1、ANSVASQC BI -1 996, ANSVASQC B2-1996, and ANSI/ASQC B3-1996 AMERICAN NATIONAL STANDARD Guide for Quality Control Charts Control Chart Method of Analyzing Data Control Chart Method of Controlling Quality During Production AMERICAN SOCIETY FOR QUALITY CONTROL 61 1 EAST WISCONSIN AVENUE MI LWAU KEE, WI
2、SCONSIN 53202 ANSVASQC B1-1996, ANSVASQC B2-1996, and ANSVASQC B3-1996 AMERICAN NATIONAL STANDARD Guide for Qualiiy Control Charts Control Chart Method of Analyzing Data Control Chart Method of Controlling Quality During Production Reatfirmation of A NSI/ASQC BI- 1985, ANSI/ASQC B2- 1985, and ANSI/A
3、SQC B3- 19851 Prepared by American Society for Quality Control Standards Committee for American National Standards Committee Z-1 on Quality Assurance An American National Standard Approved on February 26, 1996 American National Standard: An American National Standard implies a consensus of those sub
4、stantially concerned with its scope and provisions. An American National Standard is intended as a guide to aid the manufacturer, the consumer, and the general public. The existence of an American National Standard does not in any respect preclude anyone, whether he or she has approved the standard
5、or not, from manufacturing, marketing, purchasing, or using products, processes, or procedures not conforming to the standard. American National Standards are subject to periodic review, and users are cautioned to obtain the latest eitions. Caution Natice: This American National Standard may be revi
6、sed or withdrawn at any time. The procedures of the American National Standards institute require that action be taken to rem, revise, or withdraw this standard no later than five years from the date of publication. Purchasers of American National Standards may receive current informa- tion on all s
7、tandards by calling or writing the American Nationai Standards Institute. 01996 by ASQC All rights reserved. No part of this book may be reproduced in any form or by any means, electronic, mechanical, photo- copying, recording, or otherwise, without the prior written permission of the publisher. ASQ
8、C Mission: To facilitate continuous improvement and increase customer satisfaction by identifying, communicating, and promoting the use of quality principles, concepts, and technologies; and thereby be recognized throughout the world as the leading authority on, and champion for, quality. 10987 6543
9、 2 Printed in the United States of America Printed on acid-free paper Published by: ASQC 611 East Wisconsin Avenue Milwaukee, Wisconsin 53202 STD-ASd B1-ENGL L99b 075950b 0002112 b37 ANSUASQC B1.1996. ANSVASQC B2.1996. and ANSVASQC B3-1996 Contents page 1 . scope 1 3 . Specification Limits 1 5 . Var
10、iation of Quality-Assignable Causes 7 . Practical Advantages of a State of Control . ANSVASQC B1.1996. Guide for Quality Control Charts . 1 2 . TheControlChart . 1 4 . ControlLimits 1 1 6 . StateofControl . 2 2 8 . Establishment of Control Limits 2 9 . Principal Uses of the Control Chart . 3 3 3 10
11、. Use of the Control Chart for Judging Whether Control Exists . 11 . Use of the Control Chart for Attaining and Maintaining Control of Quality ANSVASQC B2.199. Control Chart Method of Analyzing Data . 6 1 . Scope 6 2 . NatureoftheDataUsed . 6 3 . ComputationoftheStatisticalMeasures 6 4 . IfrPesofCon
12、trolCharts . 7 5 . Construction of a Control Chart 7 6 . General Instructions for Control Chart Method 7 7 . Choice of Control Limits . 7 ANSYASQC B3.1996. Control Chari Method of Controlling Quality During Production . 13 1 . Introduction . 13 1.1 scope 13 1.2 %o Purposes of Inspection . 13 1.4 Qua
13、lity Characteristics Requiring Sampling Inspection 14 1.5 Need for Keeping Records of Quality 1.6 Different I)rpes of Inspection Data 15 1.3 Decrease in Lot-by-Lot Inspection Accruing from Statistical Control . 14 15 2 . Preliminary Steps in Setting up a Control Procedure . 15 2.1 Choice of Quality
14、Characteristics 2.2 Analysis of the Production Process 16 2.3 Planning How to Collect and Subgroup Data . 16 2.4 Choice of Statistical Measures . 17 3 . StartingtheControlChart . 18 3.1 Collection and Analysis of Preliminary Data . 18 3.2 Establishment of Control Limits-General-Standard Values 18 3.
15、3 Establishment of Control Limits-Charts for X and s . 18 3.4 Establishment of Control Limitsxharts for X and R . 21 3.5 Establishment of Control Limits-Charts for p or pn . 22 3.6 Establishment of Control Limits-Chart for c 23 3.7 Preparing the Control Chart for Use . 24 . 15 . lu STD-ASQ B1-ENGL 2
16、99b = 0759506 0002113 573 D ANSUASQC B1.1996. ANSUASQC B2.1996. and ANSUASQC B3-1996 4 . Using the Control Chari During Production . 24 4.1 Plotting Points on the Control Chari; Taking Action 24 4.2 When Control May Be Assumed to Exist . 24 4.3 Review of Control Chart Standards 24 4.4 ControlChartRe
17、cords 25 5 . Illustrative Examples . 25 5.1 Example 14ontrol Chart Using X and R-Control Procedure illustrated Step by Step . 25 5.2 Example 2-Contml Chart Using f3 Only-Process Control Problem 31 5.4 Example Lccontrol Chart Using x and R-Control of Spoilage for a Bulk Product 5.3 Example 3-Control
18、Chart Using Fraction Nonconforming p-lontrol of Spoilage. 100-Percent Inspection of a Product Consisting of individual Units 31 . 34 6 . Glossary of Symbols and Terms . 37 Appendix A Factors for Computing Probability Control Limits . 39 Appendix B References 41 iv Foreword (This Foreword is not a pa
19、rt of American National Standard ANSUASQC B1-1996, Guide for Quality Control Charts, nor of American National Standard ANSUASQC B2-1996, Control Chart Method of Analyzing Data, nor of ANSUASQC B3-1996, Control Chart Method of Controlling Quality During Production.) Upon request by the War Department
20、, the American Standards Association, in December, 1940, initiated a project on the application of statistical methods to the quality control of materials and manufactured products. Since, due to the national emergency, there was an urgent need for the prompt development of standards in this field,
21、the ASA Defense Emergency Procedure (later called the War Emergency Procedure) was applied to this project, and the following Emer- gency Technical Committee (later, War Committee) was appointed to develop such standards: H. F. Dodge, Bell Telephone Laboratories, Inc., Chairman A. G. Ashcroft, Alexa
22、nder Smith and Sons Carpet Company W. Edwards Deming, Bureau of the Census Leslie E. Simon, rdnance Department, U.S. Army R. E. Wareham, General Electric Company John Gaillard, American Standards Association, Secretary This committee developed these three standards. Drafts were submitted for critici
23、sm and comment to a number of key individuals in groups having a substantial interest in the subject of the standards. All of the comments were carefully reviewed by the committee and a number of changes were made in accordance with suggestions received. The revised drafts of Z1.1 and Z1.2 were unan
24、imously approved by the Emergency Technical Committee and received ASA approval as American Defense Emergency Standards (later, War Standards) on May 27, 1941. 21.3 was first approved in 1942. in November, 1952, the ASA invited the American Society for Quality Control ASQC) to accept the proprietaty
25、 spon- sorship for the three standards which had been developed by the War Committee. The invitation was accepted by ASQC in February, 1953, and the standards were turned over to the ASQC Standards Committee who assigned the designa- tions ASQC B1, ASQC B2, and ASQC B3 to the standards which were la
26、ter to become ANSI standards Zl.l,Z1.2 and 21.3 respectively. The personnel of the committee at that time was as follows: Irving W. Burr, Purdue University W. Edwards Deming, Consultant in Statistical Surveys, New York University Harold F. Dodge, Bell Telephone Laboratories, Inc. retired) and Rutger
27、s, The State University of New Jersey, Eugene L. Grant, Stanford University Ralph E. Wareham, Consultant in Quality Control Chairman One of the duties of the ASQC as Proprieary Sponsor was the establishment of a national consensus on approval of the standards by industry, In August, 1956, a canvass
28、of industry was instituted in which organizations believed to have a substantial interest in the subject of quality control were contacted. This canvass resulted in ail but three of the orga- nizations interested approving the standards as circulated by ASQC. After further review by the Standards Co
29、mmit- tee of ASQC, in the light of comments received in the course of the canvass, ASQC felt that the basic criticisms had been covered by making minor modifications and bringing the appendixes up to date, and accordingly submitted the standards to ASA for approval as American Standards. in the cour
30、se of considering the submittal for a recommenda- tion on approval, the Miscellaneous Standards Board, which had jurisdiction over this work, requested that those orga- nizations which had objected be contacted again to ascertain their present feelings in the matter. This was done with the result th
31、at the organizations involved announced that they now approved the standards. Accordingly, after receiving a favorable recommendation from the Miscellaneous Standards Board, the American Stan- dards Association approved the standards as American Standards on November 21, 1958. V ANSIIASQC B1-1996, A
32、NSVASQC B2-1996, and ANSUASQC B3-1996 In 1981, the ANSI 21 Statistical Methods Subcommittee recommended to the ASQC Standards Committee that the siandards Zl.1, Guide for Quality Control, 21.2, Control Chart Method of Analyzing Data, and 21.3, Control Chart Method of Controlling Quality During Produ
33、ction can be updated to include more modern terminology and symbols in keeping with the American National Standard ANSUASQC A1-1978, Dejnitiom, Symbols, Fomh and Tables for Control Charts and other publications such as the ASTM STPlSD, ASTM Manual on Presentation of Data and Con- trol Chart Analysis
34、. The task was assigned to the ASQC Statistics Division which formed a writing committee that completed the task of updating these standards. The writing committee combined the three standards, Zl.1, 21.2, and 21.3 under one cover as had been done previously with 21.1 and 21.2, because all three doc
35、uments are concerned with the statistical qual- ity control charts. The significant changes are: (i) the redefinition of the sample standard deviation to be s = dX(X - q2 / (n - 6; (2) the use of the words nonconforming and nonconformities to replace defectives and defects respectively; (3) the use
36、of a subscript (O) to replace the prime symbol () e.g. po used in place of p, for designating a standard value, and (4) the amplification of the material on Warning Limits. The redifinition of the sample standard deviation also required changing many factors used in control chart work (c4 in place o
37、f c2; A3 in place of Al; B5 and Bs in place of B1 and B2 respectively; E3 in place of EJ. Table 6 Fmtors for Computing Control Chart Lines and the Table A2 in the Appendix now use the updated factors. The above changes required many changes in these standards. This document is a reaffirmation of the
38、 1985 version. scope The scope of this document is intended to cover the Shewhart statistical quality control charts which are in general use in the United States manufacturing and service industries. writing committee The following individuals were members of the writing committee for the 1985 revi
39、sion of ANSI Zl.l,21.2, and 21.3 1958: Sherman L. Babcock, Chairman Hardy M. Cook, Jr, Acheson J. Duncan C. Allen Mannon Harrison M. Wadsworth, Jr. Oswald Wiliner Suggestions for improvement of this standard will be welcome. They should be sent to the Standards Administrator, American Society for Qu
40、ality Control, 611 East Wisconsin Avenue, Milwaukee, WI 53202. vi STD.ASd BL-ENGL 177b m 075750b 00021Lb 282 ANSUASQC B1-1996 ANSUASQC B1-1996 Guide for Quality Control Charts 1. SCOPE 1.1 This standard is intended as a guide for handling prob- lems concerning the economic control of quality of mate
41、- rials, manufactured products, services, etc. It has particular reference to methods of collecting, arranging, and analyz- ing inspection and test records in a manner designed to de- tect lack of uniformity of quality. For the sake of simplicity, the term product will be used hereafter, whether the
42、 object whose quality is being con- sidered is a material, a semi-finished or finished product, or a service. Thus the words “manufacturer, purchaser, production process, etc.” should be interpreted broadly to cover many fields of interest. 2. TEIE CONTROL CHART 21 It is recommended that the control
43、 chart be used for handling quality control pblems, for example, for control- iing quality during manufacture, for presenting the essential information of the quality records, as an aid in judging how well the quality is controlled, for continuing quality improvement, and as an indicator as to when
44、action should be taken to prevent quaiity problems from occurring. 2.2 The control chart is a graphical record of quality. On it are placed a pair of control limits. These limits are of assistance in judging the significance of variations of product quality around the general level, particularly wit
45、h a view to the more important function of exercising purposeful control over the quality. Moreover, they are placed such that a plotted point failing outside them dur- ing manufacture may be taken as an indication of a cause of variation that should be investigated. 3. SPECIFICATION LDaS 3.1 Before
46、 the nature of control limits is explained, it may be well to visualize the nature of specification limits. These are the limits given in the product specification to define the extreme permissible values of a quality charac- teristic, to ensure comt performance of the individual unit of product. Sp
47、ecification limits are used by the manu- facturer in his own plant, or by the purchaser when receiv- ing a consignment, as a basis for checking the quality of each individual unit inspected. A typical example of specification limits are the minimum and maximum limits for the diameter of a shaft. 4.
48、CONTROLLIMITS 4.1 The Control limits on a control chart are different from specification limits. They are used, not for checking the quality of each unit of the product, but as a basis for judging the significance of the quality variations from sample to sample, from lot to lot, or from time to time
49、. They supply a criterion for deciding whether a production process is being disturbed by causes of variation that should be investigated. Control limits apply to some measure of the collective quality of a group of units, this measure to be computed from observations made on the individual units in the group. Quality measures commonly used are (1) the average of the observed values of the individual units under consider- ation, and (2) some measure of the dispersion of the ob- served values around their average, such as the standard hiation or the range. Where individual
