You are on page 1of 15

PRODUCTIVITY & QUALITY

7 Basic Tools for Quality Control


Within any manufacturing environment, product information or "data" is collected for a variety of reasons. Some of the common reasons for having this data are: 1. 2. 3. 4. Data to assist in understanding the actual process or situation. Data for product analysis. Data for process control (SPC). Data for regulation a basis for raising or lowering a data standard, for example, temperature or thickness. 5. Acceptance or rejection data used to approve or reject products or parts. The purpose of collecting data is usually to gather information about the product or to follow up with some form of action. That is, after evaluating the actual conditions revealed by the data, some form of proper action should be taken. The first major critical step, however, is to ensure that the data represents typical conditions, or is data taken from normal circumstances. The second major critical step is to have a purpose for collecting the data. Therefore, before we actually collect data, we should ask the following questions: 1. Define What we are measuring and Why are we measuring this information? 2. Define Where and When should we measure this information? 3. Define How should we be measuring this information and at what time intervals? What measurement tool? 4. Define Who should be measuring this information? Data can be collected in many ways, depending upon the reason for the data, and the type of information we are seeking. Thus, data can be basically divided into two main groups: 1. Measurement data: continuous data of length, weight, time, torque, etc. 2. Countable data: enumerate data such as the number of defectives, percentage defective, number count of each defect, etc. Once we collect this data, it should be analyzed, and the information extracted through the use of statistical methods. For that reason, data should be collected and organized in such a way as to make data analysis more simple and meaningful. Therefore, you need to clearly record the nature of the data collected. You should also record the purpose of the measurements and their characteristics; the date; the instrument or method of measuring; the person performing the measurement; and any other pertinent information to the collection process. To properly record this data, you need to have a consistent time period, for example, measured every hour or every two hours, and make sure you are measuring production parts. In the case of collecting data to count defects, ensure you count each and every item produced, as well as the defects, during the collection period so you can compare how many defects were produced in relation to the total production of parts. Let us now summarize our data collection steps: 1. Clarify the purpose of collecting the data it is useless if there is no real reason to collect the data.
BMS/Quality & Productivity

PRODUCTIVITY & QUALITY 2. Collect data efficiently make the method reliable, consistent and specify a standard period of time. 3. Take action according to the data once you have collected the data, make it effective by analyzing the data and using it for improvement. Have an improvement action as a result of the collection process. 4. When establishing a basis for collecting data, be sure to ask and answer the What, When, Where, How & Who questions mentioned above. There are a large variety of Quality Tools and Statistical Process Control Methods (SPC) within the realm of Total Quality Management. We are, however, going to only concern ourselves with 7 Basic Quality Tools within this web site. They are: 1. 2. 3. 4. 5. 6. 7. Check Sheets Pareto Diagrams Histogram Diagram Cause-and-Effect or "Fishbone" Diagram Scatter Diagrams Control Charts NP Charts

Tool #1 - The Check Sheet


As previously mentioned, the intent and purpose of collecting data is to either control the production process, to see the relationship between cause-and-effect, or for the continuous improvement of those processes that produce any type of defect or nonconforming product. A Check Sheet is used for the purpose of collecting data to compile in such a way as to be easily used, understood and analyzed automatically. The Check Sheet, as it is being completed, actually becomes a graphical representation of the data you are collecting, thus you do NOT need any computer software, or spreadsheet to record the data. It can be simply done with pencil and paper! Check sheets have the following main functions: 1. 2. 3. 4. 5. Production process distribution checks - where the distribution lies. Defective item checks - to determine what kind of defects exist in the process. Defect location checks - to determine where the common defects on a part are located. Defective cause checks - type of defect and thus validate the cause thereof. Check-up confirmation checks - final phase of assembly to check the finished product or work.

The methods that we will concentrate on and utilize here will be for Production Process distribution and defective item checks. We will discuss the use and relevance of each individually. Production Process Distribution The size, weight, or diameter of parts, for example, are known as "continuous data". In a process where these types of data are gathered, the distribution they provide will often resemble a Histogram (Histogram is Tool #3). A histogram can be used to investigate the distribution of the process characteristics, and the average value can be calculated. Below is a sample Production Check Sheet
BMS/Quality & Productivity

PRODUCTIVITY & QUALITY

In this sample sheet, we measured torque readings. The spec limit is 2.2Nm .5. From the example sheet, you should see two dark vertical lines, on the left side it is labeled LSL (Lower Spec Limit) which indicates the 2.2 - .5, or 1.7. On the right side, another dark line marked USL (Upper Spec Limit) which indicates the 2.2 + .5, or 2.7. All product readings, or torque readings in this example, that conform (actually good product), need to fall within these limit boundaries. Anything that is measured outside these limits is termed "Non-Conforming" since they are not within proper specification limits. Every time a measurement was taken, an "X", or check mark, was made on the check sheet. From this sample sheet, you can see where most of the torque readings lie, the consistency of the distribution, and how many are actually outside the spec limit. What has also been created here on this Check Sheet is a "Histogram". We will discuss Histograms in a later lesson. Right now, what is important to note from this example chart is that we do not have a good stable process. The distribution is widespread and not well centered between the specification limits. The distribution has dual peaks, or is what is known as "bi-modal". Bi-modal means that there are two points where most of the readings taken are charted, or that is has two peaks. This means the frequency of readings rises and falls twice, rather than a more proper and even distribution with one peak.

BMS/Quality & Productivity

PRODUCTIVITY & QUALITY


(continuous data use) No.___________ 741

PRODUCTION CHECK SHEET


Product Name_________________________________ Alternator Pulley Usage________________________________________ Pulley Bolt Torque Specification__________________________________ 2.2 +/- .5 No. of Inspections______________________________ 185 Total Number__________________________________ 185 Lot Number___________________________________ 1631 Dimensions 1.5 1.6 1.7 1.8 1.9 2.0 2.1 2.2 2.3 2.4 Date_________________________________ 12- 02- 02 Factory_______________________________ Church Street Section Name__________________________ SI Line Data Collector__________________________ Sam The Man Group Name___________________________ Remarks:_____________________________ 2.5 2.6 2.7 2.8 2.9 3.0 3.1 3.2

40
SpecUSL SpecLSL

35

30

25

20

XXXXX XXXXX XXXXX XXXXX

XXXXX XXXXX XXXXX XXXX

XXXXX XXXXX XXXXX XX

15

X XXXXX XXXXX XXXXX XXXXX XXXXX XXX

X XXXXX XXXXX XXXXX XXXXX XXXXX XX

XXXXX XXXXX XXXXX XX

10

XXXXX XXXXX XXX

XXXXX XXXXX

XXXXX XXX

XXXXX XX

X XXXXX XXXXX XX
2

0
TOTAL FREQUENCY

X
1

XX

X
1

13

10

16

19

17

12

16

20

17

13

A Check Sheet is used for: 1. distinguishing between fact and opinion (example: how does the community perceive the effectiveness of the school in preparing students for the world of work?) 2. gathering data about how often a problem is occurring (example: how often are students missing classes?) 3. gathering data about the type of problem occurring (example: What is the most common type of word processing error created by the students-grammar, punctuation, transposing letters, etc.?)

BMS/Quality & Productivity

PRODUCTIVITY & QUALITY Steps to create a Check Sheet 1. Clarify the measurement objectives. Ask questions such as "What is the problem?", "Why should data be collected?", "Who will use the information being collected?", "Who will collect the data?" 2. Create a form for collecting data. Determine the specific things that will be measured and write this down the left side of the check sheet. Determine the time or place being measured and white this across the top of the columns. 3. Collect the data for the items being measured. Record each occurrence directly on the Check Sheet as it happens.

Tool #2 - The Pareto Diagram


The Pareto diagram is a graphical overview of the process problems, in ranking order of the most frequent, down to the least frequent, in descending order from left to right. Thus, the Pareto diagram illustrates the frequency of fault types. Using a Pareto, you can decide which fault is the most serious or most frequent offender. The basic underlying rule behind Pareto's law is that in almost every case, 80% of the total problems incurred are caused by 20% of the problem cause types; such as people, machines, parts, processes, and other factors related to the production of the product. Therefore, by concentrating on the major problems first, you can eliminate the majority of your problems. The few items that have the largest amount of occurrence is your more frequent problem, than are the many items that only happen once in a while. This is called the "vital few over the trivial many" rule. Quite often, once you cure several of the "big hitters" you also eliminate some of the smaller problems at the same time. So then, what exactly is a Pareto diagram? The Pareto prioritizes problem areas. Sometimes a quality problem is so cluttered with so many smaller problems, it is difficult to know just where to begin the solving process. Let's take an example. Below is a table from a manufacturing process that charted all of their quality problems. While the original defect chart listed many problems at various stages of the process, the overall problems were grouped into five main process areas. In the left column is the name of the process where the defects occur. In the next column is the amount of defects recorded from their daily check sheets, recorded during a one week period. In the third column is the percent of defectives from the overall production (N = 2165). In the fourth and final column, is the percent of the total defects. That is, for example, of all the defects recorded (416), poor Caulking is 47.6% of the entire problem. It should be obvious then, where the primary problem is and what should be focused upon first.

BMS/Quality & Productivity

PRODUCTIVITY & QUALITY

Figure 1 From the chart above (figure 1), you can now create a Pareto chart in which you can graphically display the quality problems. There is special software on the market that makes Pareto diagrams, however, an Excel barchart will basically create the same display. The below bar chart reflects the above information charted in Excel.

Pareto Example

The left vertical axis (border) shows the number of defects for each defective category, and the right vertical axis shows the percentage of each defect of the total defects. The horizontal axis (bottom) lists the defective items starting with the most frequent one on the left (Caulking), progressing over to the least frequent occurrence on the right side (Torque). Therefore, the Pareto diagram visually indicates which problem should be solved first, or in this case, the Caulking problem. With this bar graph, it is easier to see which defects are most important of all the defects that exist. If we solve all or most of the problems in Caulking, it could affect some of the problems observed in connecting, gapping, fitting, and torque. During the "brain-storming" session (we'll cover this later), it is wise to ask, "Does the Caulking problem have any impact on the other problems listed?" In some cases it might. If there was proper caulking, would part of the "Gapping" problem be eliminated?" If there were proper caulking, would the "Torque" have a better value and thus not be part of the defects? Sometimes your major problems have impact on the smaller problems. Several problem areas may all be attributed to ONE ROOT CAUSE, even though several failure modes are observed. For this reason, it is always wise to choose the most frequent problem first. STEPS IN CONSTRUCTING A PARETO CHART WITH STEP-BY-STEP EXAMPLE: 1. Determine the categories of problems or causes to be compared. Begin by organizing the problems or causes into a narrowed down list of categories (usually 8 or less). 2. Select a Standard Unit of Measurement and the Time Period to be studied. It could be a measure of how often something occurs (defects, errors, tardies, cost overruns, etc.);
BMS/Quality & Productivity

PRODUCTIVITY & QUALITY frequencies of reasons cited in surveys as the cause of a certain problem; or a specific measurement of volume or size. The time period to be studied should be a reasonable length of time to collect the data. 3. Collect and Summarize the Data. Create a three-column table with the headings of "error or problem category", "frequency", and "percent of total". In the "error or problem category" column list the categories of problems or causes previously identified. In the "frequency" column write in the totals for each of the categories over the designated period of time. In the "percent of total" column, divide each number in the "frequency" column by the total number of measurements. This will provide the percentage of the total. Error Category Punctuation Grammar Spelling Typing TOTAL Frequency Percent of Total 22 44% 15 30% 10 20% 3 6% 50 100%

1. Create the framework for the horizontal and vertical axes of the Pareto Chart. The horizontal axis will be the categories of problems or causes in descending order with the most frequently occurring category on the far left (or at the beginning of the horizontal line). There will be two vertical axes-one on the far left and one on the far right. The vertical axis on the far left point will indicate the frequency for each of the categories. Scale it so the value at the top of the axis is slightly higher than the highest frequency number. The vertical axis on the far right will represent the percentage scale and should be scaled so that the point for the number of occurrences on the left matches with the corresponding percentage on the right. 2. Plot the bars on the Pareto Chart. Using a bar graph format, draw the corresponding bars in decreasing height from left to right using the frequency scale on the left vertical axis. To plot the cumulative percentage line, place a dot above each bar at a height corresponding to the scale on the right vertical axis. Then connect these dots from left to right, ending with the 100% point at the top of the right vertical axis. 3. Interpret the Pareto Chart. Use common sensejust because a certain problem occurs most often doesn't necessarily mean it demands your greatest attention. Investigate all angles to help solve the problems-What makes the biggest
BMS/Quality & Productivity

PRODUCTIVITY & QUALITY difference? What will it cost to correct the problems? What will it cost if we don't correct this problem?

A PARETO CHART IS USED FOR:


1. Focusing on critical issues by ranking them in terms of importance and frequency (example: Which course causes the most difficulty for students?; which problem with Product X is most significant to our customers?) 2. Prioritizing problems or causes to efficiently initiate problem solving (example: Which discipline problems should be tackled first? or, What is the most frequent complaint by parents regarding the school?; solution of what production problem will improve quality most?) 3. Analyzing problems or causes by different groupings of data (e.g., by program, by teacher, by school building; by machine, by team) 4. Analyzing the before and after impact of changes made in a process (example: What is the most common complaint of parents before and after the new principal was hired?; has the initiation of a quality improvement program reduced the number of defectives?)

Tool #3 - The Histogram


The common person believes that if a part is made in mass production from a machine, all of the parts will be exactly alike. The truth is that even with the best of machines and processes, no two parts are exactly the same. The product will have a main or "mean" specification limit, with plus/minus tolerance that states that as long as the part is produced within this range, to that range, it is an acceptable part. The object is to hit the target specification, however, that is not always totally possible. The purpose of a Histogram is to take the data that is collected from a process and then display it graphically to view how the distribution of the data, centers itself around the mean, or main specification. From the data, the histogram will graphically show: 1. The center of the data. 2. The spread of the data. 3. Any data skewness (slant, bias or run at an angle). 4. The presence of outliers (product outside the specification range). 5. The presence of multiple modes (or peaks) within the data.

A HISTOGRAM IS USED FOR:


1. Making decisions about a process, product, or procedure that could be improved after examining the variation (example: Should the school invest in a computer-based tutoring program for low achieving students in Algebra I after examining the grade distribution?; are more shafts being produced out of specification that are too big rather than too small?) 2. Displaying easily the variation in the process (example: Which units are causing the most difficulty for students?; is the variation in a process due to parts that are too long or parts that are too short?)

STEPS IN CONSTRUCTING A HISTOGRAM:


1. Gather and tabulate data on a process, product, or procedure. This could be time, weight, size, frequency of occurrences, test scores, GPA's, pass/fail rates, number of days to complete a cycle, diameter of shafts built, etc.
BMS/Quality & Productivity

PRODUCTIVITY & QUALITY 2. Calculate the range of the data by subtracting the smallest number in the data set from the largest. Call this value R. 3. Decide about how many bars (or classes) you want to display in your eventual histogram. Call this number K. This number should never be less than four and seldom exceeds 12. With 100 numbers, K=7 generally works well. With 1000 pieces of data, K=11 works well. 4. Determine the fixed width of each class by dividing the range, R by the number of classes K. This value should be rounded to a "nice" number, generally a number ending in a zero. For example 11.3 would not be a "nice" number. 10 would be considered a "nice" number. Call this number i, for interval width. It is important to use "nice" numbers else the histogram created will have wierd scales on the X axis. 5. Create a table of upper and lower class limits. Add the interval width i to the first "nice" number less than the lowest value in the data set to determine the upper limit of the first class. This first "nice" number becomes the lowest lower limit of the first class. The upper limit of the first class becomes the lower limit of the second class. Adding the internal width (i) to the lower limit of the second class determines the upper limit for the second class. Repeat this process until the largest upper limit exceeds the biggest piece of data. You should have appriximately K classes or categories in total. 6. Sort, organize, or categorize the data in such a way that you can count or tabulate how many pieces of data fall into each of the classes or categories in your table above. These are the frequency counts and will be plotted on the Y axis of the histogram. 7. Create the framework for the horizontal and vertical axes of the histogram. On the horizontal axis plot the lower and upper limits of each class determined above. The scale on the vertical axis should run from zero to the first "nice" number greater than the largest frequency count determined above. 8. Plot the frequency data on the histogram framework by drawing vertical bars for each class. The height of each bar represents the number or frequency of values occuring between the lower and upper limits of that class. 9. Interpret the histogram for skew and clustering problems:

Interpreting skew problems:


Data may be skewed to the left or right. If the histogram shows a long tail of data on the left side of the histogram, the data is termed left or negatively skew. If a tail appears on the right side, the data is termed right or positively skew. Most process data should not typically appear skew. Data that is seriously skew either to the left or right may be an indication that there are inconsistencies in the process or procedures, etc. Decisions may need to be made to determine the appropriateness of the direction of the skew. It should be noted, however, that some process data is, by its very nature, skew. This situation occurs in arrival processes (for example, people arriving at a McDonalds within a fixed unit of time) and in service processes (for example, the time it takes to wait on a customer in a bank).

Interpreting clustering problems:


Data may be clustered on opposite ends of the scale or display two or more peaks indicating serious inconsistencies in the process or procedure or the measurement of a mixture of two or more distinct groups or processes that behave very differently.

BMS/Quality & Productivity

PRODUCTIVITY & QUALITY

EXAMPLE
The data below are the spelling test scores for 20 students on a 50 word spelling test. The scores (number correct) are: 48, 49, 50, 46, 47, 47, 35, 38, 40, 42, 45, 47, 48, 44, 43, 46, 45, 42, 43, 47. The largest number is 50 and the smallest is 35. Thus, the range, R = 15. We will use 5 classes, so K=5. The interval width i= R/K = 15/5=3. The we will make our lowest lower limit, the lower limit for the first class 35. Thus the first upper limit is 35+3 or 38. The second class will have a lower limit of 38 and an upper limit of 41. The completed table (with frequencies tabulated) will look like the following:
Class Lower Limit Upper Limit 1 35 38 2 38 41 3 41 44 4 44 47 5 47 50 Frequency 1 2 4 5 8

Tool #4 - Cause-and-Effect Diagram


After collecting data from a process, and then preparing a pareto or histogram diagram, it's time to consider the reasons for the variation and those defects created. This data collected will reveal that items produced do not always turn out the same on a consistent basis. That is, parts produced can vary from production line to production line, from day shift to night shift, and from day to day, and so forth. In other words, you seldom get consistent parts produced every time. What causes these differences, or variation within the process? Basically, the variation created can originate from one or more of the following sources: 1. Raw Materials 2. Machinery, equipment or tooling 3. Work method or process 4. Work force - new people, trained different, etc. 5. Measurement method, or inconsistency in ways of measurement 6. Environment - high humidity, cold temperatures, dust, etc. The real problem becomes which one of the above factors is either totally, mostly, or somewhat responsible for the cause of our problem? Or is it a combination of several causes? A Cause-and-Effect diagram is useful in sorting out the causes of dispersion and organizing mutual relationships. This is an excellent team problem solving tool, where a team can gather together to "brain storm" the potential causes and resolutions to solve the variation problem. The Cause-and-Effect Diagram was created by Dr. Kaoru Ishikawa, an engineer and professor in Japan. The Cause-and-Effect Diagram is also referred to as a "Fishbone" diagram, getting the name from its resemblance to a fish skeleton when created. The main purpose of this diagram is to define a problem, identify a possible cause, isolate the cause, and then develop a solution. Below is an example of a generic Cause-and-Effect Diagram. A CAUSE AND EFFECT DIAGRAM IS USED FOR:

BMS/Quality & Productivity

PRODUCTIVITY & QUALITY 1. Identifying potential causes of a problem or issue in an orderly way (example: Why has membership in the band decreased?; why isn't the phone being answered on time?; why is the production process suddenly producing so many defects?) 2. Summarizing major causes under four categories (e.g., People, Machines, Methods, and Materials or Policies, Procedures, People, and Plant)

STEPS IN CONSTRUCTING A CAUSE AND EFFECT DIAGRAM:


1. Write the issue (problem or process condition) on the right side of the Cause and Effect Diagram. 2. Identify the major cause categories and write them in the four boxes on the Cause and Effect Diagram. You may summarize causes under categories such as: 3. Methods, Machines, Materials, People 4. Places, Procedures, People, Policies, 5. -Surroundings, Suppliers, System, Skills 6. Brainstorm potential causes of the problem. As possible causes are provided, decide as a group where to place them on the Cause and Effect Diagram. It is acceptable to list a possible cause under more than one major cause category. 7. Review each major cause category. Circle the most likely causes on the diagram. 8. Review the causes that are circled and ask "Why is this a cause?" Asking "why" will help get to the root cause of the problem. 9. Reach an agreement on the most probable cause(s).

EXAMPLE OF COMPLETED CAUSE/EFFECT DIAGRAM:

BMS/Quality & Productivity

PRODUCTIVITY & QUALITY

Tool #5 - The Scatter Diagram


The Scatter Diagram is another Quality Tool that can be used to show the relationship between "paired data", and can provide more useful information about a production process. What is meant by "paired data"? The term "cause-and-effect" relationship between two kinds of data may also refer to a relationship between one cause and another, or between one cause and several others. For example, you could consider the relationship between an ingredient and the product hardness; between the cutting speed of a blade and the variations observed in length of parts; or the relationship between the illumination levels on the production floor and the mistakes made in quality inspection of product produced.

A SCATTER DIAGRAM IS USED FOR:


1. Validating "hunches" about a cause-and-effect relationship between types of variables (examples: I wonder if students who spend more time watching TV have higher or lower average GPA's?; is there a relationship between the production speed of an operator and the number of defective parts made?; is there a relationship between typing speed in WPM and errors made?) 2. Displaying the direction of the relationship (positive, negative, etc.) (examples: Will test scores increase or decrease if the students spend more time in study hall?; will increasing assembly line speed increase or decrease the number of defective parts made?; do faster typists make more or fewer typing errors?) 3. Displaying the strength of the relationship (examples: How strong is the relationship between measured IQ and grades earned in Chemistry?; how strong is the relationship between assembly line speed and the number of defective parts produced?; how strong is the relationship between typing faster and the number of typing errors made?)

STEPS IN CONSTRUCTING A SCATTER DIAGRAM:


1. Collect two pieces of data (a pair of numbers) on a student, process, or product. Create a summary table of the data. 2. Draw a diagram labeling the horizontal and vertical axes. It is common that the "cause" variable be labeled the horizontal (X) axis and the "effect" variable be labeled the vertical (Y) axis. The values should increase up the vertical scale and toward the right on the horizontal scale. The scale on both the X and Y axes should be sufficient to include both the largest and the smallest X and Y values in the table. 3. Plot the data pairs on the diagram by placing a dot at the intersections of the X and Y coordinates for each data pair. 4. Interpret the scatter diagram for direction and strength a. Interpreting the direction: Data patterns may be positive, negative, or display no relationship. A positive relationship is indicated by an ellipse of points that slopes upward demonstrating that an increase in the cause variable also increases the effect variable. A negative relationship is indicated by an ellipse of points that slopes downward demonstrating that an increase in the cause variable results in a decrease in the effect variable. A diagram with a cluster of points such that it is difficult or impossible to determine whether the trend is upward sloping or downward sloping indicates that there is no relationship between the two variables.
BMS/Quality & Productivity

PRODUCTIVITY & QUALITY b. Interpreting the strength: Data patterns, whether in a positive or negative direction, should also be interpreted for strength by examining the "tightness" of the clustered points. The more the points are clustered to look like a straight line the stronger the relationship.

EXAMPLES

BMS/Quality & Productivity

PRODUCTIVITY & QUALITY

Tool #6 - The Control Chart


Processes and Process Variability
The concept of process variability forms the heart of statistical process control. For example, if a basketball player shot free throws in practice, and the player shot 100 free throws every day, the player would not get exactly the same number of baskets each day. Some days the player would get 84 of 100, some days 67 of 100, some days 77 of 100, and so on. All processes have this kind of variation or variability. This process variation can be partitioned into two components. Natural process variation, frequently called common cause or system variation, is the naturally occurring fluctuation or variation inherent in all processes. In the case of the basketball player, this variation would fluctuate around the player's long-run percentage of free throws made. Special cause variation is typically caused by some problem or extraordinary occurrence in the system. In the case of the basketball player, a hand injury might cause the player to miss a larger than usual number of free throws on a particular day.

Statistical Process Control


Shewhart's discovery statistical process control or SPC, is a methodology for charting the process and quickly determining when a process is "out of control" (e.g., a special cause variation is present because something unusual is occurring in the process). The process is then investigated to determine the root cause of the "out of control" condition. When the root cause of the problem is determined, a strategy is identified to correct it. The investigation and subsequent correction strategy is frequently a team process and one or more of the TQM process improvement tools are used to identify the root cause. Hence, the emphasis on teamwork and training in process improvement methodology. It is management's responsibility to reduce common cause or system variation as well as special cause variation. This is done through process improvement techniques, investing in new technology, or reengineering the process to have fewer steps and therefore less variation. Management wants as little total variation in a process as possible--both common cause and special cause variation. Reduced variation makes the process more predictable with process output closer to the desired or nominal value. The desire for absolutely minimal variation mandates working toward the goal of reduced process variation. The process above is in apparent statistical control. Notice that all points lie within the upper control limits (UCL) and the lower control limits (LCL). This process exhibits only common cause variation. The process above is out of statistical control. Notice that a single point can be found outside the control limits (above them). This means that a source of special cause
BMS/Quality & Productivity

PRODUCTIVITY & QUALITY variation is present. The likelihood of this happening by chance is only about 1 in 1,000. This small probability means that when a point is found outside the control limits that it is very likely that a source of special cause variation is present and should be isolated and dealt with. Having a point outside the control limits is the most easily detectable out-of-control condition.

The graphic above illustrates the typical cycle in SPC. First, the process is highly variable and out of statistical control. Second, as special causes of variation are found, the process comes into statistical control. Finally, through process improvement, variation is reduced. This is seen from the narrowing of the control limits. Eliminating special cause variation keeps the process in control; process improvement reduces the process variation and moves the control limits in toward the centerline of the process.

Tool #7 - The Np Attribute Control Chart


The use of attribute control charts arises when items are compared with some standard and then are classified as to whether they meet that standard or not. The Np control chart is used to determine if the rate of nonconforming product is stable, and will detect when a deviation from stability has occurred. There are those who argue that there should only be an Upper Control Limit (UCL), and NOT a Lower Control Limit (LCL) since rates of nonconforming product outside the LCL is actually a good thing. However, if we treat the LCL violations as another search for an assignable cause, we could learn where lower nonconformity rates lie and perhaps eliminate them further.

BMS/Quality & Productivity