written report of some sort is usually required of the failure analyst, some sort of verbal presentation of the findings is usually in order as well. Even if there is no opportunity to do a formal presentation, speaking with the people who requested the work in person or on the phone can be helpful. Especially if the written format your company prefers is quite formal, the reader may misinterpret your findings or conclusions. A short summary of the salient points in spoken form, especially useful before the written report is delivered, can be Knowledge Requirements for Failure Analysts. Some companies hire people with technician backgrounds to perform failure analysis work. While an engineering degree is not necessari\ requred equired to perform this type of work in a competent manner, a wide range of skills and knowledge is Someone without an engineering background will probably take some years to develop most of the necessary skills to a level of adequate proficiency to work on a variety of components For non-corrosion failures, the basic skills required include understanding fundamental concepts in stress nalysis and mechanical property theory. While the analyst does not have to know how to perform complex stress analysis in a quantitative manner, he or she should be able to figure out where the high-stress areas are by looking at a part and having someone describe the function. This skill may be called qualitative or heuristic stress-analysis skill, to recognize unexpected failure locations or unexpected features on a subject component Since the 1950s, eng have been gathering data from mysterious failures that "should never ha happened, because the operating stresses were much lower than the known strength of the materials used to make the components. The competent analyst of structural failures must be familiar with the basic concepts of fracture mechanics, fatigue crack propagation, and the causes of residual stresses Failure analysts who work with fractures must be familiar with macrofractography. This skill is a foundation of all fracture analysis. Without skill in fractography, the wrong test location might be selected for microfractography and metallography. Evaluating a location unrelated to the crack initiation may be worse than not evaluating the material at all with these methods, because the key evidence may be destroyed during the destructive portion of the incorrect testing Metallography and interpretation of microstructures are also key skills. The analyst must be able to look at a microstructure and determine whether the material is typical of its supposed composition and specified processing. This implies knowledge of how to interpret phase diagrams and isothermal and continuous cooling curves. Basic understanding of crystallography and micro- and macroscale composition effects is also required It is difficult for someone to independently perform failure analysis work without basic literacy. People skills and understanding of human nature are also important for the failure analyst. Failures can bring out the worst in people. During the background-information collection process of a major failure, the analyst probably needs to ask questions that make people uncomfortable and defensive. It may be difficult for the analyst to determine whether correct information is being provided The analyst may need advice on how to encourage the peopl involved to tell what they know. Technical skill can help the analyst weed out some incorrect information whether intended to mislead or confuse, or given in ignorance Finally, some consideration of ethics is required of the failure analyst. Ethical issues involve decisions that may be difficult to make. Most people would probably not envy the whistle blowers who decided to inform government authorities that their companies were breaking environmental regulations and who caused fellow workers to lose their jobs. Many ethical issues are not clear. Confidentiality promised to a client by an outside failure analysis service provider may conflict with the engineer's duty to protect the public, if the attitude of the client is not in line with ethical principles. (In such a case, it may be helpful to remind the client of the consequences of a repeat failure. Anyone doing failure analysis work may wish to study some codes of ethics specifically written for engineers. There are also some interesting works on ethical systems that are conveyed by other means, including literature and the support of trusted colleagues and mentors The Failure Analysis Process: An Overview Debbie Aliya, Aliya Analytical eferences Thefileisdownloadedfromwww.bzfxw.com
written report of some sort is usually required of the failure analyst, some sort of verbal presentation of the findings is usually in order as well. Even if there is no opportunity to do a formal presentation, speaking with the people who requested the work in person or on the phone can be helpful. Especially if the written format your company prefers is quite formal, the reader may misinterpret your findings or conclusions. A short summary of the salient points in spoken form, especially useful before the written report is delivered, can be very helpful. Knowledge Requirements for Failure Analysts. Some companies hire people with technician backgrounds to perform failure analysis work. While an engineering degree is not necessarily required to perform this type of work in a competent manner, a wide range of skills and knowledge is required. Someone without an engineering background will probably take some years to develop most of the necessary skills to a level of adequate proficiency to work on a variety of components. For non-corrosion failures, the basic skills required include understanding fundamental concepts in stress analysis and mechanical property theory. While the analyst does not have to know how to perform complex stress analysis in a quantitative manner, he or she should be able to figure out where the high-stress areas are by looking at a part and having someone describe the function. This skill may be called qualitative or heuristic stress-analysis skill, to recognize unexpected failure locations or unexpected features on a subject component. Since the 1950s, engineers have been gathering data from mysterious failures that “should never have happened,” because the operating stresses were much lower than the known strength of the materials used to make the components. The competent analyst of structural failures must be familiar with the basic concepts of fracture mechanics, fatigue crack propagation, and the causes of residual stresses. Failure analysts who work with fractures must be familiar with macrofractography. This skill is a foundation of all fracture analysis. Without skill in fractography, the wrong test location might be selected for microfractography and metallography. Evaluating a location unrelated to the crack initiation may be worse than not evaluating the material at all with these methods, because the key evidence may be destroyed during the destructive portion of the incorrect testing. Metallography and interpretation of microstructures are also key skills. The analyst must be able to look at a microstructure and determine whether the material is typical of its supposed composition and specified processing. This implies knowledge of how to interpret phase diagrams and isothermal and continuous cooling curves. Basic understanding of crystallography and micro- and macroscale composition effects is also required. It is difficult for someone to independently perform failure analysis work without basic literacy. People skills and understanding of human nature are also important for the failure analyst. Failures can bring out the worst in people. During the background-information collection process of a major failure, the analyst probably needs to ask questions that make people uncomfortable and defensive. It may be difficult for the analyst to determine whether correct information is being provided. The analyst may need advice on how to encourage the people involved to tell what they know. Technical skill can help the analyst weed out some incorrect information, whether intended to mislead or confuse, or given in ignorance. Finally, some consideration of ethics is required of the failure analyst. Ethical issues involve decisions that may be difficult to make. Most people would probably not envy the whistle blowers who decided to inform government authorities that their companies were breaking environmental regulations and who caused fellow workers to lose their jobs. Many ethical issues are not clear. Confidentiality promised to a client by an outside failure analysis service provider may conflict with the engineer's duty to protect the public, if the attitude of the client is not in line with ethical principles. (In such a case, it may be helpful to remind the client of the consequences of a repeat failure.) Anyone doing failure analysis work may wish to study some codes of ethics specifically written for engineers. There are also some interesting works on ethical systems that are conveyed by other means, including literature and the support of trusted colleagues and mentors. The Failure Analysis Process: An Overview Debbie Aliya, Aliya Analytical References The file is downloaded from www.bzfxw.com
D Levy, Tools of Critical Thinking: Metathoughts for Psychology allyn and Bacon, 199 2. M. Kaku, Hyperspace, Oxford University Press, 1994 3. D P Dennies, Boeing Co., private communication Organization of a Failure Investigation Daniel P. Dennies, Ph. D, The Boeing Company Introduction on 27 JANUARY 1967, the crew of Apollo I was lost when a fire broke out inside their command module during a simulation run aboard their unfueled Saturn v launch vehicle. It was a tragic event that shook a nation into reconsidering the promise of President John F. Kennedy on 25 May 1961 that" we choose to go to the moon in this decade. Weeks later, during the federal investigation of the Apollo 1 fire, astronaut Frank Borman said it was a"failure of imagination. It was just not conceivable that men at this early stage of the space race could actually be expected to imagine all possible problems that could arise Two years, 5 months, and 23 days later, on 20 July 1969, the Lunar Module eagle landed on the moon. The Command Capsule, Service Module, and Lunar Module Eagle used for that historic space flight and landing had been improved by the rigorous and detailed failure investigation of the capsule from the Apollo 1 fire. A second capsule, the next in line, was also sacrificed to better understand the failure and to determine the root cause. During the failure investigation, many other design and manufacturing flaws were discovered, and improvements were suggested and implemented Failure investigation has come a long way since the apollo 1 fire. The men and women involved in failure investigation in the new millennium have an incredible array of sophisticated and powerful tools and equipment to assist them. Unfortunately, many failure investigations performed today are not the victims of a"failure of imagination,but the victims of a failure of organization. " Too many failure investigations are started without a clear and concise goal, direction, or plan. Somewhere along the way, an investigation may get off track and never achieve its fundamental purpose of discovering root cause The purpose of this article is to discuss the organization required at the outset of a failure investigation and to provide a methodology with some organizational tools. The main focus is on the problem-solving tool of fault tree analysis. The main point is that time spent in preparation and planning can save a lot of time and money and even help achieve a successful conclusion of the failure investigation Organization of a Failure Investigation Daniel P. Dennies, Ph D. The Boeing Company What is a failure? a good definition of a failure is the inability of a component, machine, or process to function properly Failures come in all shapes and sizes. They can be individual parts, entire machines, or a process. Broken down further, failures can be physical, paper, or thinking/cultural in origin. These possibilities expand the concept of a failure, because they introduce failures that can be touched and failures that cannot be touched. a broken crankshaft is an obvious failure. However, a specification that has poorly written requirements that fail to
1. D. Levy, Tools of Critical Thinking: Metathoughts for Psychology Allyn and Bacon, 1997 2. M. Kaku, Hyperspace, Oxford University Press, 1994 3. D.P. Dennies, Boeing Co., private communication Organization of a Failure Investigation Daniel P. Dennies, Ph.D., The Boeing Company Introduction ON 27 JANUARY 1967, the crew of Apollo I was lost when a fire broke out inside their command module during a simulation run aboard their unfueled Saturn V launch vehicle. It was a tragic event that shook a nation into reconsidering the promise of President John F. Kennedy on 25 May 1961 that “we choose to go to the moon in this decade.” Weeks later, during the federal investigation of the Apollo 1 fire, astronaut Frank Borman said it was a “failure of imagination.” It was just not conceivable that men at this early stage of the “space race” could actually be expected to imagine all possible problems that could arise. Two years, 5 months, and 23 days later, on 20 July 1969, the Lunar Module Eagle landed on the moon. The Command Capsule, Service Module, and Lunar Module Eagle used for that historic space flight and landing had been improved by the rigorous and detailed failure investigation of the capsule from the Apollo 1 fire. A second capsule, the next in line, was also sacrificed to better understand the failure and to determine the root cause. During the failure investigation, many other design and manufacturing flaws were discovered, and improvements were suggested and implemented. Failure investigation has come a long way since the Apollo 1 fire. The men and women involved in failure investigation in the new millennium have an incredible array of sophisticated and powerful tools and equipment to assist them. Unfortunately, many failure investigations performed today are not the victims of a “failure of imagination” but the victims of a “failure of organization.” Too many failure investigations are started without a clear and concise goal, direction, or plan. Somewhere along the way, an investigation may get off track and never achieve its fundamental purpose of discovering root cause. The purpose of this article is to discuss the organization required at the outset of a failure investigation and to provide a methodology with some organizational tools. The main focus is on the problem-solving tool of fault tree analysis. The main point is that time spent in preparation and planning can save a lot of time and money and even help achieve a successful conclusion of the failure investigation. Organization of a Failure Investigation Daniel P. Dennies, Ph.D., The Boeing Company What Is a Failure? A good definition of a failure is “the inability of a component, machine, or process to function properly.” Failures come in all shapes and sizes. They can be individual parts, entire machines, or a process. Broken down further, failures can be physical, paper, or thinking/cultural in origin. These possibilities expand the concept of a failure, because they introduce failures that can be touched and failures that cannot be touched. A broken crankshaft is an obvious failure. However, a specification that has poorly written requirements that fail to
discern intended criteria may go unnoticed for years. In some cases, the reason things go wrong can be traced back to the culture of a company Always remember to look at each failure as unique and different. Considering a failure to be unique is sometimes difficult when the failure looks just like the one done before. Resist the temptation to immediately assign the same root cause to failures that look alike Another good point to remember is that failures are specific to an industry and the specific requirements of that industry. It is necessary to be knowledgeable about defining the requirements of the failure at hand. The saying one mans trash is another man,s treasure"applies to failures. For example, thin dense chrome plating on Unified Numbering System(UNS)52100 martensitic steel bearings without a hydrogen bake has been used in commercial applications with great success when replacing nonplated UNS 52100 martensitic steel bearings However, once that component moved into aerospace applications as a low-cost replacement for American Iron and Steel Institute 440C stainless steel, the requirements changed, and the lack of a hydrogen bake became a liability Failures in Physical Function and Expected Performance. In a very broad view, a failure can be defined by two categories. The first category is a functional failure, when a component, machine, or a process fails and everything stops. Functional failure of a component, such as a bearing or bracket, is very obvious. Functional failure of a machine can apply to something as large as an airplane or as small as a nutcracker. The difficulty in determining the root cause of the failure increases with the complexity of the component or machine, but the root cause can usually be discovered. Functional failure of processes such as heat treatment, coating, plating, welding, mail delivery, or airline scheduling can be more complex, and the determination of the root cause is also more complex. In addition, the cause-and-effect relationship for processes may not be readily apparent The second category is a failure of expected performance, when a component, machine, or a process fails to achieve performance criteria such as life, operating limits, and specification requirements. People have an expectation for the life of certain items, such as car tires, household appliances, car engines, biotech implants, and even pens. These life criteria usually are not written down but are debated in courtrooms. More specific criteria are operating limits, such as gas consumption, production scrap rate, gas flow in a valve, modem speed, specification turnaround time, computer-chip speed, or acquisition frequency. These items usually are well defined and, in most cases, mutually agreed upon by the parties involved. Finally, specifications requirements, such as mechanical properties, plating thickness, weight, and coating optical properties, are very well defined in written documentation. The specific definitions or limits are usually created by the customer, and tests are required to ensure compliance Organization of a Failure Investigation Daniel P. Dennies, Ph. D, The Boeing Company Why Do Failures Happen? Why do failures happen? is an often-asked question, and, over the years, many categories have been devised to classify the general causes of failure. The most common reasons for failures include Service or operation conditions(use and misuse) Improper maintenance(intentional or unintentional) Improper testing or inspection Assembly errors Fabrication/manufacturing errors Design errors(stress, materials selection, and assumed material condition or properties) Today, service or operation is the first suspect when a failure occurs. As noted earlier, most components and machines, even processes, have a certain life expectancy. Under normal use, most items wear out in time. A failure occurs when an item wears out in a shorter time than is expected by the user. At this point, however, it is important to distinguish between normal use and misuse. Pencils and cars can be used in a normal manner and Thefileisdownloadedfromwww.bzfxw.com
discern intended criteria may go unnoticed for years. In some cases, the reason things go wrong can be traced back to the culture of a company. Always remember to look at each failure as unique and different. Considering a failure to be unique is sometimes difficult when the failure looks just like the one done before. Resist the temptation to immediately assign the same root cause to failures that look alike. Another good point to remember is that failures are specific to an industry and the specific requirements of that industry. It is necessary to be knowledgeable about defining the requirements of the failure at hand. The saying “one man's trash is another man's treasure” applies to failures. For example, thin dense chrome plating on Unified Numbering System (UNS) 52100 martensitic steel bearings without a hydrogen bake has been used in commercial applications with great success when replacing nonplated UNS 52100 martensitic steel bearings. However, once that component moved into aerospace applications as a low-cost replacement for American Iron and Steel Institute 440C stainless steel, the requirements changed, and the lack of a hydrogen bake became a liability. Failures in Physical Function and Expected Performance. In a very broad view, a failure can be defined by two categories. The first category is a functional failure, when a component, machine, or a process fails and everything stops. Functional failure of a component, such as a bearing or bracket, is very obvious. Functional failure of a machine can apply to something as large as an airplane or as small as a nutcracker. The difficulty in determining the root cause of the failure increases with the complexity of the component or machine, but the root cause can usually be discovered. Functional failure of processes such as heat treatment, coating, plating, welding, mail delivery, or airline scheduling can be more complex, and the determination of the root cause is also more complex. In addition, the cause-and-effect relationship for processes may not be readily apparent. The second category is a failure of expected performance, when a component, machine, or a process fails to achieve performance criteria such as life, operating limits, and specification requirements. People have an expectation for the life of certain items, such as car tires, household appliances, car engines, biotech implants, and even pens. These life criteria usually are not written down but are debated in courtrooms. More specific criteria are operating limits, such as gas consumption, production scrap rate, gas flow in a valve, modem speed, specification turnaround time, computer-chip speed, or acquisition frequency. These items usually are well defined and, in most cases, mutually agreed upon by the parties involved. Finally, specifications requirements, such as mechanical properties, plating thickness, weight, and coating optical properties, are very well defined in written documentation. The specific definitions or limits are usually created by the customer, and tests are required to ensure compliance. Organization of a Failure Investigation Daniel P. Dennies, Ph.D., The Boeing Company Why Do Failures Happen? “Why do failures happen?” is an often-asked question, and, over the years, many categories have been devised to classify the general causes of failure. The most common reasons for failures include: · Service or operation conditions (use and misuse) · Improper maintenance (intentional or unintentional) · Improper testing or inspection · Assembly errors · Fabrication/manufacturing errors · Design errors (stress, materials selection, and assumed material condition or properties) Today, service or operation is the first suspect when a failure occurs. As noted earlier, most components and machines, even processes, have a certain life expectancy. Under normal use, most items wear out in time. A failure occurs when an item wears out in a shorter time than is expected by the user. At this point, however, it is important to distinguish between normal use and misuse. Pencils and cars can be used in a normal manner and The file is downloaded from www.bzfxw.com
have an expected life. Misuse, such as chewing on the pencil or constantly driving a car at high speeds or on gh roads, may reduce the life of eltner item Improper maintenance reaches all parts of our lives. The appliances in our homes, our computers, our cars, ell as the aircraft we fly, are required to be maintained. Improper maintenance results in life reduction complete breakdown. Sometimes it is intentional, such as skipping the overhaul of a machine in order to squeeze out a few more weeks of use. Other times it is unintentional, such as the case where the person who believes that if a 50 to 50 ratio of radiator fluid to water is good then 100% radiator fluid must be better Another common outcome of poor maintenance is corrosion. The monetary loss due to corrosion in the United States has been estimated at $60 billion to $100 billion a year Improper testing can also be a root cause of failures. The cost-driven, " hire and fire "business philosophy prevalent today drives work to be performed by the lowest-cost personnel capable of performing the task Sometimes, these personnel have just enough training, and sometimes, they have too little training Unfortunately, this business attitude can foster a situation in which the personnel performing the task may cause a failure by selecting the wrong test, by performing the test incorrectly, or by reviewing the test incorrectly. A favorite example concerns bird-strike testing on jet engines. Bird strikes have been a problem during aircraft takeoffs and landings for many years. a bird may be ingested into an engine at takeoff or landing. The engine must be able to handle a bird strike without an uncontained failure, fire or engine mount failure. An engine must be able to shut down in a controlled fashion to pass this U.S. Federal Aviation Administration requirement. The jet engine industry has a test to determine the effect of a bird strike. Frozen chickens are thawed out and shot from a cannon into a jet engine running on a test stand to determine if the engine stays within its nacelle. One day, a new technician was running the test, and apparently no one told him to thaw the frozen chicken before using it in the test. Needless to say, the engine failed the test as this frozen chicken hit the engine with the destructive impact of a bowling ball Assembly errors, fabrication/manufacturing errors, and design errors, are easier root causes to identify and are more commonly noted in the public domain through lawsuits and newspaper articles. The combination of fast and cheap with the use of low-cost personnel can be disastrous. Design and manufacturing engineers believe they can control the problem by making products and fabrication processes"idiot proof. However, time and time again, failures occur that prove them wrong Benefits of Statistics. It is in the best interest of each company to keep statistics on failures. As with most databases, computers make this task easier. The type of failure, the material, the root cause, and so on, can all be easily kept in a searchable database. Because each industry, even each company, is different, they need to keep their own statistics and draw on outside sources as applicable. The reason is simple. The statistics may provide a direction as to how a company can improve profitability by indicating where the company should spend money and manpower. No company can solve every problem, so it is prudent to attack the most technically detrimental or costly problems first. The statistics can provide this direction. In addition, statistics can point to trends over time. Lastly, comparison of company statistics to industrial data indicates if a problem is company specific or industry-wide Organization of a Failure Investigation Daniel P. Dennies, Ph. D, The Boeing Company Why Is a Failure Investigation Performed? n most instances, the purpose of a failure investigation is to determine the root cause(s). Determination of root cause is good engineering practice that crosses functional boundaries within a company and is an integral part of the quality assurance and continuous improvement programs. a proper failure investigation is not a"science project. " In addition, there is a distinct difference between a failure investigation and a metallurgical evaluation A root-cause failure investigation determines the root cause of the failure and recommends appropriate corrective actions. In contrast, a metallurgical evaluation provides the answers to the metallurgical questions sed in the failure investigation. Therefore. a metallurgical evaluation is only one part of a failure Investigation
have an expected life. Misuse, such as chewing on the pencil or constantly driving a car at high speeds or on rough roads, may reduce the life of either item. Improper maintenance reaches all parts of our lives. The appliances in our homes, our computers, our cars, as well as the aircraft we fly, are required to be maintained. Improper maintenance results in life reduction or complete breakdown. Sometimes it is intentional, such as skipping the overhaul of a machine in order to squeeze out a few more weeks of use. Other times it is unintentional, such as the case where the person who believes that if a 50 to 50 ratio of radiator fluid to water is good, then 100% radiator fluid must be better. Another common outcome of poor maintenance is corrosion. The monetary loss due to corrosion in the United States has been estimated at $60 billion to $100 billion a year. Improper testing can also be a root cause of failures. The cost-driven, “hire and fire” business philosophy prevalent today drives work to be performed by the lowest-cost personnel capable of performing the task. Sometimes, these personnel have just enough training, and sometimes, they have too little training. Unfortunately, this business attitude can foster a situation in which the personnel performing the task may cause a failure by selecting the wrong test, by performing the test incorrectly, or by reviewing the test incorrectly. A favorite example concerns bird-strike testing on jet engines. Bird strikes have been a problem during aircraft takeoffs and landings for many years. A bird may be ingested into an engine at takeoff or landing. The engine must be able to handle a bird strike without an uncontained failure, fire, or engine mount failure. An engine must be able to shut down in a controlled fashion to pass this U.S. Federal Aviation Administration requirement. The jet engine industry has a test to determine the effect of a bird strike. Frozen chickens are thawed out and shot from a cannon into a jet engine running on a test stand to determine if the engine stays within its nacelle. One day, a new technician was running the test, and apparently no one told him to thaw the frozen chicken before using it in the test. Needless to say, the engine failed the test as this frozen chicken hit the engine with the destructive impact of a bowling ball. Assembly errors, fabrication/manufacturing errors, and design errors, are easier root causes to identify and are more commonly noted in the public domain through lawsuits and newspaper articles. The combination of fast and cheap with the use of low-cost personnel can be disastrous. Design and manufacturing engineers believe they can control the problem by making products and fabrication processes “idiot proof.” However, time and time again, failures occur that prove them wrong. Benefits of Statistics. It is in the best interest of each company to keep statistics on failures. As with most databases, computers make this task easier. The type of failure, the material, the root cause, and so on, can all be easily kept in a searchable database. Because each industry, even each company, is different, they need to keep their own statistics and draw on outside sources as applicable. The reason is simple. The statistics may provide a direction as to how a company can improve profitability by indicating where the company should spend money and manpower. No company can solve every problem, so it is prudent to attack the most technically detrimental or costly problems first. The statistics can provide this direction. In addition, statistics can point to trends over time. Lastly, comparison of company statistics to industrial data indicates if a problem is company specific or industry-wide. Organization of a Failure Investigation Daniel P. Dennies, Ph.D., The Boeing Company Why Is a Failure Investigation Performed? In most instances, the purpose of a failure investigation is to determine the root cause(s). Determination of root cause is good engineering practice that crosses functional boundaries within a company and is an integral part of the quality assurance and continuous improvement programs. A proper failure investigation is not a “science project.” In addition, there is a distinct difference between a failure investigation and a metallurgical evaluation. A root-cause failure investigation determines the root cause of the failure and recommends appropriate corrective actions. In contrast, a metallurgical evaluation provides the answers to the metallurgical questions posed in the failure investigation. Therefore, a metallurgical evaluation is only one part of a failure investigation
Why Determine Root Cause? The most public reason to discover the root cause of a failure is to determine the fault or innocence of a company or person during litigation. Once fault is established, monetary assessment follows. For industrial purposes, however, it is more common, that once the root cause is discovered, the corrective action to prevent future occurrences is implemented, thus saving the company time and money. In ddition, once the root cause is discovered it is also possible to determine if the failure is a unique situation or the symptom of a widespread problem The benefits of a failure investigation that discovers the root cause of the failure include being a part of quality nanagement and continuous improvement programs, assisting in a redesign, solving a manufacturing problem, saving money, saving time, and maybe in some cases, preventing injuries and/or saving lives Organization of a Failure Investigation Daniel P. Dennies, Ph. D, The Boeing Company The Four-Step Problem-Solving Process The four-step problem-solving process that most engineers are taught is a good overall view of a failure nvestigation 1. What is the problem? 2. What is the root cause of the problem? 3. What are the potential solutions? 4. What is the best solution? Therefore, a good definition of a failure investigation can be an evaluation performed to identify and/or determine the reasons for a failure that clearly identifies the root cause(s) of the failure and recommends corrective action(s). Many failure investigations end after step 2(determine root cause). It is just as important to provide steps 3 and 4. Additionally, it would also be good to provide an evaluation of the solution. Many failure investigations discover the root cause(s)and provide a recommendation for corrective action(s), but no follow up evaluations are performed to see if the corrective action worked Organization of a Failure Investigation Daniel P Dennies, Ph D. The Boeing Company Nine Steps of a Failure Investigation There are nine steps to the organization of a good failure investigation. They are 1. Understand and negotiate goals of the investigation 2. Obtain clear understanding of the failure 3. Objectively and clearly identify all possible root causes 4. Objectively evaluate likelihood of each root cause 5. Converge on the most likely root cause(s) 6. Objectively and clearly identify all possible corrective actions 7. Objectively evaluate each corrective action 8. Select optimal corrective action(s) 9. Evaluate effectiveness of selected corrective action(s) Thefileisdownloadedfromwww.bzfxw.com
Why Determine Root Cause? The most public reason to discover the root cause of a failure is to determine the fault or innocence of a company or person during litigation. Once fault is established, monetary assessment follows. For industrial purposes, however, it is more common, that once the root cause is discovered, the corrective action to prevent future occurrences is implemented, thus saving the company time and money. In addition, once the root cause is discovered, it is also possible to determine if the failure is a unique situation or the symptom of a widespread problem. The benefits of a failure investigation that discovers the root cause of the failure include being a part of quality management and continuous improvement programs, assisting in a redesign, solving a manufacturing problem, saving money, saving time, and maybe in some cases, preventing injuries and/or saving lives. Organization of a Failure Investigation Daniel P. Dennies, Ph.D., The Boeing Company The Four-Step Problem-Solving Process The four-step problem-solving process that most engineers are taught is a good overall view of a failure investigation: 1. What is the problem? 2. What is the root cause of the problem? 3. What are the potential solutions? 4. What is the best solution? Therefore, a good definition of a failure investigation can be an evaluation performed to identify and/or determine the reasons for a failure that clearly identifies the root cause(s) of the failure and recommends corrective action(s). Many failure investigations end after step 2 (determine root cause). It is just as important to provide steps 3 and 4. Additionally, it would also be good to provide an evaluation of the solution. Many failure investigations discover the root cause(s) and provide a recommendation for corrective action(s), but no followup evaluations are performed to see if the corrective action worked. Organization of a Failure Investigation Daniel P. Dennies, Ph.D., The Boeing Company Nine Steps of a Failure Investigation There are nine steps to the organization of a good failure investigation. They are: 1. Understand and negotiate goals of the investigation 2. Obtain clear understanding of the failure 3. Objectively and clearly identify all possible root causes 4. Objectively evaluate likelihood of each root cause 5. Converge on the most likely root cause(s) 6. Objectively and clearly identify all possible corrective actions 7. Objectively evaluate each corrective action 8. Select optimal corrective action(s) 9. Evaluate effectiveness of selected corrective action(s) The file is downloaded from www.bzfxw.com