Simplicity and robustness in the inspection of continuous processes
Simplicity and robustness in the inspection of continuous processes
A process or a system operating continually in time could be in one of several states, some of the states being undesirable; these states are described by 0 with 0 - 00 being the preferred value. Inspection is necessary for obtaining information about 0 , and a suitable corrective action will be taken if a departure from 00 is indicated. A process, such as a manufacturing process, may result in a continual flow of discrete items and a distinction between 0 as process quality or product quality is rode. With 0 as process quality, in the absence of a model for changes inconstant inspection intervals are used and a two state model for 0 is chosen for generating non-constant inspection intervals. A simple statistic generated by a suitable trinomial distribution is shown to be versatile and robust. Dichotomies and trichotomies generated by the division of the domain of the statistics yield simple decision rules for indicating departures from 00 and it is shown that the trichotomous rules have some attractive features including robustness. Some graphical aids for the choice of a scheme are given. With 0 as product quality, many of the available inspection schemes cause an operational problem due to the interdependence of sampling and complete inspection. Four new-schemes are developed and some graphical aids for the choice a scheme are given.
University of Southampton
Shahani, Arjan Kewalram
6218188a-9d25-4638-858c-76faa3d64d05
1977
Shahani, Arjan Kewalram
6218188a-9d25-4638-858c-76faa3d64d05
Shahani, Arjan Kewalram
(1977)
Simplicity and robustness in the inspection of continuous processes.
University of Southampton, Doctoral Thesis.
Record type:
Thesis
(Doctoral)
Abstract
A process or a system operating continually in time could be in one of several states, some of the states being undesirable; these states are described by 0 with 0 - 00 being the preferred value. Inspection is necessary for obtaining information about 0 , and a suitable corrective action will be taken if a departure from 00 is indicated. A process, such as a manufacturing process, may result in a continual flow of discrete items and a distinction between 0 as process quality or product quality is rode. With 0 as process quality, in the absence of a model for changes inconstant inspection intervals are used and a two state model for 0 is chosen for generating non-constant inspection intervals. A simple statistic generated by a suitable trinomial distribution is shown to be versatile and robust. Dichotomies and trichotomies generated by the division of the domain of the statistics yield simple decision rules for indicating departures from 00 and it is shown that the trichotomous rules have some attractive features including robustness. Some graphical aids for the choice of a scheme are given. With 0 as product quality, many of the available inspection schemes cause an operational problem due to the interdependence of sampling and complete inspection. Four new-schemes are developed and some graphical aids for the choice a scheme are given.
This record has no associated files available for download.
More information
Published date: 1977
Identifiers
Local EPrints ID: 458691
URI: http://eprints.soton.ac.uk/id/eprint/458691
PURE UUID: 429f233b-e7cc-4c1a-94e3-dbc2c6f65896
Catalogue record
Date deposited: 04 Jul 2022 16:54
Last modified: 23 Jul 2022 00:21
Export record
Contributors
Author:
Arjan Kewalram Shahani
Download statistics
Downloads from ePrints over the past year. Other digital versions may also be available to download e.g. from the publisher's website.
View more statistics