Adamson University Graduate School

MNE 624

Total Quality Management CHAPTER

ATTRIBUTE CONTROL CHARTS

Selections PART 1     PART 2 Introduction Types of Attribute Control Charts Classification Charts The p Chart for Constant Subgroup Sizes  The p Chart for Variable Subgroup Sizes  The np Chart  Count Charts PART 3  c Charts  u Charts  Limitations of Attribute Control Chart

Chapter Objectives

To discuss when to use the different types of attribute control chart To construct the different types of attribute Control Charts: p Chart, np Chart, c Chart and Chart To analyze and interpret attribute control charts To discuss the limitations of attribute control charts

PART 1
Introduction
ATTRIBUTE
 The term attribute, as used in quality, refers to those quality

characteristics that conform to specifications or do not conform to specifications.

Attributes are used:

 Where measurements are not possible - - for example, visually inspected items such as color, missing parts, scratches, and damage.  Where measurements can be made but are not made because of time, cost, or need.

Two basic types of attribute control charts:

1. Classification Charts 2. Count Charts Classification Charts  deal with either the fraction of items or the number of items in a series of subgroups that have a particular characteristics p Chart  used to control the fraction of items with the characteristics. np Chart  serves the same function as the p chart except that it is used to control the number rather than the fraction of items with the characteristics and is used only with constant subgroup sizes.

Count Charts  deal with the number of times a particular characteristic appears in so me given area of opportunity. c Chart  used to control the number of times a particular characteristic appears in a constant area of opportunity. Chart  serves the same basic function as a c chart, but is used when the area of opportunity changes from subgroup to subgroup.

Construction of a p Chart

Conditions for Use

 Each unit must be classifiable as either possessing or not possessing the characteristic of interest.  The probability that a unit possesses the characteristic of interest is assumed to be stable from unit to unit.  Within a given area of opportunity, the probability that a given unit possesses the characteristic of interest is assumed to be independent of whether any other unit possesses the characteristic.

The objectives of nonconforming charts are to:  Determine the average quality level.  Bring to the attention of management any changes in the average.  Improve the product quality.  Evaluate the quality performance of operating and management personnel.  Suggest place to use X and R charts.  Determine acceptance criteria of a product before shipment to the customer.
End of part 1

PART 2
p Chart for Constant Subgroup Size
 Constant subgroup size implies that the same number of items is sampled and then classified for each subgroup on the chart. We use discrete countable characteristic of output to construct a p Chart.
THE CENTER LINE AND CONTROL LIMIT
TOTAL NO.OF DEFECTIVE IN ALL SUBGROUPS UNDER INVESTIGATION CENTER LINE ( p ) = ( p ) = TOTAL NO.OF UNITS EXAMINED IN ALL SUBGROUPS UNDER INVESTIGATION UCL ( p ) = p + 3 p ( 1-p ) n UCL ( p ) = p - 3 p ( 1-p ) n

WHERE: p UCL ( p ) LCL ( p ) n

- Centerline - Upper Control Limit - Lower Control Limit - Constant Subgroup Size

Construction of a p Chart

Fig.7.3 Minitab p Chart for Fraction of Cracked Tiles

(p)=

183 3000

= 0.061 0.061 ( 1-0.061 ) 100 0.061 ( 1-0.061 ) 100 = 0.133

UCL ( p ) = 0.061 + 3

LCL ( p ) = 0.061 - 3

= -0.011

(p)=

2800

= 0.055 = 0.123 100

100

Fig.7.4 Minitab Revised p Chart for Fraction of Cracked Tiles

Iterative Re evaluations
It is possible that by changing the process, removing points that were out of control and recomputing the control limits and points that initially exhibited only common variation will now indicate lack of control. If and when this happens, the system must again re evaluate to eliminate the newly special cause of variation. Analysis of the process will continue to iterate in this manner until there are no longer appears to be lack of control. At some point a decision must be made to stop analysing the original data and collect new data. There is no explicit rule the point at which this should be done; only knowledge and experience with the process can dictate when to stop analysing previous data and begin collecting and analysing new data.

SUBGROUP SIZE
As general rule, Control Charts based on classification count data should have sizes large enough. Subgroup sizes should remain the same for all subgroups but occasionally circumstances require variation in subgroup size. Whether the subgroup size for p Chart varies or remain constant, the larger the subgroup size, the narrower the control limits will be.

p Chart for Variable Subgroup Size

(p)= 2569 6421 = 0.400 0.400 ( 1-0.400 ) 465 0.400 ( 1-0.400 ) 465 = 0.468 UCL ( p ) = 0.400 + 3

LCL ( p ) = 0.400 - 3

= 0.332

Fig. 7.6 Minitab p Chart for Vehicles with Exact Change

Fig. 7.5 Number of Vehicles using Exact Change, Transponders, or Tokens

Fig. 7.7 Minitab Revised Control Chart for Vehicles with Exact Change

np Chart
Traditionally np charts are use only when subgroup size are constant. Data collected for an np chart will be a series of integer, each representing the number of nonconforming (or conforming) item in its subgroup.
CENTER LINE ( np ) = ( np ) = SUBGROUP SIZE SUBGROUPS UNDER INVESTIGATION TOTAL NO.OF UNITS EXAMINED IN ALL SUBGROUPS UNDER INVESTIGATION

UCL ( np ) = np + 3 LCL ( np ) = np - 3

np ( 1- p ) np ( 1- p )

Fig. 7.8 Minitab p Chart for Cracked Tiles

End of part 2

PART 3
c Chart
 In industrial statistics, the c-chart is a type of control chart used to monitor "count"-type data, typically total number of nonconformities per unit. It is also occasionally used to monitor the total number of events occurring in a given unit of time. Examples of processes suitable for monitoring with a c-chart include:  Monitoring the number of voids per inspection unit in injection molding or casting processes  Monitoring the number of discrete components that must be re-soldered per printed circuit board  Monitoring the number of product returns per day
From Wikipedia

The number of events in an area of opportunity is denoted by c, the count for each area of opportunity. The sequence of successive c values, taken over time, is used to construct the control chart. The centerline for the chart is the average number of events observed. It is calculated as

The standard error is the square root of the mean. Adding and subtracting three times the standard error from the Centerline, c , yields the upper and lower control limits. Thus

Counts, Control Limits, and Zones

 As we have already seen with p charts and np charts, when a process is in a state of control, only very rarely will points fall beyond the control limits. Therefore, when a point does fall outside the control limits, we will consider it an indication of a lack of control and take appropriate action. When the lower limit is calculated to be negative, we will use 0 uses as the lower control limit because, just as with p charts and np charts, negative numbers of events (such as 3 defects on a radio) are meaningless.

Sample Problem Consider a firm that has decided to use a c chart to help keep track of the number of telephone requests received daily for information on a given product. Each day represents an area of opportunity. Over a 30-day period, 1,206 requests are received, or an average of 40.2 per day. This value is c, the centerline.

Given : c = 40.2
59.2 Actual counts occurring in an area of opportunity will always be whole numbers. Thus a count of 59 is within the control limits, while a count of 60 is beyond the UCL. The A,B and C zone boundaries are constructed at one and two standard errors from the centerline, respectively. The zone boundaries are:

Because the actual counts are whole numbers, the observations would fall into zones as follows: Zone Counts Upper A 53, 54, 55, 56, 57, 58, 59 Upper B 47, 48, 49, 50, 51, 52 Upper C 41, 42, 43, 44, 45, 46 Lower C 34, 35, 36, 37, 38, 39, 40 Lower B 28, 29, 30, 31, 32, 33 Lower A 22, 23, 24, 25, 26, 27

The zones each contain a reasonable number of whole numbers and are close enough in size to workable. But consider the problem that would have been encountered if the process average had been C = 2.4. Here we would get

As before, because the counts are whole numbers, the observations will fall into zones as follows:

Furthermore, keep in mind that when the average count is small, larger and larger areas of opportunity will b needed to detect imperfections. This will occur as a natural consequence of improved quality through the use of control charts.

Construction of a c Chart: An Example

 Consider the output of a paper mill: the product appears at the end of a web and is rolled onto a spool called a reel. Every reel is examined for blemishes, which are imperfections. Each reel is an area of opportunity. Result of these inspections produces the data in table below (table 1.1).

In this example the average number of imperfections per reel is Standard Error UCL (c) LCL (c) Centerline (c) = c = 6.00 = 2.45 = 6.00 + 3 (2.45) = 13.35 = 6.00 3 (2.45) = -1.35 (use 0.0) = c = 150/25 = 6.00

Fig. 1.1 Minitab c Chart for Blemishes

Small Average Counts When average counts are small, data appearing as counts will tend to be symmetric. This may lead to over adjustment (false alarms) or under-adjustment (too little sensitivity). False alarms are indications that the process is exhibiting special variation when no special variation can be found. Most often, these indications will be points on the control chart that are just beyond the upper control limit. False alarms, in and themselves can destabilize a stable process. In some cases, control limits calculated using the Equations 1.1 and 1.2 may not provide sufficient sensitivity to an indication of a special source of variation. This can result in a loss of opportunity for process improvement.

Fixed Control Limits

To avoid both of these problems, we may use a set of fixed control limits for c chart. These fixed control limits are sometimes called probability control limits and provide an excellent and economical rule for separating special and common variations when average counts are less than 20.

Therefore, for this application of the c Chart the control limits should properly have come from the Table 1.2. As 6.00 is in the 5.59 to 6.23 range, the values for the lower and upper control limits respectively are 0.5 and 13.5. These values have been used top draw by hand because Minitab does not incorporate fixed probability limits.

 Chart
 The Chart is similar to the c Chart in that it is a control chart for the count of

number of events, such as number of conformities over a given area of opportunity. The fundamental difference lies in the fact that during construction of a c chart, the area of opportunity remains constant from observation to observation, while this is not a requirement for the Chart. Instead, the Chart considers the numbers of events (such as blemishes or other defects) as a fraction of the total size of the area of opportunity in which these events were possible, thus circumventing the problem of having different area of opportunity for different observations. Examples of processes suitable for monitoring with a -chart include:  Monitoring the number of nonconformities per lot of raw material received where the lot size varies  Monitoring the number of new infections in a hospital per day  Monitoring the number of accidents for delivery trucks per day

The characteristic used for the control chart, , is the ratio of the number of events to the area of opportunity in which the events may occur. For observation , we call the number of events (such as imperfections) the observed c , and the area of opportunity a . Thus, is the ratio
=c /a Equation 1.1 The average of all the values, , provides a centerline for the control chart Centerline (u) = = c / a Equation 1.2 Standard Error = / a

Since the area of opportunity varies from subgroup to subgroup, so does the standard error. This results in control limits that vary from subgroup to subgroup:
LCL (u) = - 3
LCL (u) = + 3

/ a Equation 1.3
/a Equation 1.4

Construction of a Chart : An Example

Here is an example of the u chart template in the QI Macros for Excel. Just type data into the yellow input area and the chart will be drawn to the right. You can also cut and paste data from another spreadsheet into the template.

Consider the case of the manufacture of a certain grade of plastic. The plastic is produced in rolls, with samples taken five times daily. Because of the nature of the process, the square footage of each sample varies from inspection lot to inspection lot.
Centerline (u) = Average Number of defects /100 sq.ft. = u = 120/47.90 = 2.51

Table 3.1 Defects in Rolls of plastic

Limitations of Attribute Control Charts

As process improve and defects or defectives become rarer, the number of units that must be examined to find one or more of these events increases.

Table 4.1Control Limits for Defects in Rolls of Plastic

Fig. 4.1 Minitab u Chart for Number of Defectives in Rolls of Plastic

Attribute control charts are limited in terms of the level of process improvement, they enable. Another disadvantage of using attribute control charts is that if special variation from several different sources is present, it is difficult to identify and isolate the special sources individually.

Thank You!
PRESENTORS

JOSEILYN jo ANG ROBERTO obet LIZARDO JONAS nash JASARENO

End of part 3