IntroductiontoFMEA

1、SECTION ONEIntroductionToFAILURE MODEEFFECTS & ANALYSISFMEA1.0Definition, Purpose, and PracticesØ FMEA: A SYSTEMIZED GROUP OF ACTIVITIESDESIGNED TO1.RECOGNIZE and EVALUATE the POTENTIAL FAILURE of a PRODUCT/PROCESS and its EFFECTS2.IDENTIFY ACTIONS WHICH COULD ELIMINATE or REDUCE the CHANCE

2、 of POTENTIAL FAILURE OCCURRING3.DOCUMENT the PROCESS1.1DESIGN FMEA (DFMEA)PROCESS FMEA (PFMEA) ARE ANALYTICAL TECHNIQUESUTILIZED BYDESIGN RESPONSIBLE ENGINEER/TEAMOR MANUFACTURING RESPONSIBLE ENGINEER/TEAMTHESE TOOLS ARE USED TO ASSURE THAT POTENTIAL PRODUCT FAILURE MODESAndTHEIR ASSOCIATED CAUSESH

3、AVE BEEN CONSIDERED and ADDRESSEDIN THE DESIGN OR MANAUFACTURING PROCESS1.2DFMEASUPPORTS DESIGN PROCESSin REDUCING RISK of FAILURESbyØ AIDING IN THE OBJECTIVE EVALUATION OF DESIGN REQUIREMENTS AND DESIGN ALTERNATIVESØ AIDING IN THE INITIAL DESIGN FOR MANUFACTURING AND ASSEMBLYØ INCREA

4、SING THE PROBABILITY THAT POTENTIAL FAILURE MODES AND THEIR EFFECTS ON SYSTEMS AND PRODUCT OPERATION HAVE BEEN CONSIDERED IN THE DESIGN & DEVELOPMENT PROCESSØ PROVIDING ADDITIONAL INFORMATION TO AID IN THE PLANNING OF EFFICIENT DESIGN TESTING AND PRODUCT DEVELOPMENT PROGRAMS1.3DFMEAALSO SUP

5、PORTS DESIGN PROCESSByØ DEVELOPING A LIST OF POTENTIAL FAILURE MODES RANKED ACCORDING TO THEIR EFFECT ON THE “CUSTOMER,” THUS ESTABLISHING A PRIORITY SYSTEM FOR DESIGN IMPROVEMENTS AND DEVELOPMENT TESTINGØ PROVIDING AN OPEN ISSUE FORMAT FOR RECOMMENDING AND TRACKING RISK REDUCING ACTIONS&#

6、216; PROVIDING FUTURE REFERENCE TO AID IN ANALYZING FIELD CONCERNS, EVALUATING DESIGN CHANGES AND DEVELOPING ADVANCED DESIGNS1.4PFMEATHE PROCESS FMEAØ IDENTIFIES POTENTIAL PRODUCT RELATED PROCESS FAILURE MODESØ ASSESSES THE POTENTIAL CUSTOMER EFFECTS OF THE FAILURESØ IDENTIFIES THE PO

7、TENTIAL MANUFACTURING OR ASSEMBLY PROCESS CAUSES AND IDENTIFIES PROCESS VARIABLES ON WHICH TO FOCUS CONTROLS FOR OCCURRENCE REDUCTION OR DETECTION OF THE FAILURE CONDITIONSØ DEVELOPS A RANKED LIST OF POTENTIAL FAILURE MODES, THUS ESTABLISHING A PRIORITY SYSTEM FOR CORRECTIVE ACTION CONSIDERATIO

8、NSØ DOCUMENTS THE RESULTS OF THE MANUFACTURING OR ASSEMBLY PROCESS1.5FMEA APPLICATIONSINMANUFACTURING SETTINGSDFMEAØ SHOULD BE INITIATED BEFORE OR AT DESIGN CONCEPT FINALIZATIONØ SHOULD BE CONTINUALLY UPDATED AS CHANGES OCCUR OR ADDITIONAL INFORMATION IS OBTAINED THROUGHOUT THE PHASES

9、 OF PRODUCT DEVELOPMENTØ SHOULD BE FUNDAMENTALLY COMPLETED BEFORE THE DRAWINGS ARE RELEASED FOR TOOLING, OR OTHER MANUFACTURING NEEDSØ ADDRESSES THE DESIGN INTENT AND ASSUMES THE DESIGN WILL BE MANUFACTURED AND ASSEMBLED TO THIS INTENTØ DOES NOT RELY ON PROCESS CONTROLS TO OVERCOME PO

10、TENTIAL WEAKNESSES IN THE DESIGN, BUT IT DOES TAKE THE TECHNICAL AND PHYSICAL LIMITATIONS OF A MANUFACTURING OR ASSEMBLY PROCESS INTO CONSIDERATION1.6FMEA APPLICATIONSINMANUFACTURING SETTINGSPFMEAØ SHOULD TAKE INTO ACCOUNT ALL MANUFACTURING OPERATIONS, FROM INDIVIDUAL COMPONENTS TO ASSEMBLIES&#

11、216; DOES NOT RELY ON PRODUCT DESIGN CHANGES TO OVERCOME WEAKNESSES IN THE PROCESSØ DOES TAKE INTO CONSIDERATION A PRODUCTS DESIGN CHARACTERISTICS RELATIVE TO THE PLANNED MANUFACTURING OR ASSEMBLY PROCESSØ ASSURES THAT, TO THE EXTENT POSSIBLE, THE RESULTING PRODUCT MEETS CUSTOMER NEEDS AND

12、 EXPECTATIONS1.7KEY RESOURCES NECESSARYTO CONDUCTSUCCESSFUL FMEA PROGRAMSØ COMMITMENT OF TOP MANAGEMENTØ KNOWLEDGEABLE INDIVIDUALS, I.E. EXPERTISE IN:DESIGNMANUFACTURINGASSEMBLYSERVICEQUALITYRELIABILITYØ INDIVIDUALS ATTENTIVE TO FMEA TIMELINESS, I.E. ACHIEVE GREATEST VALUE: BEFORE A D

13、ESIGN OR PROCESS FAILURE MODE HAS BEEN UNKNOWINGLY DESIGNED INTO THE PRODUCTØ PEOPLE RESOURCES MAY BE INTERNAL OR EXTERNAL TO THE BUSINESS, OR A COMBINATION THEREOF1.8FMEAADVANTAGES(AND LIMITATIONS)ADVANTAGES:Ø ENHANCE DESIGN AND MANUFACTURING EFFICIENCIESØ ALLEVIATE LATE CHANGE CRISE

14、SØ MINIMIZE EXPOSURE TO PRODUCT FAILURESØ AUGMENT BUSINESS RECORDSØ IMPROVE “BOTTOM LINE” RESULTSØ ADD TO CUSTOMER SATISFACTION1.9FMEAADVANTAGES(AND LIMITATIONS)LIMITATIONS*Ø EMPLOYEE TRAINING REQUIREMENTSØ INITIAL IMPACT ON PRODUCT AND MANUFACTURING SCHEDULESØ FIN

15、ANCIAL IMPACT REQUIRED TO UPGRADE DESIGN, MANUFACTURING, AND PROCESS EQUIPMENT AND TOOLS*These LIMITATIONS should be recognized and treated as “short term” and “minimal” interruptions to a business, and understood to ultimately offer dynamic ADVANTAGES!1.10SECTION TWOFMEA PLANUTILIZING KEYPRODUCTION

16、 PROCESSESANDPRODUCT EVALUATION TOOLS2.0QUALITY MANAGEMENT PROCESSPRINCIPLES DRIVING CUSTOMER SATISFACTION· TEAMWORK· MANAGING PROCESSES· STATISTICAL PROCESS CONTROL2.1FMEA REQUIRESTEAMWORKBUILD QUALITY INTO PEOPLE THROUGHTRAININGandCOMMITTED LEADERSHIP2.2MANAGING PROCESSESUTILIZE A S

17、TYLE OF MANAGEMENT THAT IS ALSO PEOPLE ORIENTED IN CONTRAST TO ONE THAT IS SOLELY ORIENTED TOWARD RESULTSPROCESS-ORIENTED MANAGEMENTØ SUPPORT AND STIMULATE EFFORTS TO IMPROVE THE WAY EMPLOYEES DO JOBSØ REINFORCES “LONG TERM” OUTLOOKMANAGEMENT CRITERIA· DISCIPLINE· TIME MANAGEMENT

18、· SKILL DEVELOPMENT· PARTICIPATION & INVOLVEMENT· MORALE· COMMUNICATION2.3STATISTICAL PROCESS CONTROL(SPC)STATISTICAL TOOLS· PARETO Diagrams· CAUSE-AND-EFFECT Diagrams· HISTOGRAMS· CONTROL Charts· SCATTER Diagrams· GRAPHS· CHECKSHEETS2.4PROC

19、ESS INCORPORATIONA PROCESS INCLUDES SOME COMBINATION OF:· METHODS· MATERIALS· MACHINES· MANPOWER· ENVIRONMENT· MEASUREMENTTHESE ARE INCORPORATED TO COMPLETE TASKS, SUCH AS PRODUCING A PRODUCT OR PERFORMING A SERVICE. A PROCESS HAS MEASURABLE INPUTS AND OUTPUTS.2.5FMEA P

20、ROJECTSSELECTION & PURPOSETANGIBLE EFFECTSØ SUCCESSFUL DEVELOPMENT OF NEW PRODUCTSØ SHORTENING OF PRODUCT DEVELOPMENT TIMEØ INCREASED MARKET SHAREØ INCREASED SALES VOLUMEØ DEVELOPMENT OF NEW MARKETSØ INCREASED PRODUCTION VOLUMEØ FEWER PROCESSESØ IMPROVED Q

21、UALITYØ REDUCED DEFECT COSTSØ FEWER CUSTOMER COMPLAINTS2.6FMEA PROJECTSSELECTION & PURPOSEINTANGIBLE EFFECTSØ INCREASED QUALITY-CONSCIOUSNESS AND PROBLEM-CONSCIOUSNESSØ MORE CONFIDENCE IN NEW PRODUCT DEVELOPMENTØ IMPROVED STANDARDIZATIONØ IMPROVED QUALITY OF WORK

22、16; IMPROVED INFORMATION FEEDBACK2.7FMEA PROJECT IMPLEMENTATIONQUANTITATIVE METHODS· SEVERITY RANKING· OCCURRENCE RANKING· DETECTION RANKINGRISK PRIORITY NUMBER(RPN)2.8FMEA PROJECT IMPLEMENTATION QUALITATIVE METHODS· SIMILAR PAST EXPERIENCES· BRAINSTORMING ANALYSIS· CUS

23、TOMER(S) INPUTFINANCIAL IMPACT2.9FAILURE MODE ERRORSDESCRIPTIONSDESIGNPOTENTIAL FAILURE MODE CAN BE DESCRIBED AS THE MANNER IN WHICH A COMPONENT, SYSTEM, OR SUBSYSTEM COULD FAIL TO MEET THE DESIGN INTENT. THESE FAILURE MODES SHOULD BE DESCRIBED IN “PHYSICAL” OR TECHNICAL TERMS.2.10FAILURE MODE ERROR

24、SDESCRIPTIONSPROCESSPOTENTIAL FAILURE MODE IS DEFINED AS THE MANNER IN WHICH THE PROCESS COULD FAIL TO MEET THE REQUIREMENTS OF THE DESIGN INTENT. THE FMEA TEAM SHOULD BE ABLE TO POSE AND ANSWER THE FOLLOWING QUESTIONS:Ø HOW CAN THE PROCESS/PART FAIL TO MEET SPECIFICATIONS?Ø REGARDLESS OF

25、ENGINEERING SPECIFICATIONS, WHAT WOULD A CUSTOMER (END USER, SUBSEQUENT OPERATIONS, OR SERVICE) CONSIDER OBJECTIONABLE?2.11FAILURE MODE ERRORSEXAMPLESGROUNDEDFRACTUREDSHORT CIRCUITEDOXIDIZEDOPEN CIRCUITEDSTICKINGCRACKEDBURREDHANDLING DAMAGELEAKINGDEFORMED LOOSENEDBENT TOOL WORNIMPROPER SET-UPDIRTYTH

26、ESE ARE SOME “TYPICAL” FAILURE MODES. IT IS RECOMMENDED THAT A COMPARISON OF SIMILAR PROCESSES AND A REVIEW OF CUSTOMER (SUBSEQUENT OPERATION AND END USER) CONSIDERATIONS BE CONDUCTED.2.12FAILURE MODE ERRORSPROBABILISTIC ERRORSDEFINITION:A PROBABILISTIC ERROR IS ONE THAT CANNOT BE PREDICTED WITH CER

27、TAINTY.SAMPLE PROBLEMTHE VOLTAGES OF 50 BATTERIES OF THE SAME TYPE HAVE A MEAN OF 1.52 VOLTS AND A STANDARD DEVIATION OF 0.04 VOLT. FIND THE PROBABLE ERROR OF THE MEAN AND THE 50% CONFIDENCE LIMITS.SOLUTION:PROBABLE ERROR OF THE MEAN = 0.6745Page: 25s x= 0.6745s ¸ÖN = 0.6745s ¸ÖN

28、-1 = 0.6745(0.04 ¸Ö49)= 0.00385 VOLTSTHEN, 50% CONFIDENCE LIMITS ARE1.52 ± 0.00385 VOLTS2.13FAILURE MODETECHNICAL ANALYSISFAILURE MODE ANALYSIS AND CONTROL MUST BE ASSIGNED TO AN INDIVIDUAL WHO POSSESSES THE BACKGROUND EXPERIENCE AND NECESSARY SKILLS TO SUCCESSFULLY MANAGE THE FMEA. T

29、HIS INDIVIDUAL MAY BE AN EMPLOYEE OF THE COMPANY OR A SPECIALIST THAT IS ENGAGED BY THE COMPANY.2.14PROCESS REVIEW IN PRACTICEFAILURE MODE ERRORS INTRODUCE A RISK INTO A PROCESS (OR PROCESSES). A CONTROL SYSTEM PROVIDES A MEANS OF “CATCHING” AN ERROR BEFORE ANY DAMAGE IS INCURRED.EXAMPLE: TYPE OF ER

30、RORDESCRIPTIONInformation Validation There are no checks to Omissioncatch incorrect or incomplete information items.An administrative example could be the processing of purchase orders that could occasionally result in incorrect orders being sent to the supplier(s). This could be the case if there a

31、re no checks on orders before they are submitted. Such an error could result in an incorrect shipment of goods, and payment for undesired items. In this example, the preferable control system would be one that would more readily assure correct preparation of purchase orders. Training of staff and si

32、mplifying process steps are areas to support this goal.2.15FLOW CHARTPROCESS DIAGRAMMINGFLOW CHART FOR PURCHASING PROCEDURE2.16FMEA PLANONGOING EVALUATORY PROCEDURESTHE INDIVIDUAL RESPONSIBLE FOR DESIGN AND/OR PROCESS IS ACCOUNTABLE FOR ASSURING THATALL DFMEA AND/OR PFMEA ACTIONS RECOMMENDED HAVE BE

33、EN IMPLEMENTED OR ADEQUATELY ADDRESSED. THE FMEA IS A LIVING DOCUMENT AND SHOULD ALWAYS REFLECT THE LATEST DESIGN LEVEL, AS WELL AS THE LATEST RELEVANT ACTIONS, INCLUDING THOSE OCCURRING AFTER THE START OF PRODUCTION OPERATIONS.2.17SECTION THREEImplementingFMEA3.0IDENTIFYING POTENTIAL FAILURESPRODUC

34、T CONSIDERATIONS&PRODUCTION OPERATIONSFAILURE MODES DEFINITIONALL WAYS IN WHICH A PRODUCT (INCLUDING EACH OF ITS SPECIFIC PARTS) OR PROCESS CAN FAIL TO PERFORM ITS INTENDED FUNCTION. MORE THAN ONE FAILURE MODE CAN EXIST FOR A GIVEN PART OR PROCESS. ALSO, SOME FAILURES MAY BE GRADUAL AND/OR PARTI

35、AL, WHEREAS OTHERS MAY OCCUR IMMEDIATELY AND COMPLETELY.3.1IDENTIFYING POTENTIAL FAILURESUSE of STATISTICAL TOOLSTYPES/DEFINITIONS1.PARETO Diagrams. Classify failures according to cause. The failures are diagrammed according to priority, using a bar-graph format, with 100% indicating the total amoun

36、t of failures.2. Cause-and-Effect Diagrams. Also, known as “fishbone diagrams.” Used to analyze the characteristics of a product or process and the factors that contribute to them, and to prospective failures.3.Histograms. Frequency data obtained from measurements display a peak around a certain val

37、ue. This can be used to determine potential failures by checking dispersion shape, center value, and the nature of dispersement.3.2IDENTIFYING POTENTIAL FAILURESUSE of STATISTICAL TOOLSTYPES/DEFINITIONS4.Control Charts. Serve to detect abnormal variations. Sample data (defining potential failures) a

38、re plotted to evaluate situations and trends in a process. Comparisons can be made against required control limits.5.Scatter Diagrams. Corresponding data are plotted and the relation between the data (and impact on prospective failures) can be analyzed.BESIDES THESE ANALYTICAL (STATISTICAL) TOOLS, T

39、HERE CAN OFTEN BE MANAGEMENT SITUATIONS WHERE THE DATA NEEDED FOR PROBLEM SOLVING ARE NOT AVAILABLE. NEW PRODUCT DEVELOPMENT IS AN EXAMPLE. ANOTHER MAY BE DEVELOPING A NEW MANUFACTURING METHOD TO IMPROVE PRODUCTIVITY. THESE SITUATIONS REQUIRE COLLABORATION AMONG PEOPLE FROM DIFFERENT DEPARTMENTS.3.3

40、IDENTIFYING POTENTIAL FAILURESPROSPECTIVE QUESTIONSFOLLOWING ARE REPRESENTATIVE QUESTIONS TO CONSIDER IN SOME BASIC AREAS THAT MAY REQUIRE FMEA. THIS IS NOT ALL INCLUSIVE, BUT EXAMPLES TO EVALUATE.A. MATERIAL1. Are there any mistakes in volume?2. Are there any mistakes in grade?3. Are there impuriti

41、es mixed in?4. Is the inventory level adequate?5. Is there any waste in material?6. Is the handling adequate?7. Is the quality standard adequate?B. OPERATION METHOD1. Are the work standards adequate?2. Is it a safe method?3. Is it a method that ensures a good product?4. Is it an efficient method?5.

42、Are the temperature and humidity adequate?6. Is the lighting adequate?7. Is there adequate contact with the previous and next processes?3.4IDENTIFYING POTENTIAL FAILURESPROSPECTIVE QUESTIONSC. MACHINE1. Does it meet production requirements?2. Does it meet process capabilities?3. Does it meet precisi

43、on requirements?4. Is the layout adequate?5. Are there enough machines/facilities?6. Is everything in good working order?7. Is operation stopped often because of mechanical trouble?D. MANPOWER (OPERATOR)1. Does s/he follow standards?2. Is his/her work efficiency acceptable?3. Is he/she responsible (

44、accountable)?4. Is he/she qualified?5. Is he/she experienced?6. Is he/she assigned to the right job?7. Does he/she maintain good human relations?3.5 EFFECT AND ANALYSIS METHODOLOGIESFailure ModeEffect Severity: 1-10Causes Occurrence: 1-10Controls Detection: 1-103.6RPN = Severity XOccurrence X Detect

45、ion1-1,000EFFECT AND ANALYSIS METHODOLOGIESDEFINING FMEA TERMSRISK PRIORITY NUMBER (RPN)The RISK PRIORITY NUMBER (RPN) is the product of the SEVERITY (S), OCCURRENCE (O), and DETECTION (D) ranking:RPN = (S) x (O) x (D)The RPN is a measure of design risk. The RPN is also used to rank order the concer

46、ns in processes (e.g., in Pareto fashion).The RPN will be between “1” and “1,000.” For higher RPNs the team must undertake efforts to reduce this calculated risk through corrective action(s).3.7EFFECT AND ANALYSIS METHODOLOGIESDEFINING FMEA TERMSSEVERITY (S)SEVERITY (S) is an assessment of the serio

47、usness of the effect of the potential failure mode. SEVERITY applies to the effect only. A reduction in SEVERITY ranking index can be effected only through a design change. SEVERITY is estimated on a “1” to “10” scale.3.8EFFECT AND ANALYSIS METHODOLOGIESSEVERITY (S) Evaluation CriteriaEffectCriteria

48、: SEVERITY of EffectRankingHazardous-withoutwarningVery high severity ranking when a potential failure mode affects safe operation and/or involves noncompliance with regulations without warning.10Hazardous-with warningVery high severity ranking when a potential failure mode affects safe operation an

49、d/or involves noncompliance with regulations with warning.9Very highProduct/item inoperable, with loss of primary function.8HighProduct/item operable, but at reduced level of performance. Customer dissatisfied.7ModerateProduct/item operable, but may cause rework/repair and/or damage to equipment.6Lo

50、wProduct/item operable, but may cause slight inconvenience to related operations.5Very LowProduct/item operable, but possesses some defects (aesthetic and otherwise) noticeable to most customers.4MinorProduct/item operable, but may possess some defects noticeable by discriminating customers.3Very Mi

51、norProduct/item operable, but is in noncompliance with company policy.2NoneNo effect.1NOTE:Make sure team members agree on the criteria for ranking severity and apply them consistently.3.9EFFECT AND ANALYSIS METHODOLOGIESDEFINING FMEA TERMSOCCURRENCE (O)OCCURRENCE (O) is the likelihood that a specif

52、ic cause/mechanism will occur. OCCURRENCE is estimated on a “1” to “10” scale.3.10EFFECT AND ANALYSIS METHODOLOGIESOCCURRENCE (O) Evaluation CriteriaProbability of FailurePossible Failure RatesRankingVery High: Failure is almost inevitable³1 in 2101 in 39High: Repeated Failures1 in 881 in 207Mo

53、derate: Occasional Failures1 in 8061 in 40051 in 2,0004Low: Relatively Few Failures1 in 15,00031 in 150,0002Remote: Failure is Unlikely£1 in 1,500,0001NOTE:The team should agree on an evaluation criteria and ranking system, which is consistent, even if modified for individual product/process analysis.3.11EFFECT AND ANALYSIS METHODOLOGIESDEFINING FMEA TERMSDETECTION (D)DETECTION (D) is an assessment of the abilityof proposed *type (2) current design controls to detect a potential cause/mechanism (design weakness), or the ability of the proposed #type (3) cur