Statistical Process Control
What is a process? Inputs PROCESS Outputs A process can be described as a transformation of set of inputs into desired outputs.
Types of Measures Measures where the metric is composed of a classification in one of two (or more) categories is called Attribute data. _ Good/Bad Yes/No Measures where the metric consists of a number which indicates a precise value is called Variable data. Time Miles/Hr October 2, 212 3
Population Vs. Sample (Certainty Vs. Uncertainty) A sample is just a subset of all possible values sample population Since the sample does not contain all the possible values, there is some uncertainty about the population. Hence any statistics, such as mean and standard deviation, are just estimates of the true population parameters. October 2, 212 4
WHY STATISTICS? THE ROLE OF STATISTICS Statistics is the art of collecting, classifying, presenting, interpreting and analyzing numerical data, as well as making conclusions about the system from which the data was obtained. October 2, 212 5
Descriptive Statistics Descriptive Statistics is the branch of statistics which most people are familiar. It characterizes and summarizes the most prominent features of a given set of data (means, medians, standard deviations, percentiles, graphs, tables and charts. October 2, 212 6
Inferential Statistics Inferential Statistics is the branch of statistics that deals with drawing conclusions about a population based on information obtained from a sample drawn from that population. October 2, 212 7
WHAT IS THE MEAN? The mean is simply the average value of the data. mean ORDERED DATA SET -5-3 -1 xi -1 = x = = 2 =.17 n 12 n=12 1 3-6 -5-4 -3-2 -1 1 2 3 4 5 6 4 Mean x i = 2 October 2, 212 8
WHAT IS THE MEDIAN? If we rank order (descending or ascending) the data set,we find the value half way (5%) through the data points and is called the median value. ORDERED DATA SET -5-3 -1-1 Median 1 3-6 -5-4 -3-2 -1 1 2 3 4 5 6 4 Median Value 5% of data points October 2, 212 9
WHAT IS THE MODE? If we rank order (descending or ascending) the data set We find that a single value occurs more often than any other. This is called the mode.. ORDERED DATA SET -5-3 Mode Mode -1-1 1 3-6 -5-4 -3-2 -1 1 2 3 4 5 6 4 October 2, 212 1
WHAT IS THE RANGE? The range is a very common metric. ORDERED DATA SET -5 To calculate the range simply subtract the minimum value -3 in the sample from the maximum value. -1-1 Range = x MAX x MIN = 4 ( 5) = 9 Range 1 3-6 -5-4 -3-2 -1 1 2 3 4 5 6 4 Range Min Max October 2, 212 11
WHAT IS THE VARIANCE/STANDARD DEVIATION? The variance (s 2 ) is a very robust metric. The standard deviation(s) is the square root of the variance and is the most commonly used measure of dispersion. 2 DATA SET X i -5 X = = 2 Xi X ( X i X ) = -.17-5-(-.17)=-4.83 (-4.83)2=23.32 n 12-3 -3-(-.17)=-2.83 (-2.83)2=8.1-1 -1-(-.17)=-.83 (-.83)2=.69 ( X X) 2 2 i 6167 s = =. -1 5 6 n 1 12 1 = -1-(-.17)=-.83 (-.83)2=.69. -(-.17)=.17 (.17)2=.3 -(-.17)=.17 (.17)2=.3 -(-.17)=.17 (.17)2=.3 -(-.17)=.17 (.17)2=.3 1 -(-.17)=.17 (.17)2=.3 3 1-(-.17)=1.17 (1.17)2=1.37-6 -5-4 -3-2 -1 1 2 3 4 5 6 4 3-(-.17)=3.17 4-(-.17)=4.17 (3.17)2=1.5 (4.17)2=17.39 61.67 October 2, 212 12
Statistical Process Control (SPC) Measures performance of a process Uses mathematics (i.e., statistics) Involves collecting, organizing, & interpreting data Objective: Regulate product quality Used to Control the process as products are produced Inspect samples of finished products
CONTROL CHART Functions of a Process Control System are To signal the presence of assignable causes of variation To give evidence if a process is operating in a state of statistical control
CONTROL CHART Essential features of a control chart Variable Values Upper Control Limit Central Line Lower Control Limit Time
Show changes in data pattern e.g., trends Control Chart Purposes Make corrections before process is out of control Show causes of changes in data Assignable causes Data outside control limits or trend in data Natural causes Random variations around average
Quality Characteristics Variables 1. Characteristics that you measure, e.g., weight, length 2. May be in whole or in fractional numbers 3. Continuous random variables Attributes 1. Characteristics for which you focus on defects 2. Classify products as either good or bad, or count # defects e.g., radio works or not 3. Categorical or discrete random variables
CONTROL CHART Types of Control Charts for Attribute Data Description Type Sample Size Control Chart for proportion non conforming units Control Chart for no. of non conforming units in a sample Control Chart for no. of non conformities in a sample Control Chart for no. of non conformities per unit p Chart np Chart c Chart u Chart May change Must be constant Must be constant May Change
Control Chart Types Control Charts Variables Charts Attributes Charts R Chart ` X Chart P Chart C Chart
X Chart Type of variables control chart Interval or ratio scaled numerical data Shows sample means over time Monitors process average and tells whether changes have occurred. These changes may due to 1. Tool wear 2. Increase in temperature 3. Different method used in the second shift 4. New stronger material Example: Weigh samples of coffee & compute means of samples; Plot
R Chart Type of variables control chart Interval or ratio scaled numerical data Shows sample ranges over time Difference between smallest & largest values in inspection sample Monitors variability in process, it tells us the loss or gain in dispersion. This change may be due to: 1. Worn bearing 2. A loose tool 3. An erratic flow of lubricant to machine 4. Sloppiness of machine operator Example: Weigh samples of coffee & compute ranges of samples; Plot
Construction of X and R Charts Step 1: Select the Characteristics for applying a control chart. Step 2: Select the appropriate type of control chart. Step 3: Collect the data. Step 4: Choose the rational sub-group i.e Sample Step 5: Calculate the average ( X) and range R for each sample. Step 6: Cal Average of averages of X and average of range(r)
Construction of X and R Charts Steps 7:Cal the limits for X and R Charts. Steps 8: Plot Centre line (CL) UCL and LCL on the chart Steps 9: Plot individual X and R values on the chart. Steps 1: Check whether the process is in control (or) not. Steps 11: Revise the control limits if the points are outside.
X Chart Control Limits UCL x LCL x = x + A R 2 = x A R 2 From Tables Sub group average X = x1 + x2 +x3 +x4 +x5 / 5 Sub group range R = Max Value Min value
R Chart Control Limits UCL R = D 4 R From Tables LCL R = D 3 R
Problem8.1 from TQM by V.Jayakumar Page No 8.5
p Chart for Attributes Type of attributes control chart Nominally scaled categorical data e.g., good-bad Shows % of nonconforming items Example: Count # defective chairs & divide by total chairs inspected; Plot Chair is either defective or not defective
p Chart p = np / n where p = Fraction of Defective np = no of Defectives n = No of items inspected in sub group p= Avg Fraction Defective = np/ n = CL
p Chart Control Limits UCL p = p+ z p(1 n p) z = 3 for 99.7% limits LCL p = p z p(1 n p)
Purpose of the p Chart Identify and correct causes of bad quality The average proportion of defective articles submitted for inspection,over a period. To suggest where X and R charts to be used. Determine average Quality Level.
Problem Problem 9.1 Page no 9.3 TQM by V.Jayakumar
np CHART P and np are quiet same Whenever subgroup size is variable,p chart is used. If sub group size is constant, then np is used. FORMULA: Central Line CLnp = n p Upper Control Limit, UCLnp = n p +3 n p (1- p ) Lower Control Limit, LCLnp = n p -3 n p (1- p ) Where p = np/ n =Average Fraction Defective n = Number of items inspected in subgroup.
Problem Problem No 9.11 page No 9.11 in TQM by V.Jayakumar
c Chart Type of attributes control chart Discrete quantitative data Shows number of nonconformities (defects) in a unit Unit may be chair, steel sheet, car etc. Size of unit must be constant Example: Count no of defects (scratches, chips etc.) in each chair of a sample of 1 chairs; Plot
c Chart Control Limits UCL c = c + 3 c Use 3 for 99.7% limits LCL c = c 3 c