China
Africa
Explore chapters and articles related to this topic
Quality by Control
Published in Anthony J. Hickey, Sandro R.P. da Rocha, Pharmaceutical Inhalation Aerosol Technology, 2019
Helen N. Strickland, Beth Morgan
The first step within PPQ is to develop options focusing on consumer risk. One plan is to choose an isolated lot acceptance sampling plan indexed on a limiting quality value of 0.5% from ISO 2859-2 (1985). This plan requires sampling 800 units in total (80 per segment across ten segments). The plan will pass (i.e. the batch will be released to market) if the number of defective units from the sample of 800 is less than or equal to 1. This sampling plan has a 9.1% chance of passing at the limiting quality value of 0.5%. This means that the consumer risk for this plan is 9.1%; i.e. there is a 9.1% chance that a batch that contains 0.5% defective would be released to the market. Another option is to choose an acceptance sampling plan that allows zero defects and controls the consumer risk to be 5% as opposed to, approximately, 9%. This is a user defined acceptance sampling plan. This plan would sample 600 units (60 per segment), and the batch would be released if the number of defects observed was zero. A final option would be a user-defined plan that allowed one defective unit, but still controlled the consumer risk to be 5% at the limiting quality value of 0.5%. This acceptance sampling plan would sample 950 units (95 per segment), and the batch would be released to market if the number of defective units was less than or equal to 1. This highlights that there are three possible sampling plan options, but they do not provide the same level of consumer risk protection.
Optimal designing of two-level skip-lot sampling reinspection plan
Published in Journal of Applied Statistics, 2022
N. Murugeswari, P. Jeyadurga, S. Balamurali
Among various sampling plans designing approaches such as one-point approach, minimum angle method, and so on, two points on the OC curve approach admires the researchers because it provides more preference for producer and consumer expectations. That is, this approach simultaneously considers PQL, denoted by p1 and CQL, denoted by p2 along with their corresponding risks α and β when selecting optimal parameters. The general expectation of implementing any sampling plan is the inspection effort to be reduced and the corresponding cost should be minimized. The reduction of effort and time involved in acceptance sampling inspection can be achieved by the minimization of ASN of the concerned plan. Due to these reasons, we attempt to determine an optimal plan with minimum ASN by applying two points on the OC curve approach. The performance measures probability of acceptance, ASN and ATI of the proposed plan are given in Equations (4), (5) and (6). The value of P, probability of acceptance of the lot under reference plan, is calculated using binomial distribution and given in the following equation: p1 (or PQL) and p2 (or CQL), the probabilities of acceptance of the lot under reference plan are given by
Estimation of the probability content in a specified interval using fiducial approach
Published in Journal of Applied Statistics, 2021
Ngan Hoang-Nguyen-Thuy, K. Krishnamoorthy
Mechanical parts are manufactured to meet some tolerance specification limits so that they can be used for their intended purpose. For example, if a shaft is designed to have a ‘sliding fit’ in a hole, the shaft should be little smaller than the hole. Specifically, if a shaft with a nominal diameter of 10 mm is to have a sliding fit within a hole, the shaft might be specified with a tolerance range from 9.964 to 10 mm, and the hole might be specified with a tolerance range from 10.04 to 10.076 mm. Both the shaft and hole sizes will usually form normal distributions.1 In electrical components production, an electrical specification might call for a resistor with a nominal value of 100 ohms, but will also state a tolerance such as ±1%. Thus, in many applications, one needs to assess the percentage of parts that meet the specifications. For example, the acceptance sampling plan, an important statistical method, is commonly used in quality control. In particular, the plan is used to accept/reject a shipment of a product based on some quality characteristics of the parts in a sample from the shipment. Such methods are also used in different stages of production by a manufacturer. If an acceptance sampling plan is based on a continuous variable type data, and it is designed to accept/reject the shipment or a production process on the basis of the percentage of parts satisfy the tolerance specifications, then a confidence interval (CI) or hypothesis test for the true percentage of parts that meet specification is required to implement the acceptance sampling plan.
Design of variables sampling plans based on lifetime-performance index in presence of hybrid censoring scheme
Published in Journal of Applied Statistics, 2019
Ritwik Bhattacharya, Muhammad Aslam
Acceptance sampling plans have been equally applied for the inspection/testing of non-electronic products and electronic products both. The inspection of the lot having non-electronic products is easier than the testing of electronic products for lot sentencing. Currently, electronic devices have less chance of failure over a longer period of time (see [20]). So, for the lot sentencing of such high quality of the product, it may not be possible to wait for the failure of the product. As a consequence of that, it may be possible to use censoring techniques. Type-I and Type-II censoring are commonly used censoring techniques for the inspection of products. In Type-I censoring, the experimental time is fixed, and, in Type-II censoring, the number of failures is fixed for the inspection/testing process of electronic products. Therefore, the acceptance sampling plans based on cumulative information are designed for the testing of high quality product (see [22]). The work on sampling plan using Type-I and Type-II censoring can be seen in [11].