![](/images/default/shared/large-loading.gif)
![](http://purplepass.s3.amazonaws.com/images/shared/data/45180/176310/event_header.png?r=1538119479)
Sequential Sampling to Reduce Inspection Costs
Presented By Compliance Training Panel
Thursday, October 4th at 1:00pm EST
Online sale ends: 10/04/18 at 12:45pm EST
Online Event
This is an Online Event
Bethesda, MD 20817
-= series =-
Buy Tickets
Talent
William A. Levinson Description
ANSI/ASQ Z1.4 (formerly MIL-STD 105) is a generally accepted sampling standard for pass/fail data. This webinar will show that, while the standard's double and multiple sampling plans reduce the amount of inspection, minimal inspection can be achieved with sequential sampling; a process in which the consecutive sample sizes are 1. Attendees will learn how to convert an ANSI/ASQ Z1.4 plan into a sequential sampling plan that provides comparable protection against poor quality, and how to communicate this to the customer. Objective : • Attendees will learn how to use ANSI/ASQ Z1.4, including definition of the plan based on (1) the lot size, (2) the sampling level, and (3) the acceptable quality level (AQL), and also the switching rules for normal, tightened, and reduced inspection. • Attendees will learn (and a pre-programmed spreadsheet will be provided to help) how to convert the ANSI/ASQ Z1.4 plan into a sequential sampling plan, and also how to tabulate the sequential plan in a dozen or fewer rows to make it easy to deploy and use on the shop floor. This addresses the traditional administrative obstacle to use of sequential plans, namely, the need for a table with a hundred or more rows—one for each item that is inspected. • The webinar will also show how to calculate the operating characteristic (OC) curve for the sequential plan to demonstrate to customers that it has the same chance of rejecting a lot at the rejectable quality level (RQL) as the original ANSI/ASQ Z1.4 plan. (ANSI/ASQ Z1.4 plans do not have formal RQLs but we pretend, for computation purposes, that the RQL is the nonconforming fraction at which the plan has only a 10% consumer's risk of accepting the lot.) The average sample number (ASN) will meanwhile be consistently lower than that of the traditional plan's sample size, and even the ASN of the double and multiple sampling plans. Areas Covered in the Session : 1. Inspection is a necessary but non-value-adding activity; the less we can do while continuing to meet customer requirements, the better. 2. How to use ANSI/ASQ Z1.4 • Selection of the sampling plan based on (1) the lot size, (2) the inspection level, and (3) the acceptable quality level (AQL) • Application of the switching rules depending on process performance. Poor quality will require tightened inspection, while consistent good quality allows reduced inspection. 3. Conversion of an ANSI/ASQ Z1.4 plan into a sequential plan • The sequential plan's specifications depend on (1) the AQL and producer's risk alpha, i.e. the chance of rejecting a lot at the AQL, and (2) the RQL and consumer's risk beta, i.e. the chance of accepting a lot at the RQL. ANSI/ASQ Z1.4 does not specify RQLs but, for computation purposes, we pretend the RQL is the nonconforming fraction at which the chance of acceptance is 10 percent. • The sequential plan can then be tabulated, not in terms of acceptance and rejection numbers that depend on the sample size (which can easily result in tables with a hundred or more rows even with stopping rules that prevent inspection from continuing forever) but rather in terms of acceptance and rejection sample sizes that depend on the number of nonconforming parts that have been found. This reduces the table to a dozen or even fewer rows in most cases, which makes it easy to use on the shop floor, and is one of the webinar's most important takeaways. • The sequential plan's OC curve will intersect that of the original ANSI/ASQ Z1.4 plan at the AQL and 1 minus alpha (i.e. the chance of acceptance at the AQL) and the RQL and beta, which shows the customer that the protection against poor quality is roughly identical between the AQL and RQL. • The sequential plan's average sample number also can be calculated for comparison to that of the original plan, for any given quality level. • It may, on the other hand, be simpler to use ANSI/ASQ Z1.4's reduced sampling plan due to the need to make two decisions—one as to whether to accept the lot, and the other as to whether to go back to normal inspection. It is possible to accept the lot but find enough nonconformances to invoke the switching rules. Who Will Benefit: Manufacturing and quality engineers, technicians, and inspectors with responsibility for inspections that involve quantitative measurements of parts. About Speaker: William A. Levinson, P.E., FASQ, CFPIM, is the owner of Levinson Productivity Systems PC. He holds professional certifications from the American Society for Quality, APICS, and Society of Manufacturing Engineers. Organizer: Compliance Training Panel Email: support@compliancetrainingpanel.com Phone: 844-216-5230 |
Presented By
1:00pm to 2:00pm
General Admission: $179.00
21+
200 844-216-5230 PAST EVENTS
Tuesday, September 25th
Online Event Past Event
|