Quality Newsletter

Quality Fast Facts: Variables vs. Attributes Inspection

In basic and brief terms attributes control is at the limits, variables control is within the limits. Concerning the data that is generated by each method, attributes data is discreet whereas variables data is continuous. It is established fact that attributes control has been on a gradual but pronounced reduction in American Industry since the mid to late fifties. Some of the reasons for this trend are as follows:

Manufacturing and quality are no longer separate functions in best in class companies - indeed they should not be. Inspection for the sole purpose of outright acceptance or rejection denies the essential economic advantages inherent in a modern, well-defined quality program. It is axiomatic that passing judgment on a lot or batch of material that has already been manufactured without proper in-process control is weak and ineffective quality assurance.

A well-defined quality program requires the commitment and leadership of management to be successful. At ACD, managers continually improve process, educate workers, and express quality concerns to suppliers in a timely fashion to mitigate unnecessary queue times.

One area that comes under close scrutiny is attributes inspection. Attribute inspection is effective as a final acceptance method for product/parts/components after it is produced/manufactured. Almost without exception it is found that attributes inspection is counter-productive when used as a process quality control method. When proper comparisons to variables inspection are evaluated, the following undeniable facts emerge:

  1. Variables inspection offers complete control over the manufacturing process. With variables data trends are seen allowing for corrective action to be taken long before the product reaches the reject level.
  2. Variables data allows for the orderly and precise control over process adjustments. Changes in the process, when necessary, are easily effected. This may involve changes in tooling, fixturing, heat, etc. When completed, the results are known—not guessed at. The great bulk of dollars spent on poor quality usually evolve around the following areas:
  • Scrap
  • Sort
  • Selective Assembly
  • Downtime
  • Hunting for the Assignable Cause
  • Excessive Attribute Inspection

Variables gaging/in-process control mechanism provides for the avoidance of each of these costly areas.

  1. Variables gaging/control provides for uniformity of results. Everyone achieves the same results.
  2. With variables data communications are greatly improved. Specific rather than hazy information is transmitted.
  3. Variables gaging/control allows the use of modern statistical quality control techniques to be implemented such as control charts, capability studies, tool life studies, etc. These techniques in most cases allow for less inspection of the product itself because of the positive elements of control.
  4. Variables gaging/control is easier to calibrate and maintain. Generally, it is much faster than attributes gaging.
  5. Attributes gaging can require an operating sequence of push-pull and feel whereas variables gaging/control requires in essence (only) a scan of the readout. Experience tells us that attributes gaging on large quantities creates excessive operator fatigue and unreliable results.
  6. Attributes gaging/control is subject to greater wear. In the case of the screw thread the constant turning and screwing in and out of receiver type gaging creates sufficient abrasion to wear gage surfaces very rapidly.
  7. Since attributes gaging wears in the direction toward minimum material, rework is costly and in many cases impossible.

There are numerous other technical and economic advantages related to variables gaging/in process control. The most profound example may be when proper attributes gaging is compared to proper variables gaging and control. Cost-wise it is surprising to see that variables gaging/in process control does not create much of a cost differential when compared to the benefits of implementing the measurement system.