1、FAILURE MODE and EFFECTS ANALYSIS 失 效 模 式 及 其 影 響 分 析,Philips CFT Asia Pacific Innovation ,SxOxD,RPN125,SODDescription Action? 1011Failure unlikely to reach user 1101Frequent Minor, detectable failure 1110Minor defect reaches user infrequently 10101Serious and frequent failure 11010Frequent minor de

2、fect reaches user 10110Serious defect reaches user 101010IN TROUBLE!,Step 5 - Establish Actions and Responsibilities,1. What has to be improved : - change of design - life test - contact with supplier - control point build in, in production process etc. 2. Who is responsible 3. When is it finished,R

3、PN acceptable?,Corrective actions,DO,New RPN1,Acceptable?,End,Yes,No,No,Yes,RPN acceptable?,Corrective actions,DO,New RPNn,Acceptable?,End,Yes,No,No,Yes,Choice of Corrective Actions,Unlikely to be effectivePreferred choices *Increase inspection *Poka-yoke (fool-proofing) *Additional test*Process cha

4、nge *Additional check*design change *Re-train operators*Automation *Re-write SOP*Computerisation * Robotics,Corrective Action Preference Hierarchy,The intent of any recommended corrective action is to improve the RPN rating by reducing Severity, Occurrence, Detection or all 3 together. Emphasis shou

5、ld be on prevention, i.e. reducing Occurrence. Should prevention be impossible, the following Hierarchy of preference should be used: 1. Prevention, or potential cause elimination 2. Detection, as early in the product life cycle as possible: best at the design stage next best prior to mass productio

6、n then prior to shipment and, as last resort in the field, prior to catastrophic failure 3. Impact reduction, i.e. reduce Severity Generally, to reduce any of the three rating - O, D, S - process and/or design changes/modification are required,FMEA,is is not * structured* a show-piece * systematic*

7、a number game * rigorous* a guessing game * comprehensive* a documentation exercise * scientific* a self-consolation exercise * a discipline. * preventive. * customer-focused. * a team effort. * a cross-functional exercise. * a living document. * a continuous process* a one-off exercise,Real Life FM

8、EA Exercise,3PRODUCT D-FMEA,Process FMEA vs Product D-FMEA,*P-FMEA does not rely on product design changes to overcome process weaknesses, but it does take into consideration a products design characteristics *D-FMEA does not rely on process controls to overcome design weaknesses, but it does take t

9、he technical/physical limitations of the manufacturing process (i.e. the process capability) into consideration.,Process step,Purpose,Piece-part or sub-assy,Function,Process FMEA,Product FMEA,Process FMEA vs Product D-FMEA,*Hence causes such as “operator not inserting resistor properly” noisy; jerky

10、; cannot what the customer would observerun at high speed and/or experience Next higher level effectseffects on the engineover-heated; vibration Local effectseffects on the valve itselfdeterioration,MATERIAL,USE,DESIGN,COMPONENT,SPECIFICATION,ENVIRONMENT,PRODUCT,Product Cause - Effect Investigation,

11、Product Functions,Product Functions,Product Functions,Severity (D-FMEA),RatingDegreeEffects 1Minor* no perceptible effect on product performance * user will probably not notice the failure 3 low* customer notices slight deterioration in performance * causes slight customer annoyance 5 Moderate* caus

12、es some customer dissatisfaction * customer notices reduced product performance * gradual performance deterioration 7High* high degree of customer dissatisfaction * loss of sub-system function, eg. engine fail 10 Very high* safety-related * non-compliance with government regulations,Occurrence (D-FM

13、EA),RatingDescriptionProbability of failureDesign life failure rate 1RemoteFailure unlikely 1 ppm 3Low Relatively low failure rate 200 ppm 1000 ppm 5ModerateOccasional failures2500 ppm 1 in 80 (1.25%) 7HighRepeated failures 1 in 40 (2.5%0 1 in 20 (5.0%) 10Very highFailure almost inevitable 1 in 8 (1

14、2.5%) 1 in 2 (50.0%),Detection,ApproachDescription Application Time-tableHow early in product life cycle D-FMEA from design start to Field failure ? Fixed-gateWhats the probability of detection D-FMEA at any point (gate) in the cycle ?,Detection (D-FMEA),Time-table approach Rating Probability of det

15、ection 1 Will be detected prior to Production Release 3 Very likely to be detected prior to Final Release 4 Detectable after release but before Mass Production 5 Detectable prior to Shipment 7 Will be detected before occurrence in the field 10 Undetectable until failure occurs in the field,Detection

16、 (D-FMEA),Fixed Gate approach Rating Probability of detection 1 Will be detected in design evaluations 3 High chance of detection in design evaluations 5 Moderate chance of detection in design evaluations 7 Low chance of detection during design evaluations 10 Undetectable until failure occurs in the

17、 field,4ORGANIZATIONAL ASPECTS OFFMEA,Organisational Aspects of FMEA,General Points Participants Preparations Reporting,FMEA is a team effort covering all relevant disciplines FMEA sessions should not last longer than half a day The product FMEA should be focused on:- reliability (normal functioning

18、)- producibility The process FMEA should be focused on:- process flow- manipulations of operator- deviations in material- transport- storage,General Points,FMEA : Critical Success Factors,*Be honest *Be positive *Be thorough *Attention to details * Follow through *Involve the right people *“Rather b

19、e safe than sorry” *Management by fact -Minimise guess-work and gut-feel -Maximise use of data and quantitative methods,Participants,Participants should be directly involved Number of participants should be limited Participants from relevant departments:- Development- Engineering- Production- Proces

20、s - Quality- Specialist (if necessary)- Service,Preparations for D-FMEA,Preparations should be done by the designer, in close co-operation with the chairman of the meeting Drawings and sufficient models should be available One model should have been in circulation Tolerance calculations and contour

21、maps should be available Special preparations for a D-FMEA:- the design should be breakdown into elements- priority settings of these elements- from existing products the following data should be available:. Presale data. Service call rate figures. Fall off figures,Reporting,Result of FMEA reported

22、on standard forms Action list should be set up from the session and should be:- presented on the milestone meetings- known and tracked within the project team- discussed at every project team meeting,5MATURITY GRID,Use of the Maturity Grid,Objective: To track maturity in design of product & process

23、To facilitate Milestone decisions To visualise status: also problems to be solved in the next phases Inputs for Maturity Grid: Outstanding Design Issues (Tech Spec, Customer Requirements) Problems found in Design Evaluation / Process Trails Problems found in Reliability tests Potential problems iden

24、tified in FMEA sessions,FMEA provides input to the maturity grid, but is not the only one!,Maturity Grid for Products & Processes,Gravity Factor: S - non conformity with safety standard /other requirement A - product not producible or saleable B - not accepted by critical customer or production with

25、 big problems C - can be sold or produced with minor difficulties D - problem accepted (no actions) Evolution Factor: 4 - cause not known 3 - solution not known 2 - evaluation not yet positive 1 - solution not yet introduced 0 - solutions introduced,Product/Process maturity grid,Gravity Factor,Evolution Factor,S,A,B,C,D,4,3,2,1,0,AFM,CD,DR,CR,IR,Potential problems