Statistical Process Control for learning
- 1. WELCOME TRAINING PROGRAMME ON STATISTICAL PROCESS CONTROL
- 3. STATISTICAL PROCESS CONTROL CONTENT 6.TOOLS FOR STATISTICAL CONTROL 7.PROCESS CAPABILITY 8.PROCESS CAPABILITY INDICES 9.CONTROL CHART TYPES 10 CONTROL CHART MEHODOLOGY EVALUATION
- 4. Mistake Proofing 100% Inspection Statistical Process Control In this method more advanced and modern techniques are used which require substantial investment during its installation and maintenance. As it is detection type of technique it can’t avoid failure but rejects defective products. Requires more inspectors, more inspection times and in turn more cost. For this technique investment is very less and process is controlled on each workstation therefore defective components is not forwarded to next operation. Predictability reduces frequent adjustments & in turn increases productivity, reduces inspection cost at station & at final inspection
- 5. SPC The Use of Statistical Techniques such as Control charts to analyze the process or its output so as to take appropriate action to achieve and maintain a state of statistical Control and to Improve the process Stability.
- 6. QUALITY OF CONFORMANCE Vs CONFORMANCE TO SPECIFICATION
- 7. PROCESS CONTROL THREE TYPES OF PROCESS CONTROL PREVENTION OF DEFECTS * MISTAKE PROOFING DETECTION OF CAUSES AND LEAD TO CORRECTIVE ACTION * VISUAL CONTROL, SPC DETECTION OF DEFECTS * INSPECTION
- 8. VARIATION PRINCIPLES OF SPC VARIATION IS INEVITABLE VARIATION IS MEASURABLE
- 9. STATISTICAL PROCESS CONTROL STATISTICAL PROCESS CONTROL 10 % IS STATISTICS 90 % IS PRODUCT & PROCESS KNOWLEDGE
- 10. STATISTICS DETERMINE WHICH SAMPLE IS GOOD ? SAMPLE –1 15,15,14,15,16,15,15 SAMPLE –2 14,15,20,15,16,10,15 SAMPLE - 3 19,20,15,14,11,10,16
- 11. STATISTICS HENCE WE NEED TO CAPTURE AVERAGE RANGE STANDARD DEVIATION TO EVALUATE SAMPLES
- 12. VARIATION & CAUSES TYPES OF VARIATION RANDOM VARIATION NON RANDOM VARIATION CAUSES OF VARIATION COMMON OR CHANCE CAUSES SPECIAL OR ASSIGNABLE CAUSES
- 13. CAUSES COMMON CAUSES SPL./ASSIGNABLE FEW IN NOS. PLENTY IN NOS. VARIATION IS LOW VARIATION IS HIGH PART OF THE PROCESS VISITOR TO THE PROCESS CONSTANT VARIATION FLUCTUATING VARIATION
- 14. CAUSES COMMON CAUSES SPL./ASSIGNABLE STATISTICS APPLICABLE STATISTICS CANNOT APPLY MANAGEMENT CONTROLLABLE OPERATOR CONTROLLABLE e.g Pressure variation, Environment variation e.g Wrong setting, wrong master REDUCTION LEAD TO IMPROVEMENT ELIMINATION LEAD TO MAINTENANCE
- 15. TOOLS FOR CONTROL HISTOGRAM ----------- BELL SHAPE CONTROL CHART-- NO OUT OF CONTROL
- 16. NORMAL DISTRIBUTION TAKE A SAMPLE OF 100 ASSUME MEAN = 50.00 S. D. = 1.5 48.5 -------------51.5 (1sigma) 68.20% 47.0--------------53.0 (2Sigma) 95.4% 45.5--------------54.5 (3Sigma) 99.73%
- 17. NORMAL DISTRIBUTION WHEN AVERAGE, MEDIAN & MODE IS SAME AVERAGE OF ALL VARIABLE DATA FOLLOWS NORMAL DISTRIBUTION ELONGATION,UTS,TWIST,OVALITY, ETC., LIKE ONE SIDED DOES NOT FOLLOW NORMAL DISTRIBUTION
- 18. STATISTICAL CONTROL & PROCESS CAPABILITY A PROCESS FREE FROM ASSIGNABLE CAUSES (PREDICTABLE PROCESS) PROCESS CAPABILITY IS A MEASURE OF INHERENT VARIATION (MANAGEMENT CONTROLLABLE)
- 19. PROCESS CAPABILITY Cp=Potential Process Capability Index Cp = TOLERANCE ---------------------- TOTAL VARIATION (6 SIGMA)
- 20. PROCESS CAPABILITY INDICES TV Vs TOL PPM LEVELS 6 SIGMA > TOL 125 > 100 Cp = 0.8 HIGHER REJECTION 6 SIGMA = TOL 100 = 100 Cp = 1.0 2700 PPM 6 SIGMA < TOL 75 < 100 Cp = 1.33 64 PPM 6 SIGMA < TOL 60 < 100 Cp = 1.67 4 PPM
- 21. PROCESS CAPABILITY INDICES Cp DOES NOT SPECIFY WHERE THE PROCESS IS CENTERED. HENCE WE NEED TO HAVE ONE MORE INDEX TO MEASURE ACTUAL PROCESS CAPABILITY Cpk = ACTUAL PROCESS CAPABILITY INDEX Cpk = Min ( USL – AVG / 3 SIGMA, AVG – LSL/ 3 SIGMA) USE Cp AND Cpk TOGETHER Cpk CANNOT EXCEED Cp
- 22. PROCESS CAPABILITY INDICES Pp & Ppk Cp &Cpk PROCESS PERFORMANCE INDEX PROCESS CAPABILITY INDEX USED DURING INITIAL PROCESS STUDY DURING PPAP ONGOING PROCESS CAPABILITY STUDY CAN BE CAPTURED FOR STABLE AND CHRONICALLY UNSTABLE PROCESSES USED ONLY FOR STABLE PROCESSES
- 23. PROCESS CAPABILITY INDICES Pp & Ppk Cp &Cpk CAPTURES VARIATION DUE TO BOTH COMMON & SPECIAL CAUSES CAPTURES VARIATION DUE TO COMMON CAUSES ONLY SIGMA IS CALCULATED USING n-1 FORMULA USING ALL INDIVIDUAL READINGS SIGMA IS CALCULATED USING R bar / d2 FORMULA Ppk > 1.67 Cpk > 1.33
- 24. CONTROL CHARTS OBJECTIVES OF CONTROL CHART TO DETECT SPECIAL/ASSIGNABLE CAUSES TO MAINTAIN THE ACHIEVED PROCESS CAPABILITY TO IDENTIFY THE OPPORTUNITY FOR IMPROVEMENT
- 25. CONTROL CHART STEPS 1. GATHER DATA 2. INITIAL STUDY 3. CALCULATE CONTROL LIMITS 4. ESTABLISH ONGOING CONTROL CHART 5. MONITOR, REVIEW AND IMPROVE PROCESS CAPABILITY
- 26. 1.GATHER DATA ELIMINATE OBVIOUS DEFICIENCIES IDENDTIFY THE FACTORS AFFECTING AVG. & RANGE UNDERSTAND PROCESS THROUGH MASTER CAUSE & WHY WHY ANALYSIS PLAN SAMPLE SIZE, FREQ, CONTROL CHART,NO. OF SUBGROUPS,ETC.,
- 27. SELECTION OF CHARTS X BAR & R CHART MOST SENSITIVE CHART MEDIAN & R CHART Used to compare output of several process e.g. same parts from two supplier. X BAR & S CHART SUBGROUP IS > 9, Process is stable & in control and the objective is to reduce variation.
- 28. SELECTION OF CHARTS X & MR CHART WHERE NOT SUITABLE FOR SUBGROUP SAMPLING INSPECTION IS COSTLY/ DESTRUCTIVE IN NATURE , long time inspection .
- 29. GATHER DATA RECORD PROCESS EVENTS WHILE COLLECTION OF DATA
- 30. INTERPRETATION FOR CONTROL FOCUS ON RANGE CHART ALL 4 CONDITIONS ARE CHECKED AND APPROPRIATE ANALYSIS MADE AND CAUSES IDENTIFIED FOR ALL 4 OUT OF CONTROL CONDITIONS
- 31. INTERPRETATION FOR CAPABILITY CALCULATE SIGMA, 6 SIGMA, Cp,Cpk IF Cpk < 1.33 INITIATE CA TO IMPROVE THE VALUE INITIATE CA PLAN FOR PROCESSES NOT UNDER STATISTICAL CONTROL IF Cpk >1.33 ESATBLISH ONGOING CONTROL
- 32. EFFECTIVENESS OF SPC MEASURED BY IMPROVED PROCESS KNOWLEDGE REVIEW AND REVISION OF UCL/LCL REDUCTION IN REJECTION ( ACTUAL Vs ESTIMATED) IMPROVED PRODUCTIVITY LESS INSPECTION AND ADJUSTMENT CUSTOMER SATISFACTION DUE TO LESS VARIATION
- 33. ATTRIBUTE CHART P-CHART - PROPORTION OF UNIT NONCONFORMING - SAMPLE SIZE NEED NOT BE EQUAL np-CHART - NUMBER OF UNITS NONCONFORMING - SAMPLE SIZE MUST BE EQUAL
- 34. ATTRIBUTE CHART C – CHART - NUMBER OF NONCONFORMITIES - SAMPLE SIZE MUST BE EQUAL U – CHART - NUMBER OF NONCONFORMITIES PER UNIT - SAMPLE SIZE NEED NOT BE EQUAL
- 35. OPERATOR’s ROLE FOR ONE SUBGROUP PLOTTING 1. CHECK AS PER CONTROL PLAN 2. CALCULATE AVG/RANGE/INDIVIDUALS 3. PLOT THE SAME IN CONTROL CHART 4. CHECK THE PLOTTED POINT IS IN CONTROL ( REFER 4 CONDITIONS IN CONTROL CHART) 5. IF IT IS IN CONTROL, CONTINUE THE PROCESS 6. ELSE, STOP THE PROCESS, TAKE CORRECTIVE ACTION & DISPOSITION ACTION AS PER REACTION PLAN 7. RECORD THE PROCESS EVENTS INCLUDING OUT OF CONTROL CONDITIONS
- 36. EFFECT OF OVER ADJUSTMENT IF A STABLE PROCESS IS ADJUSTED ON THE BASIS OF EACH MEASUREMENT MADE, THEN THE ADJUSTMENT BECOMES AN ADDITIONAL SOURCE OF VARIATION OVER ADJUSTMENT WILL INCREASE THE VARIATION
- 37. Thank you