Andy Duncan
Research Paper
English H102-100
Dr. Christine Boese
12/3/98
The Bhopal Disaster: Political Ramifications and Industry's Application of Technology Toward Human-Centered Design

 In the twilight hours of the morning on December 3, 1984, forty-one metric tons of methyl isocyanate, also known as MIC, burst from a gas scrubber located at a Union Carbide pesticide plant near Bhopal, India.  The cloud of gas dispersed across the plant grounds and, by morning the toxic fog enveloped most of the area in and around Bhopal.  Officially, after one year, 1,800 Bhopal residents had died and 320,000 innocent Indian citizens had been affected by the disaster, yet, even now thirty to fifty people per month die from adverse effects of the disaster (Lepkowski 1985, Lepkowski 1991).  Alarmingly, even the event that was deemed the worst industrial disaster in history was settled for a price.  Numerous legal proceedings followed the fallout in Bhopal.  The Union Carbide parent company, located in the U.S., denied responsibility for the accident, blaming its Indian division, Union Carbide India Ltd. (UCIL), for carelessness and faulty design and construction.  In fear of economic isolation, due to Indian market abandonment by potential businesses and their reactions to Union Carbide's loss of face in the Bhopal incident, the Indian government agreed upon a $470 million liability settlement with Union Carbide for all inclusive damages in 1989.  In a 1991 report in Chemical & Engineering News, by Wil Lepkowski, over 4000 people were counted dead as a result of the Bhopal disaster. I have done some comparative math on the situation.  Each death in Bhopal, plus thousands of still-born incidents and fetal deaths, and the pain and suffering for thousands of people, all considering the 1989 settlement, was worth around $1000 per person. Rights activists for the victims of the incident feel that the government's decision was based more upon economic ramifications for the Indian business market, than a social stance that would protect and compensate the victims (Lepkowski 1991).  The people are suspicious of these dealings with the human element of the disaster as casualties at only a monetary price.  I agree on the atrocity of the fact that such an incident has actually put a price on human life and suffering.
 
The chemical process industry (CPI), the business market related to the Bhopal disaster, is the type of industry responsible for synthesizing base chemicals for components in commercial products.  Very quickly after the incident at Bhopal, safety audits became the order of the day, revealing the immediate concern that rose in the chemical process industry (Lepkowski 1985).  With these actions as the origin, safety issues in the chemical process industry have become a central focus in the last ten years.  Companies have attempted to better insure their assets through systems such as process safety management and qualitative and quantitative risk analysis, while governments have participated at the minimum level concerning regulation standards.  Process safety management is a system ensuring the easy flow of critical knowledge, pertaining to a complex system, throughout the hierarchy of participants in that system(Schreiber 1991).  Qualitative and quantitative risk analysis are methods of predicting possible failures within technical systems through analysis of past disasters and possible scenarios.  The interesting fact among all the resulting activity originating with Bhopal, is that the public, being those who are employed and affected by the chemical process industry in so many ways, are being shut out of developments in the industry.  Research studies by various engineers and scientists within the industry seem to focus on the immediate financial risks and long-term economic results of possible chemical disasters such as Bhopal.  With encouragement from governments and in the face of economically critical ramifications, the chemical process industry is turning away from the human laborer in systems, toward application of safer and more efficient technology in hopes of more human-centered design.

Government-Carbide Correlation

 The major conflict that still resides in Bhopal is many faceted in its degrees of seriousness for the people, the government, and Union Carbide.  The Indian people of Bhopal wish for compensation for the simple lifestyle and peace they lost in the Bhopal disaster through further action against Union Carbide.  In May of 1991, two Indian victim's rights activists visited the U.S. to meet with Union Carbide and request more funding and research to heal the still open wounds of their little piece of India.  The activists, a Mr. Chouhan and Mr. Sarangi, had several requests: $900 million more in liability payments, Carbides confession that cyanide was also a factor in the disaster, and the release of critical chemical research that could help in understanding the toxic effects of the spill and the possible cures.  Union Carbide denied all these demands, continuing on the story originally set up and agreed upon with the Madhya Pradesh government in 1989.  At that time, Carbide paid the $470 million in liabilities, took the position of sabotage as the reason for the accident, and denied claims to the presence of cyanide in the gas release.  I believe the difficulties that the activists encounter with Union Carbide are possibly caused by an underlying economic motive shared by the chemical process industry, especially Union Carbide, and concerning the Indian government.  The still present name for Union Carbide in and around Bhopal is "Killer Carbide."   The people hold great resentment for the company, yet, they encounter the reality that they need Union Carbide's funding, while desiring its continued punishment, to heal the community of Bhopal and prevent another disaster.  The Indian government has conflicts with itself concerning the situation.  The industrial community in Indian society, in correlation with the Indian government, does not wish to reopen the Carbide lawsuit in fear of setting high financial standards for liability responsibilities in their own plants.  Such action could disrupt the industry's business flow with growth of new institutions for bureaucracy and regulation.  This underlying partnership also holds the fear that if the Indian judicial system punishes Union Carbide too harshly that western investors may avoid India as a possible market in fear of high regulations and opposition by the Indian citizens (Lepkowski 1991).

One of the possible factors in the Bhopal disaster is the concept of blind technology transfer. This concept means establishing a highly technological chemical process factory in a technologically slow region or society.  This theory of including high levels of technology in a society that has not experienced the appropriate levels of technological evolution may have taken place in Bhopal with Union Carbide.  The Indian government, in its interest of establishing economic progress, may have allowed Union Carbide to set up shop in a community that could not comprehend its technological components (Lanza 1985).  This event would explain the carelessness in construction and design of the plant.  It would also disregard Union Carbide's sabotage theory, by introducing ignorance or lack of ability as an explanation, and raise ethical questions within the chemical process industry.  This theory may explain why the Indian government continually sides with Union Carbide, creating a concealing wall against the victim's rights activists.  Instead of a technological evolution occurring in Bhopal, the government's actions created a toxicology revolution and resentful feelings toward government involvement.  The example set by the mistakes of the Indian government-Union Carbide correlation in Bhopal have spurred a movement in the chemical process industry to insure participants against similar consequences of industrial accidents.  The main actions within this movement have included research into process safety management, qualitative risk analysis, and quantitative risk analysis (Kim, Moore, Ondrey 1991, Schreiber 1991).

Process Safety Management

Through the litigation between the Indian government, rights activists for victims, and Union Carbide, the chemical process industry has begun a process to determine on whom to place responsibility for major disasters and accidents within the business.  Ultimately, responsibility for safe operations has become the burden of management.  It is the responsibility of management to have control and knowledge of the process design and operation.  Proper management of complex systems is crucial to success within the chemical process industry.  With this idea in heavy consideration, the chemical industry has said it is the main responsibility of management to equip employees in the system with enough knowledge to improvise in crucial, possibly dangerous situations (Bell 1989).

The Center for Chemical Process Safety (CCPS) was formed in view of the newly defined responsibilities placed on management within the industry, and the changing aspects of the chemical process business.  The CCPS conducted research and began to outline a strategy for management throughout the chemical processing industry.  Management would now employ a more consistent commitment to the technical elements of safety processes.  Process knowledge obtained by management would be captured and made readily available to all participants in the process, thus, encouraging safety throughout the entire system.  This is process safety management.  I view this concept as almost childishly simple, but when applied to a massive network of technological and human interaction, this ideal can be easily lost in the noise. The ideal of process safety management ensures that regulation and safety technology works in the system, before risk analysis of system scenarios and possibly resulting disasters begins.  The new management system is ultimately directed toward a set goal concerning safety and efficiency.  Everyone becomes involved and, thus, takes part in ensuring the economic success and safety of the entire system.  The senior executives create the goal.  Management displays, explains, and directs the workers toward the goal.  At the end of the process, the workers make the goal a reality.  The difficulty of applying such a system to the present chemical process industry is the aspect of investment.  The idea of application raises such questions as: How effective is investing in a Process Safety Management program?  How will the costs and benefits be quantified for usage as proof that the system works?  What procedures and methods for assessment and comparison to industry peers will be suitable? The universal problem with such ideals as process safety management is that they require a significant investment of resources to reveal their true benefits, yet, the ideal must show its benefits to convince investors to of the soundness of their investment (Schreiber 1991).  The main focus of applying process safety management concepts to the chemical process industry seems to lye in an economic sphere of influence.  I think that this concept could be one way to satisfy both the industrial community and the inhabitants of Bhopal.  The people would become better informed, thus, breaking down the process of blind technology transfer while beginning a economically beneficial technological evolution.  If the people could accept an economic solution as a way to heal the wounds at Bhopal, while making them a more influential part of the combined industrial and social community, lives could be restored and continue.  The problem is the painful scars, physically and emotionally, that remind the people of Bhopal, the chemical process industry, and the world of the need to realize and reduce potential risk where industry interacts with its surrounding environment.

 Qualitative Risk Analysis

Since the Bhopal disaster, the majority of research for increased safety and efficiency in industry accidents has focused on qualitative and quantitative risk analysis.  Qualitative risk analysis concentrates on learning problem-solving techniques from past accidents and disasters, while creating solutions and processes for future occurrences.  This type of risk analysis is limited to the types of scenarios that exist in the present and the past (Stix 1989).  Another limit has been placed on the advance of qualitative risk analysis by the governments setting regulations for the chemical process industry.  The U.S. government's Environmental Protection Agency (EPA) proposed a risk management program only in the event of "worst case scenarios."  This was described as the complete release of hazardous chemical substances from a chemical process facility.  The government's high level of misinformation and ignorance on the subject of preventing such disasters is shown paramount by this action.  Research into such a scenario has been performed by engineers and has been determined to be impossible.  Even Isadore Rosenthal, a senior research fellow at Wharton School's Risk Management & Decision Process Center, agrees that the true risk is questionable when he says, "Although deaths have occurred within plants, no one has been killed outside a U.S. plant during an accident in the past 50 years."  This quotation by Rosenthal seems to imply that either a major disaster must occur for the public, industry, and government to notice or, the workers' lives in the plants are worth less than the inhabitants in the surrounding environment (Kim, Moore, Ondrey, 1995).  Rosenthal's position on the subject, combined with the abandonment of Union Carbide India Ltd. by the parent Union Carbide (U.S.) in the Bhopal litigation, seems to suggest that U.S. workers are worth more than workers outside the Melting Pot, even in U.S. affiliated international industry.  This contradiction of the U.S. concept of uplifting diversity and of engineering ethics, the desire to protect, aid, and value all society, suggests to me that more work should be spent on standardizing and strengthening international business relations instead of just fixing the machinery within the system.
 

Human Error Contribution
Human error in chemical process systems has become a major documented source for the cause of disasters in complex systems.  One of the major aspects to be learned from qualitative risk analysis is how human error contributes to a potentially hazardous situation.  Presently, technological institutions consist of complex control systems with few human workers.   Near eighty percent of failures in complex systems occur primarily due to complex coincidences caused by neglect or misunderstanding of the implications of an event, by a human worker.  The solution to this massive origin in human error is to design systems that compensate and lower the risk that frequently occurs in industry due to ignored warnings, careless risks, sloppiness, deception, and lying.  Note, the entire system does still consist of both human and technological components.  In his research on minimizing human-machine interface failure, John J. Sudano of Martin Marietta, suggests this process:

 simplify the design, reduce the [human]-Machine interaction, separate and form a hierarchy of the system variables, design the system to support the human decision maker, don't force the human to support the system, insulate or spatially separate or bundle the hardware, overlap the system sensors, provide integral system maintenance, build 'slower' systems, assure rapid error detection and recovery of critical functions, overdesign the system.

 
This new system would allow compensation by the system for human error before dangerous consequences occur (Sudano 1994).  One of the possible causes determined for the Bhopal disaster was a missing filter/separator, called a slip blind, forgotten by a worker to be inserted in the system (Lepkowski 1985).  If these ideals had been in place at the Union Carbide plant in Bhopal, the missing slip blind would likely not be a possible cause.

 The study done by Sudano seems to suggest that the human element of systems is the root of the problem.  As was determined through the Bhopal incident, informed management of the design and construction of complex chemical process systems is crucial to success.  I think that the human element within the system is not the problem.  From a study by Donald A. Norman, entitled Toward Human-Centered Design, it seems that the design of technological systems is at the root of accidents and the high risks present in our technological society.  The original purpose of machinery and technology was to complement human abilities.  People are good at creating and perceiving problems, while technology is useful for solving or making problems easier.  This would work except for the direction which design engineers take to applying technology to solve problems. Norman states in his essay, "We are capable of building machines that can perform flawlessly many of the things we are bad at but it is very difficult to build machines that can do the things we are good at."  The trouble now is that engineers do not first ask whether a task would be better performed by a human or a machine, but if a machine could be designed to do the task.  The general problem that inspired the need for the technological fix is no longer the focus of engineering.  Presently, application is the highest priority of engineers.  The excitement of applying technology has blinded engineers to the clumsy situation in which they put human operators within a system.  Humans are forced to adapt to the needs of the machinery, instead of the machinery leaving control in human hands.  This seems to be the prevalent reason that human error in complex systems has been calculated at such high percentages.  If humans are a component of the system, yet we have no control, then the machine can never be wrong.  We cannot react to the invisible information which machines communicate through, until the error lies in the hands of the operator.  This awkward situation within industry could be solved by creating technology that raises the level of human comprehension within a system instead of demeaning the human component to menial, repetitive labor (Norman 1993).  If human needs, especially the original problem, can be met before the requirements of technology, then risk should decrease.

Quantitative Risk Analysis

 Especially in Europe and Asia, quantitative risk analysis (QRA) is becoming an institution in the chemical process industry.  Quantitative risk analysis is a method of analysis based on mathematical models that depict the fallout of possible scenarios, disasters and accidents, in the chemical process industry.  QRA allows companies to perform self-checks on their systems.  This allows for consistently available, up-to-date information, and, with the aid of computer-generated mathematical models, new scenarios and a wider sphere of knowledge.  Computer generation of mathematical models allows predictions concerning the effects of various variables on possible system scenarios.  This technology has been employed to analyze the effects of prevailing meteorological conditions on the fallout during the Bhopal disaster.  After these observations, qualitative risk analysis could be performed upon the scenarios to learn how to handle new, similar situations that might occur in the chemical process industry.  This process of risk analysis does not necessarily solve accidents that have already occurred but, instead, it gathers scenario information and statistics that can be applied to create new quantitative scenarios for future use.  The research and implementation of QRA is still in beginning stages.  The present problem with the process is that it cannot yet handle the combination of all possible variables in a scenario.  This characteristic gives countless separate scenarios with a myriad of probabilities, yet, not a conclusive combination for easier application to everyday situations.  Such a problem occurred when quantitative risk analysis was attempted for fabricated accidents in the Netherlands.  These models would test the risk present in a system based in the Netherlands' environment, while gathering information on the possible consequences of fallout.    Meteorological conditions could be factored into the model, but it could not also handle the various effects of present land structures (i.e. cliffs).  There is, as industry improves its safety measures, a good problem for QRAs.  Basis information for risk analysis, overall, shrinks as less data of failure is produced by the industry.  Presently QRA gives predictions for overall risk within an accuracy of a factor of ten based on the mathematical models.  This amount of error provides information, but not yet reliability.  The best way that researchers in industry have seen to implement QRA is to concentrate on the idea that, "The analysis must be a living document, not something you do once and put away."  QRA must remain as a process that produces an infinite set of scenarios to continue to push industry toward 100% safety and efficiency (Joyce 1986, Kim, Moore, Ondrey 1995).

Financial Ramifications, then the Worker?
 The all-encompassing situation concerning Bhopal seems to be an indicator of the CPI's outlook on its place and influence within international society.  The fact that the Indian government settled so easily and quickly, even in the face of continuing concerns due to the Bhopal disaster, concerns me.  It appears that the CPI is not completely uniform, between its categorized businesses, in its view on priorities concerning such incidents as occurred at Bhopal with Union Carbide.  A study was done in 1996, by Flavio D. Penteado and Amy R. Ciric, both memebers of the Department of Chemical Engineering at the University of Cincinnati, Ohio, presenting a new approach to process plant design.  As I read the report, the aspect that seemed most important to the two authors was the financial risks involved for the industry and how they were effected by the new approach.  There was mention of an exchange between the appearance of risk and the increasing costs due to the employment of safety devices (Penteado, Ciric 1996).  The shocking part of the new approach was that safety precautions would be designed with the machines in mind, not the workers.  This sort of technology-centered design either implies the further elimination of the human component from complex systems, or suggests that human workers are worth less than the cost of repairing machinery.  This approach is very similar to what Norman sees as a problem in present day technology.  The worker in the system is forced to adapt to the needs presented by the machines in value and efficiency (Norman 1993).  The Penteado and Ciric study categorizes financial risk into two sections: financial risk associated with accidents at the machines in jeopardy, and financial risk due to the increased risk toward units surrounding the accident (Penteado, Ciric 1996).  Neither of these categories considers the financial ramifications toward the human component of the system.  I believe this could suggest that either, again, the human component is not worth enough, or that it costs too much to maintain, repair, and support in the event of an accident.  This idea once again suggests the removal of the human component from complex systems due to the needs of technology and economic concerns.  Norman's concept of human-centered design would supply for both of these needs while giving control and beneficial knowledge to the human component of the system.
Suspicions, Developments, Conclusions
 The incident at Bhopal and its continuing consequences seem to have created a race within the chemical process industry.  Especially concerning Union Carbide and the Indian government, there seems to be movement to concentrate on the economic ramifications for the company and the government, versus the victims and the activists that represent them.  Oppositely, there seems to be a realization within the chemical process industry that technology-centered design can no longer operate at the level present business conditions require for success due to the lesser abilities and comprehension displayed by the human component in chemical process systems.  These two scenarios create a conflict.  Union Carbide wishes to close the books on the Bhopal incident, still standing firm on its platform of sabotage as the cause of the disaster.  This would avoid laying the blame on the design and construction of the system, allowing for the presence of blind technology transfer.  If these conclusions can be avoided, Union Carbide and the Indian government can escape any further damages, punishment, or red tape from liability.  At the same time, the chemical process industry is realizing that the application of technology in concepts such as process safety management, qualitative risk analysis, and quantitative risk analysis could lead to a more efficient, more beneficial, safer, human-centered chemical processing system.  The newer system would limit human error not by eliminating the human component from the system, but by designing the system to be user friendly and user active.  Participants in the human component would be less in number but greater in knowledge, ability, and activity.  The two aspects of the conflict oppose on the same plane of influence.  Activity in the chemical process industry for more human-centered design must prove worthiness of investment, while opposed by the attempts to end discussion of liability in the Bhopal incident that would show the need for a technology-centered to human-centered shift.  I believe that human-centered design is the answer to revealing the faults at Bhopal, and to allowing technology to continue to grow while still considering safety and humanity in the face of economic ramifications.  If the technology or its design is never determined to be faulty, then mechanization could continue to grow, quantifying all aspects of humanity and eliminating the complement between creator and creation of technology.
 Bibliography

Bell, Trudy E. 1989. The limits of risk analysis. IEEE Spectrum, v26n6 June: p51.

Joyce, Edward J. 1986. To Stop Another Bhopal. Datamation, v32n5 March 1: p40-42, 44.

Kim, Irene., Moore, Stephen., Ondrey, Gerald. 1995. Risk & the CPI. Chemical Engineering, v102n2 February: p26-29.

Lanza, Guy R. 1985. Blind Technology Transfer: The Bhopal Example. Environmental Science & Technology, v19n7 July: p581-582.

Lepkowski, Wil. 1985. Special Report: Bhopal. Chemical & Engineering News, v63n48 December 2: p18-32.

Lepkowski, Wil. 1991. Indian Activists Press Bhopal Accident Issues in U.S. Visit. Chemical & Engineering News, v69n19 May 13: p22-23.

Norman, Donald A. 1993. Toward Human-Centered Design. Technology and the Future, 1997, ed.7 p246-253.

Penteado, Flavio D., Ciric, Amy R. 1996. MINLP approach for safe process plant layout. Industrial & Engineering Chemistry Research, v35n4 April: p1354-1361.

Schreiber, Sanford. 1991. Guidelines for Technical Management of Chemical Process Safety. Plant/Operations Progress, v10n2 April: p65-68.

Stix, Gary. 1989. Bhopal: a tragedy in waiting. IEEE Spectrum, v26n6 June: p47-50.

Sudano, John J. 1994. Minimizing human-machine interface failures in high risk systems. IEEE Aerospace and Electronic Systems Magazine, v9n10 October: p17-20.