THE DEVELOPMENT OF MATHEMATICAL MODEL FOR SINGLE ACCEPTANCE SAMPLING BY CONSIDERING INSPECTION ERROR
The increasing growth of the manufacture industry required the industry to always increase <br /> <br /> their operational activities in order to provide suit quality products that are in line with the <br /> <br /> consumer needs. Currently, company is not only focusing on t...
Saved in:
| Main Author: | |
|---|---|
| Format: | Final Project |
| Language: | Indonesia |
| Online Access: | https://digilib.itb.ac.id/gdl/view/26379 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| Institution: | Institut Teknologi Bandung |
| Language: | Indonesia |
| Summary: | The increasing growth of the manufacture industry required the industry to always increase <br />
<br />
their operational activities in order to provide suit quality products that are in line with the <br />
<br />
consumer needs. Currently, company is not only focusing on the quantity but also the quality <br />
<br />
of their products. The common method that used in controlling the inspection process of <br />
<br />
product acceptance is acceptance sampling. However, this method still has shortcoming in <br />
<br />
doing calculation, the assumption that used was perfect inspection or error free where <br />
<br />
basically humans are not free from errors. Therefore, to overcome this problem, a <br />
<br />
mathematical model of acceptance sampling that focus on single acceptance sampling by <br />
<br />
considering inspection error. <br />
<br />
The development of mathematical model conducted in this research is integrating the <br />
<br />
classification error formula using reference from Statistical Research Group, Columbia <br />
<br />
University. There are two error value considered 𝑒1 and 𝑒2, so that the formula for <br />
<br />
developing mathematical model for probability of acceptance is 𝑃𝑎𝑒 = Σ ( <br />
𝑛 <br />
𝑑 <br />
𝑐 ) <br />
𝑑=0 (𝑝∗)𝑑(1 − <br />
𝑝∗)𝑛−𝑑 .The performance measures that used to evaluate this sampling plan are operating <br />
<br />
characteristic (OC) curve, average outgoing quality (AOQ), and average total inspection <br />
<br />
(ATI). The error probability value that used in this research are 0.001 and 0.01. Furthermore, <br />
<br />
the sampling plan data processing is carried out based on three types of risk (producer risk, <br />
<br />
consumer risk, and producer and consumer risk). <br />
<br />
Based on data processing results, it was found that there was a differences between the <br />
<br />
assumption of perfect inspection and consider human error. The greater value of probability <br />
<br />
error, the less amount of sample acceptance received so that it has an impact on AOQ and <br />
<br />
ATI which can affect the total costs incurred by the company. In this research, researcher <br />
<br />
also conducted sensitivity analysis for each parameter α, β, AQL, and LQL to the probability <br />
<br />
changes of error parameters (such as 0.0001, 0.0005, 0.001, 0.005, 0.01, and 0.05) for each <br />
<br />
performance measure. The results of sensitivity analysis indicate that there are significant <br />
<br />
changes for each performance measure when the probability changes of error parameter <br />
<br />
become 0.05 for all parameters. |
|---|