Percent defective

With attribute data the product either has or has not the ascribed attribute - it can therefore either pass or fail the test. There are no gray areas. Attributes are measured on a go or no-go basis. With variables, the product can be evaluated on a scale of measurement. However, with inspection by attributes we sometimes use an acceptable quality level (AQL) that allows us to ship a certain percent defective in a large batch of product -... 

Application. A company has received a very large shipment of rivets. One product specification required that no more than 2 percent of the rivets have diameters greater than 14.28 mm. Any rivet with a diameter greater than this would be classified as defective. A random sample of 600 was selected and tested with a go-no-go gauge. Of these, 16 rivets were found to be defective. Is this sufficient evidence to conclude that the shipment contains more than 2 percent defective rivets ... 

The relationship between Cp and percent defective (assuming a normal distribution) is given in Table 4. 

Figure 10. Pareto chart analysis of the percent defects out of total number of wafers produced before and after process control.

Figure 4 Contour plot of burn percent defective. Abbreviations. DT, dwell time HB, hot bar temperature.

Many practitioners in the area of quality engineering recommend that a minimum value for Cp should be 1.33 in order for virtually no defects to occur due purely to the variability in the process. This value of Cp implies that the width of the specification should be eight times that of the estimated process standard deviation. With this value for Cp, only 0.006% defects will be produced. One can see immediately the nonlinear relationship between Cp and the percent defective. A more comprehensive list of Cp values versus percent of the population falling outside specification limits is given in Table 11. The results of a... 

Table IV. Reactivity Ratios and Percent Defect (head-to-head and tail-to-tail) for Aregic PVF Samples ...

The p chart is used to present percent defective. A sample of 100 to 1000 items would be examined for each data point and the percent defective would be calculated and plotted on the chart. This calculation involves taking the number of defectives divided by the number in the sample and then multiplying by 100 to get the percent. The graph can be in fractions or percent. The p chart is one of the control charts for attributes that can have a variable sample size due to the ratio calculation from the data. The thing to remember is that when making a p chart the sample size should not vary much more than 25 percent from the largest sample size to the smallest sample size. 

Tolerance interval n. The specified allowance of variation of a dimension (or other quantity) above and below the nominal or target values in a production part or product. For a process whose average level and random variation are in control, symmetrical tolerance limits should be at least six processes standard deviations apart in order to approach zero percent defective parts. Tolerances are better understood and getting much more attention now than a few decades ago when a New York molder was asked about tolerances on parts he was producing. His reply Hey, we got lotsa tolerance here We hire people no matter what color or nationality they are . 

The LWP method has several drawbacks. It is generally difficult to come up with exact criteria weights. The method also requires all the criteria values to be scaled properly if the measurement units are very different. For example, quality may be measured in "percent defectives," service may be measured by lead-time in days, capacity in hundreds of units, and price in dollars. The use of linear additive value fimction violates the normal economic principle of "Diminishing Marginal Utility." In Section 6.3, we will discuss several methods for ranking suppliers that overcome these drawbacks. 

Buyer 1 The weighted average rejects of both products received from the two suppliers (measured in percent defectives) cannot exceed 6% for Buyer 1. 

At PPI (see Chapter 3) the measurement system and the employee reward systems were based on units produced, just as they were at Signetics. And that directly related to PPFs over 14 percent defect rate. It generated what it measured, and so its solution was to add control systems, such as Statistical Process Control (SPC), which, in and of itself, is a very good system when applied as a performance enhancement tool used to measure process performance as it happens. But as a control system, which was after the fact, it had no effect on quality output. All it did was confirm that indeed the defect rate was over 14 percent. It identified a varying number of areas where errors occurred, but it did little to aid in identifying solutions. 

Taken an over 14 percent defect rate down to 2 percent... 

Reduce a more than 14 percent defect rate to below 2 percent. 

When a vehicle is declared out-of-servIce for a condition resulting from an accumulation of violations, all violations that contributed to the specific out-of-service condition must be repaired (e.g., a vehicle, or vehicles in combination declared out-of-service for 20 percent detective brake violations must have all the 20 percent defective brake violations repaired prior to being released or, a vehicle declared out-of-service for two tires at less than 1/32 inch (0.8 mm) tread depth must have both tire violations repaired prior to the vehicle being released, etc.). Once all of the contributing out-of-service violations have been repaired on any vehicle in a combination, that specific vehicle in the combination is no longer considered to be out-of-service. 

When vehicies in combination are deciared out-of-service for 20 percent defective brake violations, any vehicle within the combination that does not contain a brake violation that contributed to the 20 percent defective brake out-of-service condition is allowed to proceed providing It does not contain any other out-of-service conditions. 

P Charts. One of the most commonly used control charts for attributes is the percent defective or P chart. P charts are prepared by obtaining a series of... 

If a negative value (less than zero) of the lower control limit is obtained, it should be considered as zero on the chart since the percent defective cannot be less than zero. 

C Charts. The C chart is also known as the C-bar chart and is a special type of attributes control chart. Unlike the P chart, which portrays percent defective, the C chart uses the number of defects per unit. For example, a black speck and a deep sink mark on a molded part are two defects per part for C chart calculation but one part defective for P chart purposes. 

---