Process Quality Control Training
0 likes421 views
This document provides an overview of process quality control using statistical process control (SPC) and statistical quality control (SQC) approaches. It defines SPC and SQC, noting that SPC focuses on controlling process inputs through variables while SQC monitors outputs through attributes. The document outlines key learning objectives around these topics. It also defines key terms like process quality control and discusses the difference between SPC and SQC. Additionally, it covers process capability analysis using Minitab and controlling process inputs and monitoring outputs. Overall, the document serves as training material on quality control tools and techniques with a focus on SPC and SQC.
![ If Cp == Cpk, then the process is perfectly centered. If perfectly centered, Cp == Cpk.
Because Cpk accounts for centering (where Cp does not), Cpk can never be larger than Cp.
Both assume a stable process.
Cpk measures how close a process is performing compared to its specification limits and accounting for the
natural variability of the process.
Larger is better. The larger Cpk is, the less likely it is that any item will be outside the specification limits.
When Cpk is negative it means that a process will produce output that is outside the customer specification
limits.
When the mean of the process is outside the customer specification limits the value of Cpk will be Negative
We generally want a Cpk of at least 1.33 [4 sigma] or higher to satisfy most customers.
17Process Capability Analysis (5)](https://image.slidesharecdn.com/processqualitycontroltrainingv3-11july2020-200829220432/85/Process-Quality-Control-Training-17-320.jpg)
Process Quality Control Training
- 1. Process Quality Control – SPC, SQC Approach JULY 2020 1 Dominic E. Nwagbaraocha
- 2. Meet The Trainer DOMINIC E. NWAGBARAOCHA CHEMICAL ENGINEER (MSC, COREN) MBA, FINANCE AND INVESTMENT ISO 9001 LEAD AUDITOR CERTIFIED SIX SIGMA (CSSYB, CSSGB,CSSMBB) 9+ YEARS IN MANUFACTURING 2
- 3. Learning Objectives Process Quality Control SPC, SQC Defined Difference between SQC and SPC Controlling Process Inputs (independent variables) Process Capability with MINITAB Monitoring process outputs (dependent variables) 7QC-Tools with MINITAB 3
- 4. Process Quality Control Defined A set of Interrelated or interacting activities which transform Inputs to outputs Process A means of evaluating a process stability over a period of time hence ensuring its output (goods or services) is fit for use – DN’2020 4 Ensuring goods and services are fit for purpose Quality Evaluation of a process over time, hence ensuring its stability Control
- 5. SQC And SPC Defined Statistical process control (SPC) is the application of the 14 statistical and analytical tools (7-QC and 7-SUPP) to control process inputs (independent variables) Error Prevention Monitors process in real time Statistical quality control (SQC) is defined as the application of the 14 statistical and analytical tools (7-QC and 7-SUPP) to monitor process outputs (dependent variables) Error Detection Controls defects from reaching customer SPC SQC Although both terms are often used interchangeably, SQC includes acceptance sampling where SPC does not 5
- 6. The 14 Statistical & Analytical Tools 6 Pareto Diagram Cause & Effect Sheet Graphs & Charts Check Sheets Histogram Scatter Diagram Control Chart 7-QC Tools Essential Tools for Discovery Process Process Flowcharts 7-SUPP (additional quality tools)
- 7. WHY SPC & SQC Understanding the process and the specification limits Eliminating assignable (special) sources of variation, so that the process is stable Monitoring the ongoing production process, assisted by the use of control charts, to detect significant changes of mean or variation Reducing scrap and rework Improving productivity and overall quality Competing in today’s world markets 7 Specification Limits Set by customers What the customer wants Control Limits Set by the process What the process can do
- 8. PQC – PDCA Approach Plan • Identify the problem and the possible causes Do • Make changes designed to correct or improve the situation Check • Analyze the effect of these changes; using control charts to show the effects of changes over time. Evaluate the results, and then replicate or try something different Act • If successful, standardize the changes and work on further improvements. If unsuccessful, identify other ways to change the process or different problem causes 8
- 9. 7 Steps of QC Problem Solving STEP Pareto Diagram Cause & Effect diagram Graphs and Charts Check sheets Histograms Scatter Diagrams Control Charts 1 Select the topic 2 Understand the situation and set targets Understand the situation Set targets 3 Plan activities 4 Analyze causes Investigate relationship between causes and characteristics Investigate past and present situations Stratify Investigate temporal changes Investigate mutual relationships 5 Consider and Implement Countermeasures 6 Check Results 7 Standardize and Establish Control Very Effective Effective 9
- 10. Data Types Variables data is defined as a measurement such as height, weight, time, or length. Monetary values are also variables data. Generally, a measuring device such as a weighing scale, Vernier, or clock produces this data. It can contain decimal places e.g. 3.4, 8.2. Smaller sample size Quick to react For critical product characteristics 10 Attributes data is defined as a count such as the number of employees, the number of errors, the number of defective products, or the number of phone calls. A standard is set, and then an assessment is made to establish if the standard has been met. The No. of times the standard is either met or not is the count. E.g Label gauge. Does not contains decimal places, always whole numbers, e.g. 2, 15. Larger sample size Slow to react For summarizing characteristics
- 11. Out Of Statistical Control This happens when any of the following conditions are met Points beyond control limit Non random scatter within the control limits Run rule – nine or more consecutive points on the same side of the center line Trend rule – six or more consecutive points slowly moving up or slowly moving down. 11
- 12. Controlling Process Inputs (Independent variables) Process Control Evaluation of the process stability over time 12 Process capability Evaluation of how well process meets specification
- 13. Process Capability Analysis (1) 68.27% 99.999943% 99.9999998% 99.9937% 99.73% 95.45% -1s +1s +2s +3s +4s-2s-3s-4s x +5s +6s-6s -5s Quality should be controlled and guaranteed through proper measurement of process/machine capabilities (Cp-Cpk, Cm-Cmk). These values are used to see if the process/machine is statistically stable or not. In order to control quality by causes and not by results, process parameters capability (Cp-Cpk) should be measured If the quality characteristic is assumed to follow a normal distribution, where ±3σ includes 99.73% of the population, process capability is defined in the following way: Process capability = ±3σ or 6σ Process capability is influenced by Machine, Method, Man, Material. Machine capability is only related to machines quality parameters. The proportion of data under the curve shows that: • 68.27% of the measured values lie between ± 1s, • 99.73% of the measured values lie between ± 3s, • 99.999998% of the measured values lie between ± 6s S = standard deviation (it measure the dispersion of a set of data from its mean) μ = mean xi = individual x values N =total number of values Normal distribution 13
- 14. Process Capability Analysis (2) Cp < 1 Cp = 1 Cp = 1,33 Cp = 1,66 mean increasing capability Incapable just capable acceptable capable Tolerance LSL USL Cp: is about the spread of the Process Cpk is about the Centering the Process Cpk = 0,66 (Cp = 1,33) Cpk = 1 (Cp = 1,33) Cpk = 1,33 (Cp = 1,33) Towards the centre Out of specification = Defect risk centred X = tolerance X = X = Cp: Is the process capable to produce inside the limits? Is the car width fitting with the road width? Cpk: Is the process centred? Is the car driving in the middle of the road or decentred on left or right? If Cpk << 1 The process is not capable to produce products properly. If Cpk = 1 27 out of 10,000 produced items are out of tolerances. If Cpk >= 4/3 About 64 items out of 1,000,000 items are out of tolerances. With this low level of defect production, the process can be managed. 14
- 15. Process Capability Analysis (3) Remember the Car parking in the garage analogy? Cpk = Negative number: Your process will regularly crash the car into the wall Cpk =0.5: You have a good chance hitting the wall on entry Cpk =1: Your car may be just touching the nearest edge of the entry Cpk =2: Great! You have great clearance. You could double the width of your car before you hit the side of the garage Cpk =3: Excellent! You have excellent clearance. You could triple the width of your car before you hit the side of the garage 15 CPK can have an upper and lower value reported If the upper value Cpu is 2 and the lower CpL is 1, we say it has been shifted to the left. This tells us nothing about if the process is stable or not. We must report the lower of the 2 values. What are Good Values For CPK?
- 16. Process Capability Analysis (4) Cp Distribution Judgment Action Cp 1.67 Process capability is more than enough. Simplification of process control and cost reduction can be considered in certain cases. 1.67 > Cp 1.33 Process capability is sufficiently high. Ideal condition. Maintain it. 1.33 > Cp 1.00 Process capability is not sufficiently high, but is adequate. Control process properly and maintain it in a control state. Defects may result if Cp approaches 1. Take action if needed. 1.00 > Cp 0.67 Process capability is not sufficient. Defects have been generated. Screening inspection and process. control and Kaizen will be required. 0.67 > Cp Process capability is vary low. Cannot satisfy quality. Quality must be improved, cause must be pursued and emergency actions must be taken. Reexamine standards. >_ >_ >_ >_ Judgment on the process based on the value of Cp To use properly Cp – Cpk please refer to proper Statistical Process Control (SPC) theory 16
- 17. If Cp == Cpk, then the process is perfectly centered. If perfectly centered, Cp == Cpk. Because Cpk accounts for centering (where Cp does not), Cpk can never be larger than Cp. Both assume a stable process. Cpk measures how close a process is performing compared to its specification limits and accounting for the natural variability of the process. Larger is better. The larger Cpk is, the less likely it is that any item will be outside the specification limits. When Cpk is negative it means that a process will produce output that is outside the customer specification limits. When the mean of the process is outside the customer specification limits the value of Cpk will be Negative We generally want a Cpk of at least 1.33 [4 sigma] or higher to satisfy most customers. 17Process Capability Analysis (5)
- 18. Cp Values If the ratio is greater than one, then the Engineering Tolerance is greater than the Process Spread so the process has the “potential” to be capable (depending on process centering). If, however, the Process Spread is greater than the Engineering tolerance, then the process variation will not “fit” within the tolerance and the process will not be capable (even if the process is centered appropriately). 18 Relating Cp And Cpk If Cp == Cpk, then the process is perfectly centered. If perfectly centered, Cp == Cpk. Because Cpk accounts for centering (where Cp does not), Cpk can never be larger than Cp. Both assume a stable process. Process Capability Analysis (6)
- 19. Open the excel sheet CIwebinar.xlsx Choose Stat > Quality Tools > Capability Analysis > Normal In Single column, enter ST deg In Subgroup size, enter 1. In enter Upper Specification Limit, enter value. In enter Lower Specification Limit, enter value. Click OK 19Capability Analysis on MINITAB
- 20. Capability Analysis on MINITAB 20 Goto Assistant Click Capability Analysis Follow the process questions and choose Capability Analysis. In Single column, enter ST deg In Subgroup size, enter 1. In enter Upper Specification Limit, enter value. In enter Lower Specification Limit, enter value. Click OK
- 21. Process Capability Analysis 21
- 22. Process Capability Analysis 22
- 23. Monitoring process outputs (dependent variables) 23
- 24. Pareto Diagram • Pareto diagram is a very powerful tool to stratify data • It’s a bar graph used to make visible the vital few “issues” versus the trivial ones (many) • It can be applied for improvements in all aspects • Pareto diagrams show whether the improvement actions produced expected results • Often used both in the PLAN and CHECK phase of a PDCA cycle (before and after the implementation) 1 Tool Description Method of use Remark Pareto Diagrams A diagram on which undesirable events or costs associated with items such as quality (e.g. number of defects or non- conforming products), productivity, cost, safety and so on are stratified according to their causes or manifestations and plotted in order of importance There may be a large number of undesirable phenomena or causes of trouble. The Pareto diagram makes it easy to see which of these have the most serious effect on quality, productivity, cost, safety, etc., together with their relative proportions Plot the “other” category at the far right of the diagram and ensure that it is not too large 24
- 25. Building A Pareto Chart1 Machine Material Man Method 42.8% 28.5% 16.2% 12.5% 4M DISTRIBUTION 25 • Open the excel sheet Pareto CIwebinar.xlsx • Choose Stat > Quality Tools > Pareto Chart • In Defects or Attribute Data in, • Enter Defects • In Frequencies in, enter Counts. Click OK
- 26. Cause & Effect Diagram • Cause-effect diagram is a powerful tool to identify and visualize potential causes of a problem (defect) • It helps in picking up and arranging all possible causes without any omissions: – 4M (Material, Machine, Man, Method) – 5M = 4M + Measurement – 5M + E (Environment) • It’s extensively used during root causes analysis in the PLAN phase of a PDCA cycle 2 Tool Description Method of use Remark Cause-and- effect diagram A diagram shaped like the bones of a fish for systematically summarizing the relationships between quality characteristics, defects (effects) and their causes Useful for searching out the factors that affect the characteristics, sorting out the relationships between these factors (causes) and the characteristics (the results), and depicting these systematically Gather the opinions of as many people as possible in order to flush out all the relevant factors 26
- 27. Building A Cause & Effect Diagram • Open the excel sheet CIwebinar.xlsx • Choose Stat > Quality Tools > Cause-and-Effect • In Causes column fill in Man…….Measurement, • Enter Title and Effects • Click OK 2 Dispersion occurs during the production process, so go through the steps in the manufacturing process one by one to seek the causes. 27
- 28. Graphs Graph and charts are extremely useful to make data visual They allow a quicker understanding of the data (especially in terms of trends, for examples when we analyze quality issues over the time) Different types of graphs and charts can be used depending on the needs and the purpose of the analysis Very powerful to verify the results of any improvement implementation (in the CHECK phase of any PDCA cycle) 3 Tool Description Method of use Remark Graphs and charts Diagrams for plotting data and showing temporal changes, statistical breakdowns and relationships between different quantities. Used for organizing data. Use line graphs for showing time trends, bar graphs for comparing quantities and pie charts for showing relative proportions. Use solid lines, dotted lines, circles and crosses skillfully for clarity. 28
- 29. Choosing a Graph3 29 Goto Assistant Click Graphical Analysis Go through the objectives and click “help me choose” based on data given.
- 30. Choosing a Graph3 30
- 31. Building A Box Plot on MINITAB Open the excel sheet CI webinar.xlsx Copy production data from excel sheet into MINITAB Choose Graph > Boxplot Under One Y, choose With Groups. Click OK In Graph Variables, enter Cases/Shift In Categorical variables for grouping (1-4, outermost first), enter Team Click Scale and click Reference Lines In Show reference lines at Y (value scale) positions, enter 40 Click OK 31 3
- 32. Check Sheet Check-sheets are very useful to simplify and speed up data collection on the shop floor The document is typically a blank form that is designed for the quick, easy and efficient recording of the desired information, which can be either quantitative or qualitative Data collection is done by simply using pencil and paper After collection, data should be analyzed with statistic methods 4 Tool Description Method of use Remark Check-sheets Forms specially prepared to enable data to be collected simply by making check marks. Used for tallying the occurrences of the defects or causes being addressed and graphing or charting them directly Clarify the objective and design a check- sheet to suit it. 32
- 33. How to Use A Check Sheet In the check sheet always report the relevant information, such as: Data source Objective of the data collection Characteristics of the data (e.g. Continuous/discrete) Date and timing (hours/shifts) of the data collection Location (production line/machine) Tool and the method used Person who collected the data 4 33
- 34. Histogram Histogram is a type of graph with a wide range of applications in statistics Histograms allow a visual interpretation of data by indicating the number of data points that lie within a range of values (so called class or bin). The frequency of the data that falls in each class is depicted by the use of a bar Generally used to help in: Eliminating defects and improving quality of the product Verifying if the production process is within the specifications Studying abnormalities / process deviation Searching for causes of variations in the production process 5 Tool Description Method of use Remark Histograms Prepared by dividing the data range into subgroups and counting the number of points in each subgroup. The number of points (the frequency) is then plotted as a height on the diagram. Prepare separate, stratified histograms for each of the 4Ms and examine the relationships between the shapes of the distributions and the specifications Use at least 30 values, preferably around 100 34
- 35. Building A Histogram • Open the excel sheet CIwebinar.xlsx • Choose Graph > Histogram> Simple/With group • Click OK • In the Graph Variables: Insert Cases/Shift, • In the Category variables for grouping: Insert Teams • Click OK 355
- 36. Scatter Plot A scatter diagram is a graphical tool that shows whether or not there is a correlation between two variables The pattern of the plotted data, as well as some calculated statistics, can expose possible cause and effect relationships 6 Tool Description Method of use Remark Scatter plot Prepared by plotting paired sets of data such as hardness and tensile strength, temperature and yield, porosity and insulation resistance, etc. against each other on x and y axes Collect paired sets of data on causes and effects, and use scatter diagrams to check for correlation between the sets of data Use at least 30 values, 50 if possible 36
- 37. Scatter Plot A correlation coefficient (rxy) can be calculated to evaluate the correlation between two variables: if two random variables are 100% correlated, the correlation coefficient would be +1 or -1. If there is no correlation, the correlation coefficient would be zero (0) 6 Positive correlation Negative correlation rxy = 1 rxy = -1 An increase in y depends on increase in x. If x is controlled, y will be naturally controlled. An increase in x will causes a tendency for decrease in y. 37
- 38. Scatter Plot Open the excel sheet CIwebinar.xlsx Choose Graph > Scatterplot Choose Simple, then click OK Under Y variables, enter Job Knowledge Test Score Under X variables, enter Tenure Click OK 38
- 39. 39Scatter Plot 6
- 40. Control Chart A control chart is a simple statistical tool that graphically represents data from a process It shows how a process is performing over the time, giving early warning that process may be going “out-of-control” and could cause defective or out of specification products Used to monitor and control production process parameters like temperature or mixing speed, as well as product quality parameters as viscosity or pH Control charts can also be used to measure the impact of the improvements 7 Tool Description Method of use Remark Control charts Prepared by plotting time along the horizontal axis and a characteristic value on the vertical axis. Unlike line graphs, they also show the control limit lines Use to check whether there are too many chronic defects, too much variation, values lying outside the control limits, or undesirable trends or cycles. Control charts show whether or not a process is in control Think about the best method of stratification and pay close attention to subgrouping 40
- 41. Control Chart Control charts allow to visualize and analyze data in a dynamic way So we can see the evolution of the characteristic under analysis over the time and react immediately 7 Central line Warning limits Control limits Samples To guarantee product quality, we need to move from a product control approach (inspection to defect detection) towards a process control approach (inspection to defect prevention). Control charts are the right tool to achieve it 41
- 42. Control Chart7 42 TYPE OF CONTROL CHART TYPE OF DATA CONTROL CHART USED CONTROL CHART VARIABLE The characteristic of a product is represented in a continuous way CONTINUOUS X - R Examples • Measures (0,01 mm) • Volume (m3) • Weight (g) • Energy consumption (kWh) CONTROL CHART ATTRIBUTE The characteristic of a product cannot be measured, therefore every product is evaluated depending if it has or not some specific attributes DISCRETE PExamples: • Number of defected parts • % of defects
- 43. 43Control Chart7 Goto Assistant Click Graphical Analysis Go through the objectives and click “help me choose” middle tab. Then decide if continuous(variable data) or Attributes
- 44. Control Chart X-R An X-R control chart is one that shows both the mean value, X , and the range, R This is the most common type of control chart using indiscrete values The X portion of the chart mainly shows any changes in the mean value of the process, while the R portion shows any changes in the dispersion of the process This chart is particularly useful because it shows changes in mean value and dispersion of the process at the same time, making it a very effective method for checking abnormalities in the process 7 44
- 45. Control Chart X-R7 Anomalies 1 point outside the control limit (upper or lower) 2 points in sequence in between control and warning limits 8 points (or more) in a row from one side of the mean/goal value 45
- 46. Control Chart X-R7 Run (or series) happens when points are always from one side of the central line. The number of points on one side of the central line is called “length of the run” If a series is 8 points, it means that there is an anomaly in the process. Trend happens when a group of points are following a trend (up or down). To evaluate if there is a trend, consider 7 points in a sequence. Often points get out of the limit before the 7° point. Periodicity happens when points show the same level of variation at the same interval. The evaluation is much more complex. The suggestion is to follow the points movement closely and make a technical decision 46
- 47. How To Build A Run Chart Open the worksheet CI.webinar and copy the ST degC data into Minitab Choose Stat > Quality Tools > Run Chart In Single column,enter ST deg In Subgroup size, enter 1. Click OK 47 It helps you check whether the process is stable, you be able to understand the different types of variation 1. Same value plot 2. Clustering plot 3. Oscillation plot 4. Trend plot
- 48. How To Build An I-MR CHART Open the worksheet CI.webinar and copy the ST degC data into Minitab Choose Stat > Control Charts > Variables Charts for Individuals > I-MR In Variables, enter Accuracy Score Click I-MR Options, then click the Teststab Choose Perform all tests for special causes, then click OK in each dialog box 48
- 49. How To Build An X bar R Chart Open the worksheet CI.webinar and copy the BD data into Minitab Choose Stat > Control Charts > Variables Charts for Subgroups > Xbar-R Choose All observations for a chart are in one column, then enter BD In Subgroup sizes, enter Reading. Click OK 49
- 50. How To Build An X bar S Chart Open the worksheet CI.webinar and copy the BD data into Minitab Choose Stat > Control Charts > Variables Charts for Subgroups > Xbar-S Choose All observations for a chart are in one column, then enter BD In Subgroup sizes, enter Reading. Click OK 50
- 51. Control Chart X-R7 BEFORE Low weight Case study 51
- 52. Control Chart P Control chart type “p” are used when the characteristic of a product cannot be measured, therefore every product is evaluated depending if it has or not some specific attributes HOW TO BUILD A CONTROL CHART P Gather all the data, select the size, the frequency and number of samples. Data collection time frame should be long enough to track all the possible sources of variation. Divide data in subgroups (n). Usually n should be more than 20 and average value of defect products for every subgroup should be around 3 – 4 Calculate the percentage of defects for each subgroup Set the scale for the graph: after p values have been calculated for all the subgroups, make the scale from 0 to the approximately twice the largest p value. In this way all the value should easily fit into the graph 8 p = np/n p = % of defect products n = dimension of the subgroup np = number of defects in one subgroup 52
- 53. Control Chart P8 Calculate the p value (p) and control limits Plot all the p values and connect the dots with straight line ҧ𝑝 = ∑pn / ∑n Example 53
- 54. Control Chart P Open the excel sheet CI.webinar.xlx > Diaper Production copy the data to Minitab Choose Stat > Control Charts > Attributes Chart > P In Variables, enter Defectives Found In Subgroup sizes, enter No of Diapers Monitored. Click OK 548
- 55. Graphical Analysis on MINITAB Open the worksheet CI.webinar and copy the Viscosity data into Minitab Goto Assistant Click Graphical Analysis Go through the objectives and click “help me choose” middle tab Choose Data with no subgroup In the Y column choose “Viscosity” Click OK 558
- 56. Graphical Analysis on MINITAB 568
- 57. Thank You! Nwagbaraocha, D. E. 57