You are being redirected because this document is part of your ASTM Compass® subscription.
    This document is part of your ASTM Compass® subscription.

    STP66

    Use of Statistics in Writing Specifications

    Published: 0


      Format Pages Price  
    PDF (532K) 4 $25   ADD TO CART
    Complete Source PDF (948K) 19 $35   ADD TO CART

    Cite this document

    X Add email address send
    X
      .RIS For RefWorks, EndNote, ProCite, Reference Manager, Zoteo, and many others.   .DOCX For Microsoft Word


    Abstract

    B word “specification” as used in this paper is meant any specification employed for the purpose of defining the desired quality of any product for utilitarian or aesthetic purposes. A specification is necessary because different lots of the same product might possess the same quality characteristic in different degrees, whereas, for best results, it is desired that the particular quality characteristic be maintained at a definite level and within very definite limits. It is the function of the specification to define this level of quality and the limits within which the quality level may be allowed to vary. For this purpose the specification must possess two operating characteristics: 1. Accuracy.—The specification must define the desirable quality level and its permissible variation with sufficient clarity as to allow definite discrimination between good lots and bad lots. 2. Economy.—The specification must permit discrimination between good lots and bad lots without the requirement of 100 per cent inspection, or the examination of all units in a lot. This is particularly important in the case of destructive testing, where the test operation itself would leave us no “good parts.” The fundamental characteristic of a manufactured product which makes the application of statistical measurements essential is the variation inherent in the product. No two pieces produced are exactly alike regardless of the refinement of the process. Although in some cases the differences are smaller than can be detected with the measuring devices on hand, we can be assured that differences do exist.


    Author Information:

    Goffman, Casper
    Staff Assistant and Director, Westinghouse Electric Corp., East Pittsburgh, Fa.

    Manuele, Joseph
    Staff Assistant and Director, Westinghouse Electric Corp., East Pittsburgh, Fa.


    Committee/Subcommittee: E11.20

    DOI: 10.1520/STP47661S