Abstract
An expected-cost model is developed for the np-control chart when curtailed sampling procedures are followed. The total expected cost consists of the cost of sampling, the cost of investigating and possibly correcting the process when an out-of-control signal is received, and the cost of producing defectives. A Hook–Jeeves search procedure is used to identify the minimum-cost sampling policy as a function of three design variables: Rejection number, maximum sample size, and intersample interval. Numerical examples are presented, and comparisons are made to corresponding complete sampling policies. These results indicate that curtailed sampling plans provide cost improvements over traditional complete sampling plans and are, therefore, to be recommended whenever it is appropriate to use them.
Original language | English (US) |
---|---|
Pages (from-to) | 57-63 |
Number of pages | 7 |
Journal | Technometrics |
Volume | 27 |
Issue number | 1 |
DOIs | |
State | Published - Feb 1985 |
Externally published | Yes |
Keywords
- Average sample number
- Binomial sampling
- Cost function
- Inverse binomial sampling
- Markov process
- Truncated sampling
ASJC Scopus subject areas
- Statistics and Probability
- Modeling and Simulation
- Applied Mathematics