Product Acceptance Determination with Measurement Error Using the Neutrosophic Statistics
The variable data is obtained from the measurement process which is not fully complete or clear in nature due to measurement error. The neutrosophic statistics which is the extension of classical statistics can be applied in the industry for the lot senescing when observations or parameters are uncertain or indeterminate or unclear. In this manuscript, a new sampling plan for the measurement error using the neutrosophic statistics is designed. The proposed sampling plan has two neutrosophic parameters, namely, sample size and acceptance number. The neutrosophic operating function is also given. The neutrosophic plan parameters will be determined through the neutrosophic optimization problem. Some tables are given for some specified parameters. From the comparison study, it is concluded that the proposed sampling plan is more flexible, adequate, and effective in the uncertainty environment as compared to the existing sampling plan under the classical statistics. A real example is given for the illustration purpose.
The achieving of the high quality of the product at a low cost is desired during the production process. The sampling plans have been widely used for inspecting the indeterminate or finished product. Every inspection system is based on a well-designed sampling plan and has been widely used in the industry for the lot sentencing. Using any inspection system, there is a chance for accepting a nonconforming unit or rejecting a confirming unit. Authors in  mentioned that this inspection error should be estimated and a corrective action by industrialist should be taken when it is large. Authors in  designed a sampling plan for inspection error. Authors in  worked on the relationship between inspection error and lot sentencing. According to  “The requirement that the measurement of an individual item does not exceed some specified limit is sometimes more important than the requirement that the mean and variability for the items be at or near some predetermined value”. So, the acceptance sampling plan is an important tool for inspecting whether the measurement of quality interest exceeds the given specification limits or not. The sample selection process is important during the production process as the inspection cost directly depends on the size of a sample. The fate of lot is based on sample information; an accepted lot will be sent to the marker and rejected lots are recertified or bad items are replaced with good items. Several authors designed sampling plans for various aspects to study inspection error, including, for example, [1, 4–15].
In practice, usually, the experimenters are not certain about the proportion of defective product. In this case, for lot sentencing of the product, an approach called the fuzzy sampling plans can be applied for inspection of the product. The fuzzy sampling plans have been widely used in the industry for various situations. Several authors contributed in this area, including, for example, [16–32].
The existing sampling plans with measurement error are designed using the classical statistics. The classical statistics assumes precision or determinate and crisp observations in the measurement process. It was mentioned by  that “all observations and measurements of continuous variables are not precise numbers but more or less nonprecise. This imprecision is different from variability and errors. Therefore also lifetime data are not precise numbers but more or less fuzzy. The best up-to-date mathematical model for this imprecision is so-called nonprecise numbers”. The neutrosophic statistics can be applied when the observations or plan parameters are not clear or precise. Authors in  introduced the neutrosophic statistics and it was applied by [34, 35]. Recently, Aslam (2018) introduced the neutrosophic statistics in the area of acceptance sampling plan. Reference  proposed sampling plan using neutrosophic process loss consideration with indeterminacy in plan parameters. More details about neutrosophic sampling plan can be seen in [37–39].
By exploring the literature and to the best of our knowledge, there is no work on the designing of sampling plan with measurement error using the neutrosophic statistics. In this manuscript, a new sampling plan for the measurement error using the neutrosophic statistics is designed. The proposed sampling plan has two neutrosophic parameters, namely, sample size and acceptance number. The neutrosophic operating function is also given. The neutrosophic plan parameters will be determined through the neutrosophic optimization problem. Some tables are given for some specified parameters. A real example is given for the illustration purpose.
2. Design of Proposed Plan
Suppose that a neutrosophic random variable follows the neutrosophic normal distribution with neutrosophic mean and neutrosophic standard deviation (NSD) . The neutrosophic probability density (npdf) of the neutrosophic normal distribution is defined by  and given byThe corresponding neutrosophic standard normal distribution is defined bySuppose that is neutrosophic random error; the neutrosophic normal distribution of observed measurement . Suppose that be the neutrosophic correlation between true neutrosophic observation and observed neutrosophic observation. By following , is defined bywhere is NSD of observed neutrosophic measurement.
Note here that and are independent, so can be defined byThe correlation between the size of measurement error and is given bywhere .
The quality of interest beyond the upper specification limit (U) or the lower control limit (L) is called the nonconforming item. We suppose that the quality of interest has some unclear or indeterminate observations. Based on the above information, we propose the following sampling plan
Step 1. Specify and calculate .
Step 2. Take a random sample from the submitted lot and compute statistic ; ; , and , where and ; .
Step 3. Accept the lot of the product of ; where is allowed neutrosophic number of defectives.
The proposed neutrosophic plan with measurement error has two neutrosophic plan parameters, namely, sample size and neutrosophic acceptance number . The proposed neutrosophic plan with measurement error is extension of  plan. The proposed neutrosophic plan with measurement error reduces to  when and .
The neutrosophic operating function (NOF) of the proposed neutrosophic plan is derived in the following steps.
According to the operational process of the plan, the lot of product will be accepted if By following [40, 41], ; the NOF of the proposed neutrosophic plan is given byThe producer is interested in using the sampling plan such that the lot acceptance probability should be greater than at acceptable quality level (AQL=), where is producer’s risk. Similarly, the consumer desires that the lot acceptance probability should be smaller than at limiting quality level (LQL=), where is consumer’s risk. To satisfy both parties risk, the neutrosophic operating characteristic curve (NOCC) should pass through the points and . The following neutrosophic optimization problem will be used to find the neutrosophic plan parameters.The purpose of the proposed neutrosophic plan is to find the neutrosophic plan parameters when is minimum. The neutrosophic plan parameters of the proposed plan will be determined by the grid search method. At specified values of AQL and LQL, several combinations of neutrosophic plan parameters and will be obtained by satisfying the conditions of and . But, the neutrosophic plan parameters where are selected and reported in Tables 1–3 for various values of , , , AQL, and LQL.
From Tables 1–3, we note that decreases as increases from 2 to 4 at the same levels of other parameters. Also, the internal between and decreases as LQL increases. We also presented the curves of plan parameters and when , =0.05, LQL=0.001, and in Figures 1, 2, and 3, respectively. From Figures 1–3, we note the decreasing trend in and as AQL values increase. We also note the indeterminacy interval in parameters increases at the higher values of AQL.
3. Comparative Study
The comparison of the proposed sampling plan under the neutrosophic statistics is compared with the sampling plan proposed by  under the classical statistics. Reference  mentioned that, under the uncertainty environment, a method which provides the plan parameters in indeterminacy interval rather than the determined values is known as the most effective and adequate method. We placed the values of a sample size of the proposed sampling plan and from the existing plan in Table 4. From Table 4, we note that, for the proposed sampling plan, the sample size is expressed in indeterminacy interval. For example, when AQL=0.001 and LQL=0.002, under the uncertainty environment, the experimenter can select a random sample between 359 and 720. On the other hand, the existing sampling plan provides the determined value which is 359. From the comparison, we conclude that the proposed sampling plan is more effective and flexible in the uncertainty environment than the plan under the classical statistics.
4. Industrial Example
In this section, we will discuss the application of the proposed sampling plan in the steel industry. In this industry, the measurement of tensile strength (X) is very difficult and costly. It is noted that the hardness (Y) is correlated with tensile strength and easy to study with low cost. As the data is obtained from the measurement process some observations are not precise or unclear. Furthermore, for the testing purpose, the experimenters are not sure about the random sample selection from a lot of the product. Therefore, in the testing of steel product, there is indeterminacy in variable of interest and plan parameters. So, the proposed plan can be applied when there is indeterminacy either in plan parameters or in observations or in both. Suppose that, for this testing, , =0.10, =2, USL=200, AQL=0.001, and LQL=0.02. From Table 1, we have plan parameters and . Based on plan parameters information, the experimenter is decided to select a random sample from the production process. The data showing some indeterminacy in variables of interest is reported in Table 5.
The required statistics is given byThe statistic is computed asThe proposed plan is implemented as follows.
Step 1. Specify and calculate .
Step 2. Take a random sample from the submitted lot and compute statistic .
Step 3. Reject the product as .
5. Concluding Remarks
In this manuscript, a new sampling plan for the measurement error using the neutrosophic statistics is designed. The neutrosophic plan parameters are determined and reported for the industrial application. The application of the proposed plan is shown using the steel data. The proposed plan can be applied in the industry such as in the steel industry and building testing material where the measurement data is not precise. The existing sampling plan, in this case, cannot be applied for the testing of material when indeterminacy is in the observations or parameters or both. The proposed sampling plan using other sampling schemes can be extended for future research.
The data is given in the paper.
Conflicts of Interest
The author declares no conflict of interest regarding this paper.
The authors are deeply thankful to the editor and the reviewers for their valuable suggestions to improve the quality of this manuscript. This work was supported by the Deanship of Scientific Research (DSR), King Abdulaziz University, Jeddah. The author, Muhammad Aslam, therefore, acknowledge with thanks DSR technical support.
E. Jackson, “Effect of inspection errors on waste and on quality of outgoing product,” Industrial Quality Control, vol. 14, no. 6, pp. 5–8, 1957.View at: Google Scholar
T. Diviney and N. David, “A note on the relationship between measurement error and product acceptance,” Journal of Industrial Engineering, vol. 14, no. 4, pp. 218-219, 1963.View at: Google Scholar
E. I. Waziri and O. J. Braimah, “Impact of inspection errors on single, double and chain inspection sampling plans,” Journal of Statistical and Econometric Methods, vol. 5, no. 3, pp. 33–59, 2016.View at: Google Scholar
S. Balamurali, M. Aslam, and A. Liaquat, “Determination of a new mixed variable lot-size multiple dependent state sampling plan based on the process capability index,” Communications in Statistics—Theory and Methods, vol. 47, no. 3, pp. 615–627, 2018.View at: Publisher Site | Google Scholar | MathSciNet
B. Sadeghpour Gildeh, E. Baloui Jamkhaneh, and G. Yari, “Acceptance single sampling plan with fuzzy parameter,” Iranian Journal of Fuzzy Systems, vol. 8, no. 2, pp. 47–55, 2011.View at: Google Scholar
P. Divya, “Quality interval acceptance single sampling plan with fuzzy parameter using poisson distribution,” International Journal of Advancements in Research and Technology, vol. 1, no. 3, pp. 115–125, 2012.View at: Google Scholar
E. B. Jamkhaneh and B. S. Gildeh, Acceptance Double Sampling Plan Using Fuzzy Poisson Distribution 1, 2012.
E. B. Jamkhaneh and B. S. Gildeh, “Sequential sampling plan using fuzzy SPRT,” Journal of Intelligent & Fuzzy Systems: Applications in Engineering and Technology, vol. 25, no. 3, pp. 785–791, 2013.View at: Google Scholar
A. Venkateh and S. Elango, “Acceptance sampling for the influence of TRH using crisp and fuzzy gamma distribution,” Aryabhatta Journal of Mathematics and Informatics, vol. 6, no. 1, pp. 119–124, 2014.View at: Google Scholar
R. Afshari and B. S. Gildeh, “Construction of fuzzy multiple deferred state sampling plan,” in Proceedings of Fuzzy Systems Association and 9th International Conference on Soft Computing and Intelligent Systems (IFSA-SCIS), 2017 Joint 17th World Congress of International, IEEE, 2017.View at: Google Scholar
S. Elango, A. Venkatesh, and G. Sivakumar, A Fuzzy Mathematical Analysis for the Effect of TRH Using Acceptance Sampling Plans, 2017.
R. Afshari, B. Sadeghpour Gildeh, and M. Sarmad, “Fuzzy multiple deferred state attribute sampling plan in the presence of inspection errors,” Journal of Intelligent & Fuzzy Systems: Applications in Engineering and Technology, vol. 33, no. 1, pp. 503–514, 2017.View at: Publisher Site | Google Scholar
F. Smarandache, “Introduction to neutrosophic statistics,” in Infinite Study, 2014.View at: Google Scholar
J. Chen, J. Ye, and S. Du, “Scale effect and anisotropy analyzed for neutrosophic numbers of rock joint roughness coefficient based on neutrosophic statistics,” Symmetry, vol. 9, no. 10, p. 208, 2017.View at: Google Scholar
A. J. Duncan, Quality Control and Industrial Statistics, 1974.View at: MathSciNet