X CHART CONTROL LIMITS BASED ON A SMALL NUMBER OF SUBGROUPS
Abstract
AD-286 384Div. 15 U (TISTP/WH) OTS price 3.60Applied Mathematics and Statistics Labs., Stanford U., Calif. X CHART CONTROL LIMITS BASED ON A SMALL NU BER OF SUBGROUPS, by Frederick S. Hillier. 4 ep 62, 29p. incl. tables, 21 refs. (Technical rept. no. 63) (Contract Nonr-22553, Proj. NR-042-002) Unclassified report DESCRIPTORS: *Groups (Mathematics), *Analysis of variance, *Quality control, *Statistical processes, Statistical distributions. Conventional X chart control limits provide a very unreliable basis for indicating when the process ha fallen out of statistical control unless these limits are based on a fairly large number of subgroups, say 25 or more. A method is presented which provide statistically sound control limits regardless of the smallness of the number of subgroups observed. This permits the reliable use of X charts during the crucial initiation of a new process, during the start-up of a process just brought into statistical control again, or for any process whose total output is not very large. The method is simple, requiring only the use of Table 4.1, and the interpretation of results is the same as with conventional control limits. Finally, even hen conventional control limits are used, the results given in Table 3.1 permit a precise probabilistic evaluation of these control limits. (Author)
Document Details
- Document Type
- Technical Report
- Publication Date
- Sep 04, 1962
- Accession Number
- AD0286384
Entities
People
- Frederick Stanton Hillier
Organizations
- Stanford University