Historical review
The AMEF discipline was developed in the United States Army by engineers from the National Agency of Space and Aeronautical (NASA), and was known as the MIL-P-1629 military procedure, entitled "Procedure for the Execution of a Mode of Failure, Effects and Criticibility Analysis »and prepared on November 9, 1949; This was used as a technique to assess reliability and to determine the effects of equipment and system failures on the success of the mission and the safety of personnel or equipment.
In 1988 the International Organization for Standardization (ISO) published the ISO 9000 series of standards for quality management and assurance; the requirements of this series led many organizations to develop quality management systems focused on the needs, requirements and expectations of the client, among these the QS 9000 emerged in the automotive area, this was developed by the Chrysler Corporation, the Ford Motor Company and the General Motors Corporation in an effort to standardize supplier quality systems; Under QS 9000 standards, automotive suppliers must employ Advanced Product Quality Planning (APQP), which must necessarily include design and process FMEAs, as well as a control plan.
Subsequently, in February 1993, the industrial automotive action group (AIAG) and the American Society for Quality Control (ASQC) registered the AMEF standards for their implementation in the industry, these standards are the equivalent to the technical procedure of the Society of Automotive Engineers SAE J - 1739.
The standards are presented in the FMEA manual approved and supported by Chrysler, Ford, and General Motors; This manual provides general guidelines for the preparation and execution of the FMEA.
Currently, the FMEA has become popular in all American automotive companies and has begun to be used in various areas of a wide variety of companies worldwide.
What is AMEF?
The Mode and Effects Analysis of potential failures, FMEA, is a systematic process to identify potential failures in the design of a product or process before they occur, in order to eliminate them or minimize the risk associated with the same.
FMEA can be considered as a standardized analytical method to detect and eliminate problems in a systematic and total way.
Main objectives AMEF:
• Recognize and evaluate potential failure modes and causes associated with product design and manufacturing
• Determine the effects of potential failures on system performance
• Identify actions that may eliminate or reduce the opportunity for failure to occur potential
• Analyze system reliability
• Document the process
Although the FMEA method has generally been used by the automotive industries, it is applicable for the detection and blocking of the causes of potential failures in products and processes of any kind of company, whether they are in operation or in the phase of draft; as well as it is applicable for administrative and service systems.
AMEF requirements
To do an FMEA the following is required:
• A team of people with the commitment to improve the design capacity to satisfy the client's needs.
• Schematic and block diagrams of each level of the system, from sub-assemblies to the complete system.
• Component specifications, parts list, and design data.
• Functional specifications of modules, subassemblies, etc.
• Manufacturing requirements and details of the processes to be used.
• AMEF forms (paper or electronic) and a list of special considerations that apply to the product.
AMEF benefits
Eliminating potential failure modes has both short-term and long-term benefits.
Short-term: represents savings on repair costs, repetitive testing, and downtime.
In the long term: it is much more difficult to measure since it relates to customer satisfaction with the product and its perceptions of quality; This perception affects future purchases of products and is decisive in creating a good image of them.
AMEF supports and reinforces the design process since:
• Assists in the selection of alternatives during design
• Increases the likelihood that potential failure modes and their effects on system operation will be considered during design
• Provides additional information to aid in planning conscientious testing programs and efficient
• Develops a list of potential failure modes, ranked according to their likely effect on the customer
• Provides an open documented format to recommend actions that reduce risk to track them
• Detects failures where self-correcting characteristics are needed or of slight protection
• Identifies known and potential failure modes that might otherwise go unnoticed
• Detects primary, but often minimal, failures that can cause certain secondary failures
• Provides a fresh perspective on understanding the functions of a system.
Format and elements of the AMEF:
To facilitate the documentation of the analysis of potential failures and their consequences, the Ford company standardized a format for carrying out the FMEA:
However, since each company represents a particular case, it must be prepared by a multidisciplinary team made up of personnel with experience in design, manufacturing, assembly, service, quality and reliability.
It is very important that, even when modifications are made, the following elements are maintained:
Header
• Type of FMEA: it must be specified if the FMEA to be carried out is design or process.
• Name / Part Number Or Process: the name and number of the part, assembly or process that is being analyzed must be registered. Use suffixes, change letters and / or the Change Report / Change Request (CR / CR) number, as appropriate.
• Design / Manufacturing Responsibility: Record the name of the operation and manufacturing plant that has primary responsibility for the machinery, equipment or assembly process, as well as the name of the area responsible for the design of the component, assembly or system involved.
• Other Areas Involved: List any area / department or organizations affected or involved in the design or function of the component (s), as well as other manufacturing operations or plants involved.
• Suppliers and Affected Plants: List any supplier or manufacturing plants involved in the design or manufacture of the components or assemblies that are being analyzed.
• Vehicle (S) / Model Year (depends on where it is being made): Records all the vehicle lines that will use the part / process being analyzed and the model year.
• Engineering Release Date: Indicates the last Engineering Release level and date for the component or assembly involved.
• Key Production Date: Record the appropriate production date.
• Prepared by: Indicating the name, telephone number, address and company of the engineer preparing the AMEF.
• AMEF Date: Enter the date on which the original AMEF was developed and subsequently, enter the date of the last revision of the AMEF.
Description / purpose of the process
Write down a simple description of the process or operation being analyzed and indicate as briefly as possible the purpose of the process or operation being analyzed.
Potential failure mode
It is defined as the way in which a part or assembly can potentially fail to meet engineering release requirements or specific process requirements. Each potential failure mode is listed for the particular operation; To identify all possible failure modes, it is necessary to consider that these can fall into one of five categories:
1. Total Failure
2. Partial
Failure 3. Intermittent
Failure 4. Gradual Failure
5. Overfunction
Effects of potential failure
The next step in the FMEA process, after defining the function and failure modes, is to identify the potential consequences of the failure mode; This activity must be carried out through brainstorming and once these consequences have been identified, they must be introduced into the model as effects.
It should be assumed that effects occur whenever failure mode occurs. The procedure for Potential Consequences is applied to record remote or circumstantial consequences, through the identification of additional failure modes, the identification procedure is as follows:
• It begins with a failure model (MF-1), and a list of all its potential consequences
• Separate those consequences that are assumed as a result whenever MF-1 occurs, these are identified as MF-1 effects
• Modes are written of additional failure for the remaining consequences (consequences that could result if MF-1 occurs, depending on the circumstances under which it occurs). New failure modes imply that unusual consequences will occur by including the circumstances under which they occur.
• Separate the assumed consequences will result whenever failure modes and their special circumstances occur; these should be identified as effects of the additional failure modes.
Severity
The first step in risk analysis is to quantify the severity of the effects, these are evaluated on a scale of 1 to 10 where 10 is the most severe.
Below are the tables with the evaluation criteria for process and design:
Table 1.- Evaluation criteria and suggested graduation system for the severity of the effects for an AMEF design.
| Effect | Criteria: Severity of effect for FMEA | Row |
| Dangerous alert | The incident affects the safe operation of the product or involves non-compliance with government regulation without alarm. | 10 |
| Dangerous with alarm | The incident affects the safe operation of the product or involves non-compliance with government regulation with the alarm. | 9 |
| Very high | The product is inoperable with loss of primary function. | 8 |
| Tall | The product is operable, but at the reduced level of performance | 7 |
| Moderate | The product is operable, but the item (s) of comfort or convenience is inoperable. | 6 |
| Low | The product is operable at a reduced level of operation. | 5 |
| Very low | Most customers notice the flaws. | 4 |
| Minor | Average customers notice the flaws. | 3 |
| Very minor | The setting, the item and the end of the rattle do not conform. The demanding ones notice the defects. | two |
| None | No defect | one |
Table 2.- Evaluation criteria and suggested graduation system for the severity of the effects in an AMEF process.
| Effect | Criteria: Severity of effect for FMEA | Row |
| Dangerous without alarm | It may endanger the operator of the assembly. The incident affects the operation or safe non-compliance of the product with government regulation. The incident will occur without alarm. | 10 |
| Dangerous with alarm | It may endanger the operator of the assembly. The incident affects the operation or safe non-compliance of the product with government regulation. The incident will occur with alarm. | 9 |
| Very high | Major disruption to the production chain, 100% of the product can be discarded. The product is inoperable with loss of primary function. | 8 |
| Effect | Criteria: Severity of effect for FMEA | Row |
| Tall | Minor disruption to the production chain. The product can be classified and one and a portion discarded. The product is operable, but at a reduced level of performance. | 7 |
| Moderate | Interruption is of minor importance to the production chain. A portion of the product can be discarded (not classified). The product is operable but some comfort / convenience item (s) is inoperable. | 6 |
| Low | Interruption, it is of minor importance to the production chain, 100% of the product can be returned to work. The product is operable, but some comfort / convenience items operate at a reduced level of performance. | 5 |
| Very low | Interruption is of minor importance to the production chain. The product can be classified and a portion can be returned to work. Most customers notice the defect. | 4 |
| Minor | Interruption is of minor importance to the production chain. A portion of the product can be returned to work online only out-of-season. Average customers notice the flaw. | 3 |
| Very minor | Interruption is of minor importance to the production chain. A portion of the product can be returned to work online only in-season. Demanding customers notice the flaw. | two |
| None | Failure mode has no effect. | one |
| Effect | Criteria: Severity of effect for FMEA | Row |
| Tall | Minor disruption to the production chain. The product can be classified and one and a portion discarded. The product is operable, but at a reduced level of performance. | 7 |
| Moderate | Interruption is of minor importance to the production chain. A portion of the product can be discarded (not classified). The product is operable but some comfort / convenience item (s) is inoperable. | 6 |
| Low | Interruption, it is of minor importance to the production chain, 100% of the product can be returned to work. The product is operable, but some comfort / convenience items operate at a reduced level of performance. | 5 |
| Very low | Interruption is of minor importance to the production chain. The product can be classified and a portion can be returned to work. Most customers notice the defect. | 4 |
| Minor | Interruption is of minor importance to the production chain. A portion of the product can be returned to work online only out-of-season. Average customers notice the flaw. | 3 |
| Very minor | Interruption is of minor importance to the production chain. A portion of the product can be returned to work online only in-season. Demanding customers notice the flaw. | two |
| None | Failure mode has no effect. |
Special features
The AIAG defines a special product feature as a characteristic product for which it reasonably anticipated the variation could significantly affect a product's safety or compliance with government standards or regulations, or is likely to significantly affect customer satisfaction with a product.
Ford Motor Company divides special features into two categories:
• Critical Characteristics - are defined by Ford as product or process requirements that affect compliance with government regulation or the safe function of the product, and that require special actions or controls.
In an FMEA design, the critical characteristics of the potential are considered. A potential critical feature exists for any severity rating greater than or equal to 9.
In the AMEF process, they are referred to as actual critical characteristics, any characteristic with a severity of 9 or 10 that requires special control to ensure detection is a critical characteristic.
Examples of product or process requirements that could be critical features include dimensions, specifications, testing, assembly sequences, tooling, joints, torques, welds, connections, and component applications. The special actions or controls necessary to meet these requirements may involve manufacturing, assembly, a supplier, shipping, monitoring, or inspection.
• Significant characteristics: they require special controls because they are important for customer satisfaction. The degrees of severity between 5 and 8 were combined with an occurrence that classified greater than 3 indicate significant characteristics. In an FMEA design, Significant Features are potential. In the FMEA process, if special control is required to ensure detection then an actual significant characteristic exists. Companies have not standardized a method for grouping and denoting special product characteristics. The nomenclature and notation will vary.
Causes of potential failures
After the effects and severity have been listed, the causes of the failure modes should be identified.
In design AMEF, the causes of failure are design deficiencies that produce a failure mode. For the process FMEA, the causes are specific errors described in terms of something that can be corrected or controlled.
Idea
The causes are evaluated in terms of occurrence, this is defined as the probability that a particular cause occurs and results in a failure mode during the expected life of the product, that is, it represents the remote probability that the customer experiences the effect failure mode.
The value of the occurrence is determined through the following tables, in case of obtaining intermediate values, the immediate superior is assumed, and if the probability of failure is totally unknown, an occurrence equal to 10 must be assumed.
Table 3.- Evaluation criteria and suggested graduation system for the occurrence of the incident in an AMEF design.
| Incident probability | Failure rate | Row |
| High above: the incident is almost inevitable | 1 in 2 ³ | 10 |
| 1 in 3 | 9 | |
| High: repetitive incidents | 1 in 8 | 8 |
| 1 in 20 | 7 | |
| Moderate: occasional incidents | 1 in 80 | 6 |
| 1 in 400 | 5 | |
| 1 in 2000 | 4 | |
| Low: relatively few incidents | 1 in 15,000 | 3 |
| 1 in 150,000 | two | |
| Remote control: the incident is implausible | 1 in £ 1,500,000 | one |
Table 4.- Evaluation criteria and suggested graduation system for the occurrence of the incident in an AMEF process.
| Incident probability | Incident rates | C pk | Row |
| High above: the incident is almost inevitable | 1 in 2 ³ | <0.33 | 10 |
| 1 in 3 | 0.33 ³ | 9 | |
| High - Generally associated with similar processes that have previously failed | 1 in 8 | 0.51³ | 8 |
| 1 in 20 | 0.67³ | 7 | |
| Incident probability | Incident rates | C pk | Row |
| Moderate: Generally associated with previous similar processes that have experienced occasional incidents, but not in significant proportions | 1 in 80 | 0.83 ³ | 6 |
| 1 in 400 | 1.00³ | 5 | |
| 1 of 2000 | 1.17³ | 4 | |
| Low: Isolated incidents were associated with similar processes | 1 in 15,000 | 1.33³ | 3 |
| Very low: only isolated incidents are associated with almost identical processes | 1 in 150,000 | 1.50³ | two |
| Remote control: the incident is implausible | 1 in £ 1,500,000 | 1.67³ | one |
Current controls
Current controls are descriptions of measures that prevent failure mode from occurring or detect failure mode if they do occur.
The design and process controls are grouped according to their purpose:
• Type 1: These controls prevent the cause or mode of failure from occurring, or reduce their occurrence
• Type 2: These controls detect the cause of the failure mode and lead to corrective action
• Type 3: These controls detect the mode failure before the product reaches the customer
Detection
Detection is an evaluation of the probability that the proposed process controls (listed in the previous column) will detect the failure mode, before the part or component leaves the manufacturing or assembly location.
Random quality control checks are unlikely to detect the existence of an isolated defect and therefore will not result in a noticeable change in the degree of detection. A valid detection control is sampling done on a statistical basis.
Table 5.- Evaluation criteria and suggested grading system for the detection of a cause of the incident or failure mode in an FMEA design.
| Detection | Criteria: probability of design control detection | Row |
| Absolute uncertainty | Design control does not detect a potential cause of the incident or subsequent failure mode, there is no design control | 10 |
| Very far | The far-away probability that design control will detect a potential cause of the incident or subsequent failure mode | 9 |
| Remote | The remote probability that design control detected a potential cause of the incident or subsequent failure mode | 8 |
| Very low | Very low probability that design control detected a potential cause of the incident or subsequent failure mode | 7 |
| Low | Low probability of design control detecting a potential cause of the incident or subsequent failure mode | 6 |
| Moderate | The moderate probability that the design control detected a potential cause of the incident or subsequent failure mode | 5 |
| Moderately high | Moderately high probability that design control detected a potential cause of the incident or subsequent failure mode | 4 |
| Tall | The high probability that design control detected a potential cause of the incident or subsequent failure mode | 3 |
| Very high | Very high probability of design control detecting a potential cause of the incident or subsequent failure mode | two |
| Almost sure | Design control will almost certainly detect a potential cause of the incident or subsequent failure mode | one |
Table 6.- Evaluation criteria and suggested graduation system for detecting a cause of the incident or failure mode in an FMEA process
| Detection | Criteria: probability of design control detection | Row |
| Almost impossible | None of the available controls detect incident mode or cause | 10 |
| Very far | Current controls have a very remote probability of detecting failure mode or cause | 9 |
| Remote | Current controls have a remote probability of detecting failure mode or cause | 8 |
| Very low | Current controls have a very low probability of detecting failure mode or cause | 7 |
| Low | Current controls have a low probability of detecting failure mode or cause | 6 |
| Moderate | Current controls have a moderate probability of detecting failure mode or cause | 5 |
| Moderately high | Current controls have a moderately high probability of detecting failure mode or cause | 4 |
| Tall | Current controls have a high probability of detecting failure mode or cause | 3 |
| Very high | Current controls have a very high probability of detecting failure mode or cause | two |
| Almost sure | Current controls almost certainly detect the mode or cause of failure. Reliable detection controls are known through similar processes | one |
NPR
• The risk priority number (NPR) is the mathematical product of the severity, the occurrence and the detection, that is:
NPR = S * O * D
• This value is used to identify the most serious risks to look for corrective actions.
Recommended actions).
When failure modes have been mandated by the NPR, corrective actions should first address the highest-grade issues and points and critical items. The intention of any recommended action is to reduce the degrees of occurrence, severity and / or detection. If no action is recommended for a specific cause, it should be so indicated.
A process FMEA will be of limited value if it does not have effective and corrective actions. It is the responsibility of all affected activities to implement effective monitoring programs to meet all recommendations.
Responsible area / individual and completion date (of recommended action)
The area and person responsible for the recommended action are recorded, as well as the target completion date.
Taken actions.
After an action has been completed, record a brief description of the current action and effective or completion date.
Resulting Npr, after having identified the corrective action, estimates the degrees of occurrence, severity and final detection. The resulting NPR is calculated, this is the product of the severity, occurrence and detection values.
The process engineer is responsible for ensuring that all recommended actions are properly implemented and monitored. The AMEF is a living document and should always reflect the latest level of design.
Sequence of Procedures for the elaboration of the AMEF
Once the elements of the FMEA have been identified, it is necessary to know how it should be carried out, that is, the logical order that the operations must carry out; This sequence is best expressed through the flow chart presented below:
It should be noted that the team responsible for executing the FMEA must have been previously defined, as well as a prior analysis for data collection.
The role of the FMEA in quality systems
• Prevention and problem solving can be considered as the main objectives of any quality system.
• For the prevention of problems, the quality systems use the display of the Quality Function (QFD), the Failure Tree Analysis (FTA), the Reverse Failure Tree Analysis (RFTA), the Advanced Product Quality Planning (APQP) and the AMEF, the latter is used both directly and indirectly through the APQP and the Design of Experiments (DOE), which is an important element for the prevention and solution of problems; As for the latter, the quality systems mainly use Continuous Improvement, the Quality Operating System (QOS), the eight disciplines for problem solving (8D) and the Control Plan, whose preparation requires directly from the AMEF,Statistical Process Control (SPC) tools and the consideration of the special characteristics established through the AMEF.
Amef's Relationship With Iso 9000 Standards
• ISO 9000 standards only define guidelines and models, they do not indicate procedures to be implemented or the corresponding strategies that must be defined by each company.
• The ISO 9000 series is especially applicable when it is necessary to check the client, as a contractual requirement, that they are being considered a set of previously established quality parameters. In these cases, the client contractually demands the quality check, not only of the development project.
Among the requirements established in the 9000: 2000 standard, reference is made to design control and process control. Clauses establish a requirement for their verification, including an analysis of failures and their corresponding effects. This verification must confirm that the data resulting from the project meets the established requirements, through project control activities, such as carrying out and recording the critical analysis of the project. The FMEA can be considered particularly as one of the most useful and efficient methods for this purpose.
conclusion
By completing this report, we establish the great importance and scope of benefits provided by the Potential Failure Mode and Effects Analysis as a tool to examine all the ways in which a product or process may fail; In addition, a review of the action to be taken to minimize the probability of failure or its effect is made.
Since for most products and processes it is not economical to carry out the FMEA for each component, it is necessary to carry out the critical elements that must be subjected to it.
Although FMEA is very valuable as an early warning technique, the definitive proof is given by the use of the product by the client. However, the field experience comes too late, and it is here that he highlights the importance of having it preceded by the AMEF so that companies can simulate the use of their products and processes in the field of work.
