《FMEA在生产管理质量中的作用(英文版).pptx》由会员分享，可在线阅读，更多相关《FMEA在生产管理质量中的作用(英文版).pptx（43页珍藏版）》请在taowenge.com淘文阁网|工程机械CAD图纸|机械工程制图|CAD装配图下载|SolidWorks_CaTia_CAD_UG_PROE_设计图分享下载上搜索。
1、1Define the scope for the FMEA2Identify team members, customer, and knowledge experts3Develop roles and responsibilities for team members4Study the product / process / service5Develop rating scales for Severity, Occurrence, and Detection6Brainstorm failure modes, known and potential7List the effects
2、, causes, and current controls for each failure mode8Assign ratings for Severity, Occurrence, and Detection9Calculate RPN10 Decide on a threshold RPN11 Prioritize based on RPN and severity, or as required by the customer12 Identify corrective actions13 Implement and verify effectiveness of correctiv
3、e action14 Complete the documentationDefine the boundaries within which the product, process, or service will be studied, identifying the operating assumptions, current conditions, and constraints if any. Example 1: Conduct a Process FMEA on PCB manufacturing line 1, starting withoperation 10 and en
4、ding with operation 250. The assumption is that design of parts is correct, and no shortagesThe line is setup as shown in process flow drawing # 809901Example 2: Conduct Design FMEA on the pressure relief valve (Part #JJJKKK)for potential failures under extreme operating conditionsAssume no preventi
5、ve maintenance will be conducted once installedDocument on the FMEA startup formlPoints to consider:Who are the customers for the FMEA, and what are their requirements / expectationsExample: Conduct (Process) FMEA on the assembly line for Product A, covering operations 110 through 230 When will this
6、 FMEA be considered complete under the given conditions (Remember, FMEAs are considered living documents, as such they need to be updated / revised as things change.)Example: FMEA will be considered complete when the team has implemented all corrective actions as agreed upon based on the RPN thresho
7、ld.SystemSub systemComponentProcessServiceLevel FocusFailure CauseEffectModeSystem CarOverheating Cooling syst. Stranded on failurethe roadSub System Cooling SystemInadequate Thermostat Car Overheatingcoolant flow stuck closedSub Assembly ThermostatStuck closed Bi Metal SpringInadequate brokencoolan
8、t flowComponentBi Metal SpringBroken FatigueThermostat stuck closed Safety, LubricationEngine, Braking, Exhaust Trans., Cooling Thermostat Diaphragm, Bi Metal SpringRadiatorPump V BeltCompanyPlant 1Plant 2Plant 3Plant 4PCB LinePower SuppliesFHP MotorsOp 10Op 20. Op 250Level FocusFailure CauseEffectM
9、odeSystem Factory (Plant 2)Late deliveries Low FPY Lost to customer on Line 2businessSub System Line 2 (Power Supplies) Low FPY High ReworkLate deliveries at / after testingOperation TestingHigh rework Testing resultsLow FPY Op 210 inconsistent Op 120Op A20Op A10Op 10Op 210Op 240Op 250Get the team t
10、ogether, identify the customer and involve them in the process. Some preliminary work must be done to understand the product / process / service, so the right team can be assembled. Example:Team members include John Patton, Kirit Chawla (Customer rep) Kim Heaston, and Betty Chung (Team Lead) No one
11、knows enough about the process at this time.Document on the FMEA startup formWho does what in the team. Clearly define the roles and responsibilitiesfor each of the team members. At this time, it is important to consider thestrengths of each individual and leverage the same. Example:Team Lead: Coord
12、inate the activities of the team, and keep on trackAct as cheer leader, and interface with managementEnsure the decisions are made with consensusCustomer:Define requirementsParticipate in developing rating scales, especially for Severity Knowledge Expert:As consultant provide info on details of prod
13、uct / process or serviceThe objective of this step is to acquire as much knowledge as possible. It is highly recommended that all team members delve into this, sincethere is no other way to familiarize yourself with the product / process / service.Example 1: The team responsible for conducting Proce
14、ss FMEA on gas panels production line spends 2 weeks on the line building gas panels, and testing the same. Example 2: The FMEA team responsible for Design FMEA on a medical implantstudies the way the product is going to be handled in the operating room, and used by the surgeon.Example 3: Four out o
15、f five members of the FMEA team spend 3 days with the phone operators, and order processing group to get first hand understanding of all the steps involved in booking hotel rooms.Once you have gained enough knowledge about the product / process / serviceand about the way it may be used by the custom
16、er / end user, you are in a position to develop meaningful rating scales. Most rating scales are on a scale from 1 to 10, however some may use a 1 to 5scale. Which of the two, is less important than ensuring that the same scaleis used throughout the company. You may have to develop / adapt the scale
17、s to suit your needsFor examples, see next page Causes Failure Modes* Effects Occurrence Detection SeverityHow frequent?Who has the most info?If eliminated, will failure be avoided?How easy to detect with current controls?Who has the most info?If prevented will effects be eliminated?How severe?Who h
18、as the most info?* Causes may be used in place of Failure Modes in some cases Severity Occurrence Detection Effect Cause Failure Mode or Cause1 Remote chance, 1 ppm2 Very low 1 to 53 Very low 5 to 104 Moderate 10 to 1005 Moderate 100 to 1K6 High 1K to 10K7 High 10K to 100K 8 Very High 100K to 1M9 Ve
19、ry High 1M to 10M10 Almost certain .1 Certain to be detected2 Very high chance of detection3 High chance of detection4 Moderately high chance of .5 Moderate chance of detection6 Low chance of detection7 Very low chance of .8 Remote chance of .9 Very remote chance of .10 Slips by undetected1 No effec
20、t 2 Minor effect- fit / cosmetics3 Major effect-fit / cosmetics4 Minor reduction in Function5 Customer dissatisfied6 Major reduction in function7 Totally dysfunctional8 Presents minor hazard9 Presents injury hazard10 Fatal if usedSeverity - effect1 No effect on operators, users, or current and down
21、stream processes2 Minor re work at current step / no dysfunction introduced in product3 Rework at current / subsequent steps causing some delayed schedules, or product is rendered partially dysfunctional4 Heavy rework at subsequent operation, major delays in schedules, or product function is severel
22、y impacted5 Risk of injury or fatal Occurrence - cause1 Very rare occurrence, X00 ppb or once in 3 to 5 years2 Rare occurrence, X to X000 ppm, 2 to 5 occurrences in 3 years3 Frequent occurrence, X000 to X0,000 ppm, 1 to 3 occurrences in 1 year4 High frequency of occurrence, X0,000 to XM ppm, over 4
23、occurrences per year5 Almost certainDetection - failure mode (or cause in some cases) 1 Certain to be detected, before any measurable damage is done2 Automated inspection / detection or strong process controls exist at current step or within the system boundaries3 Sampling plans used for inspection
24、purposes4 100% manual inspection methods used within system boundaries 5 Defect unlikely to be detected, will slip by unnoticed Severity Occurrence Detection Effect Cause Failure Mode or CauseFailure modes should be described in “physical” or technical terms, not as a symptom noticeable by the custo
25、mer.The idea is to identify as many failure modes as possible. This will involve gettingall team members and stake holders involved, however, the properapproach to brainstorming must be used. Defer all judgment, maybe evento a subsequent meeting. The biggest risk in this step is that participants wi
26、ll start mixing up failure modeswith causes, and effects. To prevent this, it is highly recommended that the meeting leader do a dry run to familiarize the group with the differences between failure modes, effects, and causes. A listing of the product / process / services specifications, or outputs
27、and requirements will benefit here.Post it pads are a convenient way of collecting this information, and grouping it into logical groups. If you write these on a work sheet, make sure you leave enough space for effects and causes Example:OverheatingSlow accelerationSkids Catches fireExcessive stoppi
28、ng distancePoor gas mileageWobbles 50 mphPoor corneringOil leakRattling of rear door Latch does not holdWipers ineffectiveHorn doesnt workStiff steeringand so on.Question:Are all the above failure modes for the car? Causes Failure Modes EffectsCar and its performancerequirementsUsers / Customers Com
29、ponent parts, & sub-systems Lead to Lead toWhen you sense getting into such a loop, askWhat is the FMEA being conducted on (focus)? What are the performance requirements / specifications of the (focus)?Example:Excessive backlashNoise exceeds specificationsDefects per unit 25% in excess of standardTe
30、st fixture operation unsafe Oil leaks Question:Are all the above failure modes for the process?Op 120Op A20Op A10Op 10Op 210Op 240Op 250Machine Weld brackets Integration Testing Crate & Ship Assemble AxleMount WheelsSystem Boundaries Causes Failure Modes EffectsOperation 240Users / Customers / Opera
31、torsDown stream operationsPlant deliverables Work instructionsOperator errorsPrior operations Lead to Lead toEffects of Failure are defined as the effects of the failure mode on the function, as perceived by the customer.When you sense getting into such a loop, askWhat part of the is the FMEA being
32、conducted on (focus)? What are the performance requirements / specifications of the (focus)?Failure Mode 2Cause 1Cause 2Cause 3Cause 4Effect 1Effect 2Effect 3Failure Mode 1Failure Mode 3 Cause of Failure is defined as an indication of a design weakness, the consequence of which is the failure mode.F
33、ailure Modes: Will share causes, and effectsCauses: The reasons for failure, can be one or many for a given failure modeEffects:The results of the failure, can be one or many for a given failure modeOver heatingBrakesbindingCause 2Low tirepressureCause 4Poor GasMileageStalled on roadEffect 3Poor acc
34、elerationFailure Mode 3Cause Failure Mode EffectImproper SeatingMissingInstructionsCause 2LooseNutsCause 4Failstorque testUnsafe atTestingEffect 3Assembled backwardsFailure Mode 3Cause Failure Mode EffectTo capture the effects, causes, and existing controls if any for the failure mode in question. I
35、nput from customer, knowledge experts will be especially helpful at this step.Caution:Many teams commit errors of wishful thinking here, especially with regard to current controls. You should focus on the current state, and not on what could be done, or what ought to be done. Now start documenting o
36、n the worksheet, knowing that each failure mode can have multiple effects, and multiple causes. ProductFailure EffectSCauseOCurrentDRPNProcessModeControlsor ServiceCarOverheatsStranded on CoolingTemp GageRoadSystemBindingNonebrakesOverloadedLoad capacitylisted Here we accomplish the conversion of ea
37、ch of these factors into a numerical value. Depending on the quality of the rating scales this step can be very easy or somewhat frustrating. There will be times when the team will disagree on the values assigned to a particular cause, effect, or failure mode. Rather than voting, it is recommended t
38、hat the team try to reach consensus. In case of disagreements, knowledge experts opinion can be used to increase understanding, or more data can be sought. In any case the teams decision should prevail. Document on the worksheetProductFailure EffectSCauseOCurrentDRPNProcessModeControlsor ServiceCarO
39、verheatsStranded on 7 Cooling3 Temp Gage4RoadSystemBinding3 None10brakesOverloaded8 Load capacity8listed on side of doorSimply assigning ratings for Severity, Occurrence, and Detection doesnt helpquantify the risk into a single measurable value. Multiplying the rating scoresthus fills this gap. Note
40、 that in this approach we have maintained equal weight for each of the three S, O, and D. Example:ProductFailure EffectSCauseOCurrentDRPNProcessModeControlsor ServiceCarOverheatsStranded on 7 Cooling3 Temp Gage484RoadSystemBinding3 None10210brakesOverloaded8 Load capacity8448listed on side of doorId
41、entify the failure modes you will choose to work on. It may not be practical to address all of them. This critical step must be conducted with customer involvement, and consensus building. Many teams run into difficulties here, especially when the customer insists all items need to be addressed. It
42、needs to be understood thatsome element of risk will always be present. The question therefore is “What level of risk is acceptable”?Some of the different approaches that have been used to define the thresholdare; Pareto Analysis, Top cause for each failure mode, etc.Document on FMEA startup form to
43、 indicate this was discussed, and agreement reached on the threshold value. CausesCumulative RPNCumulative RPN Curve80% of Cumulative RPNProductFailure EffectSCauseOCurrentDRPNProcessModeControlsor ServiceCarOverheatsStranded on 7 Cooling3 Temp gage484RoadSystemBinding3 None10210brakesOverloaded8 Lo
44、ad capacity8448listed on side of doorExtreme3 Listen to 5105hot weatherforecastsSkidsAccident10 Slick roads7 Listen to 5350InjuryforecastsWorn tires4 Yearly insp.4160Driving habit7 ABS and 5350traction controllArrive at the order in which you will start addressing the failure modes and causes for co
45、rrective action. Although RPN and severity based prioritization seems the most logical, your decision to proceed must be made with customer input. lThis step is easily accomplished by sorting the data, in descending order. You can do this within each failure mode to maintain data integrity.lDocument
46、ation on work sheet is important to ensure the whole team is in sync. ProductFailure EffectSCauseOCurrentDRPNProcessModeControlsor ServiceCarOverheatsStranded on road 7 Overloaded8Load capacity listed on side of door8448Binding brakes3 None10210Extreme hot weather3Listen to forecasts 5105Cooling sys
47、tem3 Temp gage484SkidsAccident - injury10 Slick roads7Listen to forecasts5350Driving habit7ABS and traction control5350Worn tires4 Yearly insp.4160lIn this step you identify the corrective actions needed to eliminate / controlthe causes, and reduce the risk level to or below the agreed upon threshol
48、d.There will be times when you cannot work on the root cause; in such casesthe best course of action will be via addition of controls / adequate warningdevices / mechanisms of impending failure. Re-evaluate the ratings for S, O, and D, followed by RPN calculation.lDiscussion with the customer is imp
49、ortant again since the corrective actions willimpact them. Knowledge experts may also provide valuable input as to what kinds of corrective action will be best suited for the given cause / failure mode. lFinally, from all possible corrective actions, the team may choose one based on cost / benefit a
50、nalysis, speed of implementation, and the new level of risk. Product / Process / ServiceFailure ModeEffectSCauseOCurrent ControlsDRPNCorrective ActionAssigned ToComp DateSODRPNCarOverheatsStranded on road 7 Overloaded8Load capacity listed on side of door8448Install load capacity signs easily visible