13-Sampling%20Procedures-rev%202011

13-Sampling%20Procedures-rev%202011 - Section 13 Sampling...

Info iconThis preview shows pages 1–3. Sign up to view the full content.

View Full Document Right Arrow Icon
Section 13 - Sampling Procedures UGA FDST 4090/6090 - (c) W.C. Hurst 2010 1 Sampling Procedures and Sampling Plans William C. Hurst FDST 4090/6090 (2010) Why sample? • Population or lot too large and expensive • Measurements destructive or dangerous • Boredom and fatigue cause more errors FDST 4090/6090 © W.C. Hurst 2010 SAMPLE MUST BE UNBIASED random, homogenous, independent Non-Statistical Sampling Choices 1. Zero Sampling – incidental ingredient (common salt used in brine for canned vegetables) 2. 100% Sampling – FDST 4090/6090 © W.C. Hurst 2010 a. Check weighers – filled containers b. Optical color sensors – tomato packing line c. Electric eye – segregate pecan halves 3. Judgment Sampling 4. Convenience Sampling 5. Quota Sampling Constant Percentage, Square Root • Larger lots are subjected to more scrutiny than smaller lots, FDST 4090/6090 © W.C. Hurst 2010 which are not examined closely enough SEE WORKSHEET Statistically Valid Procedures for Determination of Sample Size Sample size is based on three criteria: 1. Maximum error of estimate 2 Level of confidence desired FDST 4090/6090 © W.C. Hurst 2010 2. Level of confidence desired 3. Variability of population studied SEE WORKSHEET Sample Size Formula n E Z = 2 ˆ σ FDST 4090/6090 © W.C. Hurst 2010 Z = level of confidence E = margin of error tolerated σ = estimated variability in population ˆ SEE WORKSHEET
Background image of page 1

Info iconThis preview has intentionally blurred sections. Sign up to view the full version.

View Full DocumentRight Arrow Icon
Section 13 - Sampling Procedures UGA FDST 4090/6090 - (c) W.C. Hurst 2010 2 Sample Size Example • Suppose a quality control manager wants to estimate the true mean filled weight of 5-pound bags of sliced radishes on a new packaging machine in use. FDST 4090/6090 © W.C. Hurst 2010 • How large a sample must be taken to determine the performance of this machine if a 99% confidence level of an accuracy of 0.75 oz. is required and previous experience has demonstrated a 2-oz. variation in filled weight of these bags? SEE WORKSHEET Statistically Valid Sampling Plan Acceptance Sampling • Statistical sampling procedure employed to decide if a lot (population) of items is to be accepted or rejected on the basis of the FDST 4090/6090 © W.C. Hurst 2010 be accepted or rejected on the basis of the quality characteristics (i.e., number of defectives, measurement values) found in a randomly taken sample from the lot Considerations for Any Sampling Plan • Less information in sample • Represents “middle • Risk of rejecting “good” lots (Type I error) and accepting “bad” lots (Type II FDST 4090/6090 © W.C. Hurst 2010 Represents middle ground” between 100% inspection and no inspection error) Benefits of Acceptance Sampling • Can determine sampling error (risk) • Brings same discriminatory power to all FDST 4090/6090 © W.C. Hurst 2010 lots Types of Acceptance Sampling Attribute Sampling Plans – Inspect/ classify defectives/ defects in a random sample taken Variable Sampling Plans – Measurement taken and recorded for each item in the sample FDST 4090/6090 © W.C. Hurst 2010 from a lot – Total number of defectives in sample is compared with acceptance number and a decision is made about the lot
Background image of page 2
Image of page 3
This is the end of the preview. Sign up to access the rest of the document.

This note was uploaded on 09/10/2011 for the course FDST 4090 taught by Professor Hurst during the Fall '10 term at UGA.

Page1 / 25

13-Sampling%20Procedures-rev%202011 - Section 13 Sampling...

This preview shows document pages 1 - 3. Sign up to view the full document.

View Full Document Right Arrow Icon
Ask a homework question - tutors are online