fine for violations of environmental regulations.

This incident was due to a major tower release of gasoline that resulted in a vapor cloud explosion.3 The discharge from a relief flowed to a disposal system with an atmospheric vent, expelling the material to the atmosphere. The U.S. Chemical Safety Board (CSB) found that the company’s inadequate process safety culture was the root cause of this incident, which resulted in many fatalities and large facility and environmental losses.4 See Section 1-3.

3“Texas City Refinery.” 2017. http://en.wikipedia.org/wiki/Texas_City_Refinery_(BP)#Legal_action.

4“BP Texas City Incident: Baker Review.” www.hse.gov.uk/leadership/bakerreport.pdf.

The CSB found that the company had (1) a work environment that encouraged operations personnel to deviate from established procedures, (2) a lack of emphasis on effective communication for shift changes and hazardous operations, (3) ineffective supervisory oversight and technical assistance during unit startup, (4) insufficient staffing, (5) inadequate operator training, and (6) a failure to establish safe operating limits.

Lessons Learned

The following requirements are practiced to prevent incidents related to the lack of process safety culture:

  1. Establish safety as a major element in job performance and a critical factor when making salary and promotion decisions.

  2. Provide effective leadership and training in the area of process safety.

  3. Establish and implement a management system that continuously identifies and reduces safety risks.

  4. Train to develop a positive, trusting, and open attitude toward process safety.

  5. Maintain a high awareness of process hazards and potential consequences of incidents and continuously search for and correct weaknesses.

  6. Empower employees to successfully fulfill their safety responsibilities.

  7. Emphasize training and knowledge, and bring in outside experts when specific expertise is required.

  8. Create an environment of open and effective two-way communications so all employees know the process information relevant to their work.

  9. Develop a trusting environment where employees and supervisors accept and encourage constructive criticisms.

  10. Manage employment decisions to focus on process safety.

Notice that improvements in all of these areas are made by measuring results and making appropriate adjustments as guided by the measured results. In other words, you can’t improve what you don’t measure. Measured results are used continuously to identify and reduce process safety culture deficiencies.

14-2 Compliance with Standards

The organization must appropriately use applicable regulations, standards, codes and company requirements that are developed by national, state/provincial, local governments, professional organizations, and companies. See Section 1-10. Train and retrain to correctly understand and use these requirements.

Case History: Dust Explosions at a Pharmaceutical Plant Due to Inadequate Training on the Use of Standards

In January 2003, an explosion in a pharmaceutical plant killed 6 employees and injured 38, including 2 fire fighters who responded to the incident.5 Primary and secondary dust explosions ignited fires throughout the facility.

5“Dust Explosion at West Pharmaceutical Services.” 2017. www.csb.gov/assets/document/West_Digest.pdf.

The CSB investigation found that a major dust hazard had developed in the plant over a period of years.6 Combustible dust had accumulated on hidden surfaces above the production area, creating the fuel for the massive explosions and fires. Due to a poor design of the ventilation system, the dust was drawn above a false ceiling where the dust accumulated. The dust gradually accumulated to a thickness of about one-half inch on the ceiling tiles.

6Center for Chemical Process Safety. Introduction to Process Safety for Undergraduates and Engineers (New York, NY: American Institute of Chemical Engineers, 2016), p. 16.

The root cause of the incident was a lack of compliance with standards. The designers and operators were inadequately trained in compliance to standards. Several organizations provide readily available standards to identify and control hazards due to dusts.

Lessons Learned
  1. Train operators and designers to understand and use available codes and standards for handling hazardous materials, including Recognized and Generally Accepted Good Engineering Practices (RAGAGEP) and the National Fire Protection Association (NFPA) fire codes and standards; see Section 1-10.

  2. Regularly train and retrain operators, maintenance personnel, and engineers about the dangers of handling hazardous materials.

  3. Ensure that standards are followed with frequent audits. Examples of compliance with standards include (1) using explosion proof electrical fixtures when handling flammable materials, (2) designing ventilation systems to prevent the accumulation of dust, and (3) cleaning to remove collected dusts, including above false ceilings.

14-3 Process Safety Competency

All companies must have skills and resources in the right places to manage its process hazards. The company must also verify these skills and resources and must use this information in succession planning and management of organizational changes.

Case History: An Explosion of a Blender Due to Inadequate Knowledge of Chemical Process Safety

In April 1995, a blender containing a mixture of reactive chemicals exploded at a specialty chemical plant in Lodi, New Jersey.7 Five employees were killed, and more were injured. Property damage was about $20 million. Three hundred residents of the neighboring community were evacuated, and other businesses in the area were destroyed. The cause of the incident was the inadvertent addition of water to a water-activated reactive chemical. The root cause was the lack of process safety competency since the operators were not aware of this hazard and did not have the correct information to handle it properly.

7“EPA/OSHA Joint Chemical Accident Investigation Report.” https://archive.epa.gov/emergencies/docs/chem/web/pdf/napp.pdf.

Lessons Learned
  1. Train and retrain all employees regarding the hazards in their plant.

  2. Continuously improve knowledge and competency.

  3. Ensure that all the information needed is available to all employees.

  4. Continuously apply the knowledge that is learned. This includes an understanding of (a) the detailed process and chemical hazards, (b) conditions for failures and safeguards, (c) operating limits and conditions that will cause incidents when exceeded, (d) consequences of exceeding limits, (e) how to conduct a hazards analysis, (f) normal operating conditions, (g) how to safely handle abnormal conditions, (h) startups after normal and emergency shutdowns, (i) emergency evacuation conditions and procedures, (j) limitations of process equipment, (k) communication methods for on-site and off-site personnel including responders, and (l) conditions necessary for reentering a plant after emergency evacuations.

14-4 Workplace Involvement

Broad involvement of operating and maintenance personnel in process safety activities is necessary to ensure that lessons learned by the people closest to the process are considered and addressed.

Case History: A Fatality in a Ribbon Blender Due to an Inadequate Lock-Out/Tag-Out Permit System

Two maintenance workers were replacing part of a ribbon in a large ribbon mixer.8 The main switch was left energized; the mixer was stopped with one of three start–stop buttons. As one mechanic was completing his work inside the mixer, another operator on an adjoining floor pushed, by mistake, one of the other start–stop buttons. The mixer started, killing the mechanic between the ribbon flight and the shell of the vessel.

8Case Histories of Accidents in the Chemical Industry, Vol. 2 (Washington, DC: Manufacturing Chemists’Association, January 1969), p. 225.

Lock-out/tag-out (LOTO) procedures were developed to prevent incidents of this kind. A padlocked switch at the starter box disconnect, with the key in the mechanic’s pocket, can prevent this type of incident. After the switch gear lock-out, the mechanic should also verify the dead circuit by testing all switches. The root cause of this incident was a lack of workplace involvement, since other workers in adjacent areas were not aware of how their operations might affect other units.

Lessons Learned
  1. Develop a plan to involve all employees in the process safety program.

  2. Include employees of all types and at all levels. In this incident, employees lacked the appropriate knowledge, including supervisors, mechanics, and operators. Clearly, they were not appropriately involved in the safety program.

  3. Involve employees and contractors in the development, implementation, and continuous improvement of all RBPS elements.

  4. Include all employees who are directly involved in the work to protect their own safety.

14-5 Stakeholder Outreach

A process is necessary for identifying, engaging, and maintaining good relationships with appropriate external stakeholder groups. This would include the surrounding community, suppliers of raw materials, customers, government agencies and regulators, professional societies, and contractors, among others.

Case History: Increased Consequences in an Adjacent Community Due to Inadequate Outreach

In December 2007, an explosion of a 2450-gallon chemical reactor killed 4 four employees and injured 32, including 4 employees and 28 members of the public who were working in adjacent companies. The explosion damaged buildings within one-quarter mile of the facility. It occurred during the production of the 175th batch of MCMT, a gasoline additive. The direct cause of this incident was a loss of sufficient cooling during the reaction process, which resulted in an uncontrolled runaway reaction leading to excessive temperatures and pressures.9 Adjacent companies were not aware of the hazards associated with this operation and did not have proper protection or emergency response.

9T2 Laboratories. “Runaway Reaction.” 2017. http://www.csb.gov/t2-laboratories-inc-reactive-chemicalexplosion/.

Lessons Learned
  1. Increase communications with local communities; other plants on the site and off the site; and local, state, and federal authorities, to inform them of the hazardous chemicals handled in the plant. Also discuss possible incidents, and develop, together with stakeholders, actions to minimize problems.

  2. Work with other companies that make the chemical to share information and hazards and develop and implement best practices.

14-6 Process Knowledge Management

The company must ensure assembly and management of all information needed to perform process safety activities. This includes verification of the accuracy of this information and confirmation that this information is correct and up-to-date. This information must be readily available to those who need it to safely perform their jobs.

Case History: A Runaway Reaction and Explosion Due to Inadequate Process Knowledge Management

The previous case history on stakeholder outreach is also relevant to this element. The CSB found that the company’s management (including one chemist and one chemical engineer) did not understand the runaway reaction hazards associated with producing MCMT. The major technical deficiencies were the dependence on a single temperature controller and an undersized pressure relief. These professionals did not understand and appreciate the importance of redundant controls when controlling potential runaway reactions, nor did they have the necessary knowledge regarding properly sizing pressure relief for runaway reactions. Three of the engineers who designed this plant had MS degrees in chemical engineering, but no industrial experience or education on process hazards and their control. The CSB recommended that undergraduate programs in chemical engineering include hazard recognition as part of the curriculum.

Lessons Learned
  1. CSB recommended that the American Institute of Chemical Engineers (AICHE) and the Accreditation Board for Engineering and Technology (ABET) should work together to add reactive hazard awareness to the baccalaureate chemical engineering curricula requirements.

  2. AICHE should also inform all student members about the AICHE’s SACHE Process Safety Certificate Program and encourage participation.

This RBPS Process Knowledge Management element includes (1) gathering of process information and (2) interpretation and understanding of this information. Specifically, the process information includes design drawings, equipment specifications, chemistry including reactivity and flammable properties, process conditions with limits of temperature and pressure, calculation methods including those for sizing reliefs, process hazard reviews, and emergency conditions and responses.

14-7 Hazard Identification and Risk Analysis

This element consists of the identification of process hazards and their potential consequences. It includes (1) the definition of the risk posed by these hazard scenarios and (2) recommendations to reduce or eliminate hazards, reduce potential consequences, and reduce frequency of occurrence. Analysis may be qualitative or quantitative depending on the level of risk.

Case History: A Chemical Release and Fire Due to Inadequate Identification of Brittle Metal Failure

In September 1998, a gas plant at Longford in Victoria, Australia, suffered a major fire.10-12 Two men were killed and the state’s gas supply was interrupted for two weeks, causing chaos in the local industry and considerable hardship in homes that were dependent on the plant’s gas. A valve on a warm liquid (known as the “lean oil”) failed to close, allowing a metal heat exchanger to become very cold and therefore brittle. When operators reintroduced the warm lean oil, the cold metal fractured and released a large quantity of gas that found an ignition source. In 2004, the company was ordered to distribute $32.5 million to businesses that suffered property damage during this incident. The major reason for this incident was the lack of hazard identification and risk analysis. A risk study was planned for three years prior to this incident but not conducted.

10A. Hopkins. “Lessons from Esso’s Gas Plant Explosion at Longford.” 2017. http://www.futuremedia.com.au/docs/Lessons%20from%20Longford%20by%20Hopkins.PDF.

11Center for Chemical Process Safety. Introduction to Process Safety for Undergraduates and Engineers (New York, NY: American Institute of Chemical Engineers, 2016), p. 22.

12Wikipedia. “Esso Longford gas explosion.” 2017. https://en.wikipedia.org/wiki/Esso_Longford_gas_explosion.

Lessons Learned
  1. Identify hazards and risks, and eliminate risks with improved controls, equipment, and procedures.

  2. Update and revalidate hazards identification and risk analysis (HIRA) studies on a regular basis (U.S. OSHA PSM requires HIRA studies to be completed and validated every 5 years).

  3. Include in the HIRA study low-frequency and high-consequence incidents.

  4. Include in the HIRA study knowledgeable people who have experience and understand (a) previous incidents, near misses, leaks, and alarms; (b) conditions exceeding limits;(c) consequences of exceeding limits; (d) maintenance records; (e) work practices; (f) normal and emergency procedures; and (g) conditions negatively effecting equipment.

14-8 Operating Procedures

The company needs written instructions for a manufacturing operation that describe how the operation is to be carried out safely, explaining the consequences of deviation from procedures, describing key safeguards, and addressing unusual situations and emergencies.

Case History: A Fatality from a Runaway Reaction Due to Inadequate Training on the Use of Procedures

In August 1997, a plant experienced a runaway reaction with a phenol–formaldehyde polymerization reactor.13 The result was one fatality and seven injuries, along with environmental damage. The runaway reaction was triggered when, contrary to standard operating procedures, all the raw materials and catalyst were charged to the reactor at once, resulting in all the charged material reacting almost instantaneously. The procedure correctly specified the use of a semi-batch reaction procedure in which the reactants are added slowly to the reactor, controlling the rate of heat release. The root cause of this incident was the lack of administrative controls to ensure that operators are trained to use standard operating procedures appropriately.

13Environmental Protection Agency. How to Prevent Runaway Reactions, Report 550-F99-004. August 1999. www.epa.gov.

Case History: Runaway Reaction and Explosion Due to Inadequate Procedures

A plant manufactured a dye by mixing and reacting two chemicals.14 On April 8, 1998, a runaway reaction caused an explosion and flash fires that injured nine workers. The resulting high temperature led to a secondary runaway decomposition reaction, causing the explosion. The runaway reaction was triggered by starting the reaction at a temperature higher than specified in the normal operating procedure. In scaling up the process from the laboratory phase to production scale, this company also changed the process from a semi-batch process to a batch process. Operating procedures were inadequate—that is, they did not cover the safety consequences of deviations from normal operating limits, such as runaway reactions, or specify steps to be taken to avoid or recover from such deviations.

14U.S. Chemical Safety Board. Chemical Manufacturing Incident, Report 1998-06-I-NY. www.chemsafety.gov/reports/2000/morton/index.htm.

Lessons Learned
  1. Train and retrain operators and maintenance personnel to understand the procedures, such as the limits of conditions, and consequences of deviations from the norm.

  2. Make periodic and precise audits of procedures.

  3. Use procedures effectively, such as lock-out/tag-out, hot work, vessel entry, emergency, operating, startup, and other safety procedures.

  4. Use HIRA procedures during the design phases of projects, including new installations or modifications to existing systems.

  5. Use a permit procedure for opening vessels that are normally under pressure.

  6. Add cautions to the procedures.

  7. Communicate operating changes to other operations that may be affected by the change.

  8. Recognize that batch reactors can be very hazardous, whereas semi-batch are less hazardous. In a semi-batch reactor, the reactants are added slowly over a period of time, controlling the heat generation from the reaction. In a batch reactor, all of the reactants are added at the same time, causing the entire heat release to occur almost instantaneously.

  9. Write procedures for troubleshooting systems when they are not operating normally.

14-9 Safe Work Practices

Each company needs adequate safe work practices to safely maintain and repair equipment, such as permits to work, line breaking, and hot work permits. This applies to nonroutine operations.

Case History: An Explosion Due to a Missing Hot-Work-Permit System

On July 17, 2001, an explosion occurred at a plant in Delaware City, Delaware.15, 16 A crew was welding a grating on a catwalk in a sulfuric acid storage tank farm when a spark from the welding ignited flammable vapors in one of the adjacent storage tanks. A maintenance contractor was killed, and eight others were injured. The incident also resulted in significant damage to nearby aquatic life.

15“Motiva Enterprises Accident.” 2001. http://www.csb.gov/assets/1/19/Motiva_Final_Report.pdf.

16Center for Chemical Process Safety. Introduction to Process Safety for Undergraduates and Engineers (New York, NY: American Institute of Chemical Engineers, 2016), p. 26.

This incident was the result of inadequate safe work practices—in particular, no hot work permit to control ignition sources. Hazards in areas adjacent to the welding should have been identified, evaluated, and controlled prior to the start of the welding.

Lessons Learned

Document and use safe work practices to eliminate hazards when working in potentially hazardous process areas.

  1. Include safety practices such as lock-out/tag-out, hot work, line breaking, vessel entry, confined-space entry, and others. These practices are requirements for safely conducting maintenance, inspections, and operations.

  2. Train and retrain operators and maintenance personnel to understand safe work practices and procedures, such as the mechanical integrity checks, limits of conditions, consequences of deviations from the norm, and the flammable and reactive hazards of the chemicals being used in the process.

  3. Use HIRA procedures during all phases of a project, including new installations or modifications to existing systems.

  4. Use management of change procedures and safe work practices for all plant modifications.

  5. Use frequent mechanical integrity checks on all equipment and controls.

  6. Conduct frequent audits to ensure safe work practices are used as intended.

14-10 Asset Integrity and Reliability

Each company must manage activities that ensure that important equipment remains suitable for its intended purpose throughout its service. This element includes proper selection of materials of construction; inspection, testing, and preventive maintenance; and design for maintainability.

Case History: A Catastrophic Pipe Rupture Due to an Inadequate Asset Integrity Program

On August 6, 2012, a refinery in Richmond, California, experienced a catastrophic pipe rupture.17 The ruptured 8-inch pipe released flammable light gas oil, which then partially vaporized into a large vapor cloud. Two minutes following the release, an explosion occurred. Six employees suffered minor injuries during the incident and subsequent emergency response efforts. Approximately 15,000 people from the surrounding communities sought medical treatment at nearby medical facilities. The rupture was due to the failure of a short pipe section due to a corrosion mechanism known as sulfidation corrosion. The plant was aware of the hazard and did have an ultrasonic testing program to measure pipe wall thicknesses. However, the pipe that failed was a short length that was not tested.

17Chevron U.S.A. 2017. file:///C:/Users/Owner/Downloads/Chevron_Final_Investigation_Report_2015-01-28%20(1).pdf.

The main cause of this incident was the failure of the plant’s program for verifying asset integrity and reliability.

Lessons Learned
  1. Document and utilize procedures to verify asset integrity and reliability, such as frequent inspections and tests to verify and record the integrity of equipment in relation to specifications, and include a system to correct deficiencies promptly.

  2. Include requirements that assure the equipment is correctly designed and installed according to specifications and remains fit for use until decommissioned.

  3. Use audits to verify that this requirement is conducted as intended.

14-11 Contractor Management

Practices are needed to ensure that contract workers can perform their jobs safely, and that contracted services do not add to or increase facility operational risks.

Case History: Fire and Fatalities in a Tunnel Due to Poor Management of Contractors

On October 2, 2007, five people were killed and three others injured when a chemical fire occurred 1000 feet underground in a tunnel. The fire started when industrial painting contractors were recoating a portion of an enclosed tunnel with a highly flammable epoxy coating product. The fire trapped five workers who were killed by smoke inhalation inside the tunnel. The CSB report18 stated that contractors did not receive comprehensive formal safety training—specifically, training on company policies and practices, site-specific instruction addressing confined space safety, and the handling of flammable liquids.

18U.S. Chemical Safety and Hazard Investigation Board. “Contractor Safety Digest.” 2018.

Lessons Learned

Incidents due to failures in managing contractors can be prevented by using the following practices:

  1. Check the contractor’s safety performance records before selecting contractors. This should be a major condition for employment.

  2. Provide detailed training to all contractors about process hazards. Their training should be equivalent to the training provided to employees.

  3. Verify the contractors have the knowledge necessary for working safely on this site.

The contractor selection process should include criteria and procedures for prequalifying or disqualifying contractors based on specific safety performance measures. It should also require a comprehensive review and evaluation of contractor safety policies, procedures, and safety performance, including working in confined spaces.

Company policy should also include periodic safety audits of contractor selection and oversight to ensure continuous adherence to safety policies, procedures, and practices.

14-12 Training and Performance Assurance

The company must provide practical instruction in job and task requirements and methods for operation and maintenance workers, supervisors, engineers, leaders, and process safety professionals. It must also verify that the trained skills are being practiced proficiently.

Case History: An LPG Leak and BLEVE Due to Inadequate Training

On January 4, 1966, operators were attempting to drain water out of the bottom of a liquid propane tank.19,20 They partially opened two valves in series at the bottom of the tank. As a result of propane evaporation, the temperature was cold enough to freeze open both valves. The subsequent continuous release of liquified petroleum gas (LPG) caught fire and created conditions for a boiling liquid expanding vapor explosion (BLEVE), and a large explosion occurred. Eighteen people were killed and 81 were injured. The major cause of the incident was the lack of training and performance assurance: The operators were not trained to properly handle this situation.

19Center for Chemical Process Safety. Introduction to Process Safety for Undergraduates and Engineers (New York, NY: American Institute of Chemical Engineers, 2016), p. 32.

20“Fire and Explosion of LPG Tanks at Feysin, France.” 2017. http://www.sozogaku.com/fkd/en/hfen/HC1300001.pdf.

Lessons Learned
  1. Train operators and maintenance personnel about the hazards of the plant so they can conduct their tasks without creating safety problems.

  2. Include classroom training and training in the field.

  3. Train and retrain to have workers meet documented performance standards.

  4. Conduct training and performance checks before workers are allowed to work in the plant.

  5. Periodically check performance, and conduct retaining when deficiencies are noted.

14-13 Management of Change

The company needs a process for reviewing and authorizing proposed changes to facility design, operations, organization, procedures, personnel changes, activities prior to implementing them, and for ensuring that the process safety information is updated when changes are made.

Case History: An Explosion Due to Missing Management of Change Procedure

An incident occurred in Flixborough, England, on June 1, 1974.21 The Flixborough plant produced caprolactam, a basic raw material for the production of nylon. The process in which the incident occurred consisted of six reactors in series. In these reactors, cyclohexane was oxidized to cyclohexanone and then to cyclohexanol using injected air in the presence of a catalyst. Prior to this incident, it was discovered that the fifth reactor needed to be removed due to a crack in the vessel wall. It was decided to continue operating by connecting reactor 4 directly to reactor 6 in the series. The connection was made temporarily with readily available 20-inch pipe stock and with flexible bellows sections at both ends.

21“Flixborough Explosion 1st June 1974.” http://www.hse.gov.uk/comah/sragtech/caseflixboroug74.htm.

This temporary piping also had an inadequate support that allowed flexing as a result of internal fluctuating reactor pressures. After a short period, the bypass connection failed. An estimated 30 tons of cyclohexane volatilized and formed a large vapor cloud; this cloud was then ignited by an unknown source. The resulting explosions leveled the entire plant. Twenty-eight people died, and 36 others were injured. The major cause of this incident was the lack of using a management of change (MOC) safety procedure. The design for the temporary pipe section was drawn on the floor of the machine shop without further review.

Lessons Learned
  1. Conduct a MOC review for any changes in plants that are not “like for like” or “in-kind” changes.

  2. Use a review procedure that is proportional to the magnitude of the potential hazards. If the hazards are large, then the review should be a HAZOP or equivalent.

  3. Use people with the appropriate expertise and with supervisors.

  4. Ascertain that the changes do not inadvertently add hazards.

  5. Design changes with an engineering designer with the appropriate knowledge of design standards.

  6. Notify all operators and maintenance personnel of the details of the recommended changes.

  7. Confirm that the documented MOC process specifies that anyone at any time can initiate an MOC.

  8. Add to the MOC procedure a follow-up requirement to update process information, operating conditions, procedures, maintenance requirements, drawings, and training, as required.

14-14 Operational Readiness

Evaluation of the process must occur before startup or restart to ensure the process can be safely started. This element applies to restart of facilities after being shut down or idled as well as after process changes and maintenance. It also applies to startup of new facilities.

Case History: A Fatality in a Ribbon Blender Due to an Inadequate Pre-Startup Safety Review

This case history is the same as that in Section 14-4 for workplace involvement. In this case history, a ribbon blender was being repaired by a maintenance worker inside the unit. The blender was activated by another operator on another floor. This case highlights additional requirements before startups to ensure that the system is safe prior to startup.

Oftentimes inadequacies in two or more RBPS elements are needed to initiate incidents. In the ribbon blender incident, the fatality could have been prevented with the adequate practices of any one of the following three RBPS elements: safe work practices (use the LOTO practice to prevent startup problems), workplace involvement (include workers in reviews and decisions to emphasize practices to protect their safety), and operational readiness (walk the line to identify pre-startup problems before initiating startup).

Lessons Learned
  1. Perform a pre-startup safety review (PSSR) prior to startup.

  2. Include in the review startups after normal, emergency, and maintenance shutdowns.

  3. Confirm that the system utilizes the specific design standards for this plant.

  4. Verify that the operating, maintenance, and emergency procedures are in place.

  5. Assure that all controls are operational and in the proper condition for startup.

  6. Confirm that all maintenance on equipment and instruments is completed and ready for startup.

  7. Confirm that all training and retraining are completed, and verify that the skills and competence are satisfactory.

  8. Ensure that all action items from reviews, audits, and inspections are completed before startup.

  9. Instruct operators to “walk the line,” so as to be sure the piping, equipment, controls and valves are in the appropriate status before startup.

  10. Supervision must verify that all responsibilities of the operational readiness review are completed (documented completions), before giving the operators written approval for startups.

14-15 Conduct of Operations

This element includes means by which management and operational tasks required for process safety are carried out in a deliberate, faithful, and structured manner. Managers ensure that workers carry out the required tasks and prevent deviations from required and expected performance.

Case History: Explosions in a Refinery Due to Inadequate Conduct of Operations

On January 21, 1997, an explosion and fire occurred at the hydrocracker unit of a refinery in Martinez, California, resulting in one death, 46 worker injuries, and precautionary sheltering-in-place for the surrounding community.22,23 The incident involved the release and autoignition of a mixture of flammable hydrocarbons and hydrogen that were under high temperature and pressure. An overheated pipe ruptured.

22“EPA Report: Tosca Avon Refinery in Martinez, California.” November 1998.

23Center for Chemical Process Safety. Introduction to Process Safety for Undergraduates and Engineers (New York, NY: American Institute of Chemical Engineers, 2016), p. 37.

Prior to the incident, a temperature excursion occurred, but readings were ignored because temperature readings were historically unreliable. Radio communications from the field were ignored because they were routinely inaudible. The operators continued to run the plant, even when the process information was unreliable.

Case History: A Toxic Release Due to Inadequate Conduct of Operations

On December 3, 1984, a pesticide plant in Bhopal, India, released an estimated 40 metric tons of methyl isocyanate (MIC) into the atmosphere.24, 25 An estimated 3000 local community residents died, and more than 10,000 were injured.

24 Wikipedia. “Bhopal disaster.” https://en.wikipedia.org/wiki/Bhopal_disaster.

25Center for Chemical Process Safety. Introduction to Process Safety for Undergraduates and Engineers (New York, NY: American Institute of Chemical Engineers, 2016), p. 110.

The Bhopal incident was caused by water entering a MIC storage tank. This influx led to an exothermic reaction between the MIC and water, heating the contents, causing the pressure to increase, and releasing the MIC vapor.

Significant maintenance deficiencies contributed to this release: (1) a failed pressure and temperature alarm on the storage tank could have initiated corrective actions; (2) a refrigeration system that was not in operation on the storage tank could have prevented or mitigated the accumulated heat and the magnitude of the release; (3) a relief vent scrubber was not operating and could have absorbed the MIC; (4) an existing flare that would have burned the exiting MIC was not operating; and (5) a water curtain at the exit of the flare was not designed correctly to absorb the gas from the flare. These were all conduct of operations issues.

Lessons Learned
  1. Assign supervisors and managers to plants who accept the responsibility to operate and maintain their plants safely.

  2. Supervisors and managers must manage a documented system to be sure (a) operators and maintenance workers are trained and retrained; (b) procedures are developed, updated, and maintained; (c) process hazards analysis (PHAs) are updated and adequate; and (d) operators and maintenance workers have an appropriate safety culture that prevents cutting corners that may jeopardize safety.

  3. Use operating procedures and maintain controls and equipment to operate as intended and shut down the plant when the controls or equipment are not operating satisfactorily.

14-16 Emergency Management

The company needs plans for possible emergencies that define actions in an emergency; resources to execute those actions; practice drills; continuous improvement, training, appropriate and frequent communications with employees, contractors, neighbors, and local authorities; and communications with stakeholders in the event an incident does occur.

Case History: An Ammonium Nitrate Explosion Due to Inadequate Emergency Management

On April 16, 1947, a large explosion occurred in a cargo ship containing 1400 tons of ammonium nitrate.26, 27 A fire occurred, possibly caused by a cigarette, and the fire was not properly extinguished. The resulting explosion killed 581 people, including all but one member of the Texas City fire department. The root cause of the incident was attributed to a grossly deficient emergency management plan. The fire fighters were not aware of the hazards of ammonium nitrate and were not trained to properly handle the situation.

26“Texas City Disaster.” https://en.wikipedia.org/wiki/Texas_City_disaster.

27Center for Chemical Process Safety. Introduction to Process Safety for Undergraduates and Engineers (New York, NY: American Institute of Chemical Engineers, 2016), p. 38.

Case History: An Explosion in a Pesticide Plant Due to Inadequate Emergency Management

On August 28, 2008, a large explosion occurred that killed 2 people and injured 8; 40,000 area residents were requested to shelter-in-place following the event.28, 29 A root cause was poor emergency management due to poor communications between the plant, responders, and community leaders. During the incident, the company refused to share critical safety information with the surrounding community. The CSB investigation found that operators were poorly trained, startup procedures were inadequate, and operators, with management approval, bypassed a critical control loop that resulted in a violent runaway reaction.

28“Animation of Bayer Crop Science Pesticide Waste Tank Explosion” [Video]. https://www.csb.gov/videos/.

29“CSB Issues Report on 2008 Bayer Crop Science Explosion: Finds Multiple Deficiencies Led to RunawayChemical Reaction; Recommends State Create Chemical Plant Oversight Regulation.” https://www.csb.gov/csbissues-report-on-2008-bayer-cropscience-explosion-finds-multiple-deficiencies-led-to-runaway-chemical-reactionrecommends-state-create-chemical-plant-oversight-regulation/.

Lessons Learned

Companies must develop an emergency management plan, including the following steps:

  1. Identify incident scenarios based on hazards.

  2. Plan response actions for each credible incident scenario.

  3. Verify that communications are adequate.

  4. Specify duties and responsibilities.

  5. Update procedures including decontamination.

  6. Specify areas for emergency equipment and purchase equipment and tools for responding appropriately.

  7. Train and retrain, including field drills.

  8. Periodically review the emergency plans and continuously make improvements.

  9. Assign supervisors and managers the responsibility to manage this RBPS element.

Emergency responses need to be discussed with the community, and positive actions taken before, during, and after emergencies.

14-17 Incident Investigation

A process is needed for reporting, tracking, and investigating incidents and near misses to identify root causes; taking corrective actions; evaluating incident trends; and communicating lessons learned.

Case History: Space Shuttle Fatalities Caused by Inadequate Incident Investigations

On January 16, 2003, a space shuttle had an in-flight failure that resulted in the deaths of the seven-member crew.30 The root cause of this incident was the failure to adequately investigate abnormal results in previous space shuttle flights.

30“Columbia Accident Investigation Board Report.” http://s3.amazonaws.com/akamai.netstorage/anon.nasa-global/CAIB/CAIB_lowres_full.pdf.

Case History: Explosions in a Sugar Refinery Due to Inadequate Incident Investigations

On February 7, 2008, huge explosions and fires occurred at a sugar refinery, causing 14 deaths and injuring 38 workers.31, 32 The company had not taken effective actions over many years to control dust explosion hazards—even as smaller fires and explosions continued to occur in its plants. The company clearly had an inadequate incident investigation process that should have identified and eliminated the causes of these previous incidents, including managed follow-up actions to ensure recommendations were completed. Correcting the causes of the previous incidents would have prevented this incident.

31“Imperial Sugar Company Dust Explosion and Fire.” https://www.csb.gov/imperial-sugar-companydust-explosion-and-fire/.

32“CSB Report: Sugar Dust Explosion and Fire.” https://www.csb.gov/assets/1/20/imperial_sugar_report_final_updated.pdf?13902.

The facility management was aware of sugar dust explosion hazards, the importance of properly designed dust-handling equipment and good housekeeping practices to minimize dust accumulation. However, management did not take appropriate actions to minimize and control sugar dust hazards. A root cause of this incident was the failure to adequately investigate incidents and eliminate abnormal results that had caused previous incidents.

Lessons Learned

Develop and document an incident investigation process, including the following steps:

  1. Investigate and evaluate all incidents and near misses.

  2. Identify the causes of the incidents.

  3. Manage the investigation and recommendation processes to eliminate the causes.

  4. Include, as members of the investigation team, people with the investigation skills and competence in the process being investigated.

  5. Include, in the documented incident investigation description, a follow-up method to be sure the investigation is adequate and recommended responses are completed rapidly.

14-18 Measurement and Metrics

Leading and lagging indicators of process safety performance should include incident and near-miss rates as well as metrics that show how well key process safety elements are being performed. This information is used to drive continuous improvements in process safety.

Case History: Flight Failure of Mars Orbiter Due to Inadequate Analysis of Flight Path Deviations

On September 23, 1999, the Mars Climate Orbiter was lost during its attempt to enter into orbit around Mars.33 During the nine-month flight, the flight path had to be adjusted ten times. The navigation team recognized that the results of the maneuvers were not mathematically accurate; that is, actual results were not as predicted. After the incident they discovered that one of the maneuvering calculations was done using the wrong units—specifically, English units versus SI units. The incident could have been prevented if the team had evaluated the reasons for the abnormal deviations during the flight. The purpose of the measurement and metrics element is to identify process problems while operating a plant and adjust performance to prevent incidents. This deficiency was the major cause of the Mars Climate Orbiter incident.

33“Mars Mishap Investigation Board Report.” http://sunnyday.mit.edu/accidents/MCO_report.pdf.

Case History: Explosions in an Oil Refinery Due to Inadequate Focus on Process Safety Metrics

On March 23, 2005, an incident occurred during the startup of a refinery’s octane-boosting isomerization unit.34, 35 Explosions and fires killed 15 people and injured another 180, alarmed the community, and resulted in financial losses exceeding $1.5 billion. A tower was overfilled and a pressure relief device opened, resulting in a flammable liquid geyser from a stack that was not equipped with a flare. The release of flammables led to an explosion and fire. All of the fatalities occurred in or near office trailers located close to the stack.

The company had responded to previous incidents with a variety of metrics aimed at improving safety. However, the focus of these initiatives was on improving procedural compliance and reducing occupational injury rates, while catastrophic safety risks remained unaddressed. Unsafe and antiquated equipment designs were left in place, and unacceptable deficiencies in preventive maintenance were tolerated. The company’s focus was on slips, trips, and falls, rather than management systems, equipment design, and preventive maintenance programs to prevent the risk of major process accidents.

34“CSB Investigation of BP Texas City Refinery Disaster Continues as Organizational Issues Are Probed.”

https://www.csb.gov/csb-investigation-of-bp-texas-city-refinery-disaster-continues-as-organizational-issues-areprobed/.

35“BP Refinery Explosion and Fire.” file:///C:/Users/Owner/Downloads/CSBFinalReportBP%20(1).pdf.

Lessons Learned

To improve safety performance, plants need to measure a combination of leading and lagging indicators. The purpose of this element is to measure process indicators, evaluate deviations, and adjust results when deviations are unacceptable. See Section 1-6 and Table 1-5 for additional information on leading and lagging indicators.

14-19 Auditing

Periodic critical reviews of process safety management system performance must be performed by auditors to identify gaps in performance and identify improvement opportunities, with closure of these gaps being tracked to completion.

Case History: Explosion in a Gas Plant Due to an Inadequate Audit of Asset Integrity and Reliability

This is the same case history used in Section 14-7 on hazard identification and risk assessment. Following the Longford explosion, the Australian Royal Commission criticized the company’s management system and its auditing process.36 Although the company completed an audit of its management system one year before the incident, the audit failed to uncover the fact that the critical hazard identification procedure (HAZOP) had not been carried out. The audit also failed to identify the problems that caused this explosion.

36A. Hopkins. “Lessons Learned from the Longford Explosion.” http://www.futuremedia.com.au/docs/Lessons%20from%20Longford%20by%20Hopkins.PDF.

Lessons Learned

The audit process is very important: It assures that the plant practices the RBPS requirements as intended. A good audit process includes the following components:

  1. Conduct frequent and thorough audits to identify and eliminate problems before an incident occurs.

  2. Audit all elements of the RBPS process to ensure all requirements are fulfilled as documented.

  3. Conduct audits with teams of knowledgeable workers – both outside and inside the facility being audited. Also include outside consultants with specific skills in this audited area.

  4. Use the results to correct problems and to improve the documented audit process.

  5. Verify that training, maintenance, and other periodic tasks are scheduled and completed appropriately.

14-20 Management Review and Continuous Improvement

Managers at all levels must set RBPS process safety expectations and goals with their staff and review performance and progress toward those goals. This activity may take place in a staff or“leadership team” meeting or on a one-on-one basis. It may be facilitated by a process safety leader but is owned by the line manager.

Case History: An Explosion Due to the Failure of Many RBPS Elements

The incident presented in Section 14-9 on operating procedures had failures in other RBPS elements, including management of change, asset integrity and reliability, and operating procedures.37, 38

37Chemical Safety Board Investigations. http://www.csb.gov/investigations/completed-investigations/?Type=2.

38Chemical Safety Board Video Room. http://www.csb.gov/videos/.

Lessons Learned

Identify problems with the entire RBPS management process by frequently reviewing or auditing the entire RBPS system for the following purposes:

  1. Find problems and deficiencies with the documented process and make corrections and improvements when the performance is not as intended.

  2. Conduct these audits with knowledgeable workers and, occasionally, skilled consultants.

  3. Update (as soon as possible) all documented elements of the RBPS system to include the recommendations of this review.

  4. Study the improved system to ensure new problems have not been inadvertently added.

  5. Communicate the details of the improved system with all workers in the plant.

  6. Verify that training, maintenance, and other tasks are scheduled appropriately, and not waiting to be initiated by audits.

14-21 Summary

The 20 RBPS elements are an effective management system to prevent incidents. The effectiveness of the RBPS 20 elements was illustrated in this chapter using case histories.

Suggested Reading

Case Histories of Accidents in the Chemical Industry, Vol. 1 (Washington, DC: Manufacturing Chemists’ Association, July 1962).

Case Histories of Accidents, Vol. 2 (January 1966).

Case Histories of Accidents, Vol. 3 (April 1970).

Case Histories of Accidents, Vol. 3 (April 2013).

T. A. Kletz. “Friendly Plants.” Chemical Engineering Progress (July 1989): 8–26.

T. Kletz. Learning from Accidents, 3rd ed. (Boston, MA: Butterworth-Heinemann, 2011).

T. A. Kletz. Plant Design for Safety (New York, NY: Hemisphere Publishing, 1991).

T. Kletz. What Went Wrong? Case Histories of Process Plant Disasters and How They Could Have Been Avoided, 5th ed. (Boston, MA: Butterworth-Heinemann, 2009).

S. Mannan, ed. Lees’ Loss Prevention in the Process Industries, 4th ed. (London, UK: Butterworth- Heinemann, 2012).

S. Mannan, ed. Lees’ Loss Prevention in the Process Industries, 8th ed. (New York, NY: McGraw-Hill, 2017).

R. E. Sanders. Chemical Process Safety: Learning from Case Histories (London, UK: Butterworth-Heinemann, 2015).

Problems

14-1. All 20 RBPS elements are equally important and cannot be prioritized. Based on your personal and/or company experiences, what are two elements you have found to be inadequately addressed? State the background and reasons for these inadequacies.

14-2. As a follow-up to Problem 14-1 and based on your personal and/or company experiences, list several additional RBPS elements that you have personally found to be inadequately addressed and state the background and reasons for these inadequacies.

14-3. Review the CSB video on the October 21, 2016, MGPI Processing facility incident and describe (a) the incident and (b) the RBPS elements that were inadequately addressed. Explain why those elements were not addressed properly.

14-4. Review the CSB video on the December 2007 T2 Laboratories incident: “Runaway: Explosion at T2 Laboratories.” (a) Describe the incident. (b) Summarize the CSB recommendations.

14-5. Review the CSB video on the February 2008 Imperial Sugar Refinery explosions incident: “Combustible Dust: An Insidious Hazard.” (a) Describe the incident. (b) Summarize the CSB recommendations.

14-6. Review the CSB video on the ExxonMobil Refinery 2013 incident: “Dangerously Close: Explosion in West Texas.” (a) Describe the incident. (b) Identify which RBPS elements were inadequately addressed and why.

14-7. Review the CSB video: “CSB Video Excerpts from Dr. Trevor Kletz.” Describe Dr. Kletz’s major teachings.

14-8. Review the CSB Video: “Fire in Baton Rouge.” Describe (a) the incident and (b) the CSB findings.

14-9. It is recommended to show training videos to engineers and students to encourage them to recognize that safety is especially important. Review the Internet information concerning the video “Remember Charlie.” Describe the video and give your conclusions.

14-10. On the Internet, you will find a case history about a “Gas Station Fire.” Describe the incident and make recommendations to prevent this type of incident.

Additional homework problems are available in the Pearson Instructor Resource Center.

..................Content has been hidden....................

You can't read the all page of ebook, please click here login for view all page.
Reset