91 2. The second step consists of evolvin, tools, equipments, input materials, production techniques and final products, Th e inspection and testing procedures would help the manufacturer to determine whether the pre-determined quality standards are achieved or not, . The third step consists of Periodical review and evaluation of the various methods and procedures of inspection and quality control is essential to measure their effectiveness as well as efficiency of the system. On the basis of such review and evaluation if some changes are essential in the inspection procedures, these should be incorporated without any delay. In addition to all this, random checking of the Products as well as machines / equipments would lead to quality improvement as per design specifications, Ss (CAL QUALITY CONTROL (SQC): iE “Aquality Control System performs inspection, testing and analysis to conclude whether the quality of each product is as per laid standard or not. It is called Statistical Quality Control when statistical techniques are employed to control quality. SQC makes inspection more reliable and at the same time less costly. Using statistical technique, SQC collects and analyses data in controlling product quality. It (SQC) permits a more fundamental control and scientifically fixes the process tolerances. SQC uses the following techniques in achieving the quality the product: 1. Control Charts 2. Sampling Inspection Scanned with CamScannerDT d/S<_S’SCZ i ee . 92 gC ean be further summarized as follows sl sQC has been a tool used to control the inspection. quality of end products based on In recent years the emphasis has been shifte it d to preventing defects from occurring during the production cycle, . Inspections are made to assess the quality of the entire population by examining a random sample. The Purpose is to ensure that the number of defects is below a maximum permitted number, 4. Control charts are used to analyse the process trend, to enable appropriate actions to be taken before the Process goes out of the limits set by the quality standards. Control Limits A normal distribution curve is shown in Figure below. In this curve — Bell shaped Normal curve If xi] ------------ l | | | Arithmetic mean of distribution curve Standard deviation +6 represents 68% of the area + 2c represents 95% of the area X-+ 30 represents 99.7% of the area 1 pl Qt Normal Distribution Curve Normal Distribution Curve Scanned with CamScannerrr 9 Normal Curve is symmettical and bell shaped: 2o limits occupy 95% of the area of the normal curve and indicate that one is 95% confident that a random observation will fall in this area. Advantages of SQC The advantages of SQC are listed below: 1. Provides a way of deleting error at inspection 2. More uniform quality of production . Reduces inspection cost . Reduces number of rejects 3 4. 5. Provides a basis for attainable specification 6. Points out the bottlenecks of trouble spots. 7, . Provides or way to determine the capacity of manufacturing process, Process Control When a product is examined, certain amount of variations are observed in the quality characteristics which the customer wants. These variations are due to two reasons: . 1. Chance Variation 2. Variations due to assigned causes. 1. Chance Variation: Chance variation occur in random manner and there is very little we can do about them to control. These are natural to the process. For example quality of raw materials, performance of operator, changes in weather conditions. These variations are uncontrollable inherent, random and commonly present. Scanned with CamScanner> Variations due to Assigned Cais, ‘Vhere variations are not random Wis generally caus sometimes by more fyetors, Por e: and ave relatively larger, i and ave relatively: larger ed by only one factor but NuMple s poor quality from a Hew supplier, poor surfree raw material finish due to worn out tool etc, These jations are assignable ya & specific. Chance variation cause about 85% of the problems while assignable variation accounts for only 15% of problem. Chance variations may be reduced by changing the process and this can be done through management effort ‘To do so, there may be more efforts needed in better machine procurement, better training, better environment & weather (temperature) control ete. gnable variations, may be reduced by controlling the process with proper identification & rectification, Workers, who are close to machines, can understand the causes & rectify them. Hence, it is workers responsibility to control the assignable causes. This can be achieved by first identifying the variations, relating them to the causes and finally solving them by proper rectification measures. For the identifications of assignable variations, we use Control Charts. THEORY OF CONTROL CHARTS : ae information. A control Feud Chart is a graphical presentation of collected information. f i tur the specified Chart detects variations and warns if there is any departure from pt art is a diagnostic technique, yells cause and eliminating 1 Ch it immediately tells the tolerance limits. Control Chi in exploring the undesired variations and helps ™ exploring manufacturing troubles. Scanned with CamScannerAdvantages of Control Charts 1. Control Charts indicate whether X- chart R=chart L_ 6~chart [— Variable or — Measurement chart Control Charts ___| [~ P=chart [7 aP~chart ~~ _ C= chart Attribute chart ‘— U-chart Vayiable Charts 1. Variable Charts involve the measurement of the job dimensions & an item is accepted or rejected if its dimensions are within or beyond the fixed tolerance limits. 2. Variable Charts are more detailed and contain more information. 3. Variable Charts are relatively expensive because of greater cost of collecting data. rc bute Charts 1. Attribute Charts only differentiates between a defective & non- defective item without going into the measurement of its dimensions. Scanned with CamScanner| 2. Attribute Charts, being based u PON go and no 20 dat sample size. ‘a, require bigger For example : Checking Color or Finish of a product. In such cases the answer is either Yes or No. Types of Control Charts There are varieties of Control Charts but we shal (a) & R - charts (variable chart) (b)P—Charts (attribute chart) (c)C-Charts (attribute chart) Il be dealing with - (a) X & R— Charts : These are the charts for variables (measurable) « Average and Range (X, R) (i) X- Chart, Control limits (CLs) =%+ A, R Gi) R- Charts, UCL =D,R LCL=D;R * Average and standard deviation (X ,o) (b) P- Charts: These are the charts for attributes (good or bad) d defects n Fraction defectives =P == Sample size 52d tive = P= Average Fraction defect Control Limits = CLs= Scanned with CamScanner(o) C-Chart ¢ are charts for attribute and used when the number of possible defects is large but the expected occurrence is small. Control limits = CLs = € +3 VG Where, C=Number of defects in sample n= Number of samples =c Average defect = C= verage defect = C in ¢& R-Chart Example: Ten samples of size 5 each were drawn. The following observations were made (i) Determine UCL, LCL for X — Chart and R-Chart (ii) Draw the chart (Given: for 1=5, Ax=0.58D3=0, Dy=2.11). |Sample 1 2 3 4 5 6 7 8 9 10 | No. | x 70 |7.5 |8 To 195 [ill fils |4 35 «(4 R 2 3 2 2 3 4 3 2; 3 2 Solution: Yx=76 YR=26 Scanned with CamScannerOg 3 4s 678 9 10 Sample Number» C1 2°53 4 5 6728 9 10 Sample Number R= Chart zx -2=76 a Number of samples 10 Scanned with CamScanner| 99 Control limits - For X- Chart, UCL = + AR = 7.6 + 0.58 x 2.6 = 9,11 LCL = — A2R = 7.6 + 0.58 x 2.6 = 6.09 For R- Chart, UCL =D, R= 2.11 x 2.6 = 5.49 LCL =D; R= 0.x 2, =0 ircan be seen from X — Chart that the process goes out-of control from the fourth sample onwards. p-Chart (Percentage Defective or Fraction Defective) The following example shows how this chart is drawn Example: Ten samples of size 300 each were inspected. The results are shown below. Draw a chart for fraction defectives, Sample No. 1 {2 ]3 [4 ]s J6 [7 ]s Jo Jo } Sample Size 300 | 300 | 300 | 300 | 300 | 300 | 300 | 300 | 300 | 300 Number of defective pieces in |25 |30 |35 |40 |45 |35 |40 |30 |20 |so0 the sample Solution: n=300; k=10 Sample No. T 2 9 | 2 a 3 eA Pieces inspected=n 1300 [300 |300 [300 [300 [300 [300 | 300 }300 Number of defective Pieces found = d 25 30 |35 40 45 135 40 30 |} 20 Fraction defective “p” | 0.083 | 0.10 | 0.1167 | 0.1333 [0.15 | 0.1167 | 0.133 | 0.10 | 0.0666 wf i8 7 oe Percent defective= [83% | 10% | 11.67 13.33 [1S [1167 [13.33 10 [6.67 1100p | ft ft Scanned with CamScanner100 = 200 3000 = 9.1167 i average fraction defective z= fraction Defective a= No. of defects ample size UCL = —— 0. ca 1167+3 1 eatertacoi67) = 04723 0.17 0.16 | { UCL = 0.1723 0.15 ¢ 4 0.13 + Mean = 0.1167 0.11 + 0.10 + , N 0.09 ;- / \ 0.08 |. \ p07 | Vo | Lei=0.0611 Fraction Defective (p) ————> 0.06 L t 10 oy 2 3 4 5 678 9 Sample number — fraction defective Chart showin. Scanned with CamScannerhc Chart he following example shows how the char j is drayp, gxample: Carpets Were inspected, The results ape, tabulated below, Cons control chat for the chat for the number Of defects, mune [No. of carp [[Link] carpet (ny | pale 5{61714| OA [ No. of defects (ce [Xe of detects fe) [2 | 56 [i}s}r ta \5 | Solution: Average number of defects a retil number ofdtecsing carpets _ Ye © = n0, of defects in samples n n= no, of samples, a _ 245454641454147 = a =% 8 Means =4 and ‘ UCL = +30 =443V4=10 LCL=C-— 3vt =4-3/4=-2=0 (as there can not he negative defect UCL = 10 Mean = 4 Number of defects —————> Co 10 789 oy 2 3 4 5 6 umber of carpets. ——— Scanned with CamScanner