The International journal of analytical and experimental modal analysis ISSN NO: 0886-9367

AN IMPACT OF FUZZY LOGIC ON QUICK SWITCHING SINGLE DOUBLE

SAMPLING PLAN – ACCEPTANCE NUMBER TIGHTENING

R.Nandhinidevi*, G.Uma**, D.Manjula***

* Ph.D Research Scholar, Department of Statistics, PSG College of Arts and Science
Coimbatore – 641014
** Assistant Professor, Department of Statistics, PSG College of Arts and Science
Coimbatore – 641014
*** Ph.D Research Scholar, Department of Statistics, PSG College of Arts and Science
Coimbatore – 641014

Mail ID: amug072000@gmail.com & nilaani1491@gmail.com

Abstract

This paper elaborates the concept of an impact of fuzzy logic on Quick switching single
double sampling plan which contains the Single sampling plan as the Normal plan and the
Double sampling plan as the tightened plan. The implementation of this paper depends upon the
Fuzzy Poisson parameter. The Acceptance number tightening is considered in this paper and the
definitions, OC functions and OC curves are provided. This paper focused on the new sampling
system which helps to solve the situations of Uncertainty and Vagueness by using the Fuzzy
concept.
Keywords: Fuzzy logic, Quick Switching System, Single sampling plan, Double sampling plan,
Operating Characteristic curve, Uncertainty, Probability of Acceptance

Introduction:
This article shows how to evaluate and select quick switching system (QSSs) for
acceptance sampling which introduces a new method of describing the production provided by
sampling plans during periods of changing quality called transitive operating characteristic (OC)
curves. The Quick Switching System explored in this article consists of two sampling plans
named Normal and Tightened plans which is along with a set of rules for switching between
them. The first sampling plan, called the Normal plan, is intended for use during periods of good
quality. It has a smaller sample size which helps to reduce inspection costs. The second sampling
plan, called the Tightened plan which is used mostly when problems are encountered. It is
specially designed to give a high level of protection. The switching rules ensure that the correct
plan is used. One of the main usages of QSS is, it is easy to use and reacts quickly when the
changes occurred in the Quality of a product. QSSs concentrate one‟s inspection effort where it

Volume XII, Issue I, January/2020 Page No:337

The International journal of analytical and experimental modal analysis ISSN NO: 0886-9367

will do the most good. Further, for processes running at low level of inspection but that react
severely to the first hint of a problem.

QSSs were originally proposed by Dodge (1967) and investigated by Romboski (1969)
and Govindaraju (1991). Taylor (1992) contains tables of QSSs and a program to select and
evaluate QSSs with some modifications.

The Quick Switching System to be considered consists of two single sampling plans. But
in this article the new sampling system named QSSDSS is used. In this plan the single sampling
plan is considered as the reference plan for Normal plan and the Double sampling plan is
considered as the reference plan for tightened plan. The QSSs investigated in this article differ
from those originally proposed by Dodge in that the decision to switch is made separate from the
decision to accept/reject. In this article the Fuzzy logic concept is applied to handle the situations
of uncertainty and vagueness present in the parameter.

Review of Literature:

Soundarajan and Arumainayagam have analyzed QSS-r, r=1,2,3 with reference plans as
single sampling plan, double sampling plan, chain sampling plan and repetitive group sampling
plan . Arumainayagam and Uma studied QSS-r, r=1, 2, 3 with three stage multiple sampling plan
as a reference plan. Suresh and Jeyalakshmi used multiple deferred sampling plans (MDS) as
reference plan in QSS-1. Suresh and Kaviyarasu have analyzed QSS-r with conditional repetitive
group (CRGS) sampling plan, two stage CRGS plan and multiple repetitive groups sampling
plan as a reference plans. Arumainaygam and Vennila [21] have applied two different types of
reference plans in QSS-1 and find that the resulting system is more advantages than the system
using same reference plan for normal and tightened inspection. This paper extends this method to
a system involving the switching roles of QSS-3.

Acceptance Sampling and Fuzzy Set Theory:

In different acceptance sampling plans the fraction of defective items, is considered as a
crisp value, but in practice the fraction of defective items value must be known exactly. Many
times these values are estimated or it is provided by experiment. The vagueness present in the
value of p from personal judgment, experiment or estimation may be treated formally with the

Volume XII, Issue I, January/2020 Page No:338

The International journal of analytical and experimental modal analysis ISSN NO: 0886-9367

help of fuzzy set theory. As known, fuzzy set theory is powerful mathematical tool for modeling
uncertain resulting. In this basis defining the imprecise proportion parameter is as a fuzzy
number. With this definition, the number of nonconforming items in the sample has a binomial
distribution with fuzzy parameter. However if fuzzy number p is small we can use the fuzzy
Poisson distribution to approximate values of the fuzzy binomial.

Definitions:
Parameter „p’ (probability of a success) of the crisp Binomial distribution is known
exactly, but sometimes we are failed to obtain the exact value of „p‟, some uncertainty in the
value „p’ and is to be estimated from a random sample or from expert opinion. The crisp Poisson
distribution has one parameter, which we also assume is not known exactly.
Definition 1: The fuzzy subset 𝑁 of real line IR, with the membership function 𝜇𝑁 : 𝐼𝑅 → [0,1] is
a fuzzy number if and only if (a) 𝑁 is normal (b) 𝑁 is fuzzy convex (c) 𝜇𝑁 is upper semi
continuous (d) supp (𝑁) is bounded.
Definition 2: A triangular fuzzy number 𝑁 is fuzzy number that membership function defined by
three numbers 𝑎1 < 𝑎2 < 𝑎3 where the base of the triangle is the interval [𝑎1 , 𝑎3 ] and vertex is at
𝑥 = 𝑎2 .
Definition3: The α - cut of a fuzzy number 𝑁 is a non-fuzzy set defined as
𝑁 𝛼 = {𝑥 ∈ 𝐼𝑅; 𝜇𝑁 (𝑥) ≥ 𝛼}. Hence 𝑁 𝛼 = [𝑁𝛼𝐿 , 𝑁𝛼𝑈 ] where
𝑁𝛼𝐿 = 𝑖𝑛𝑓{𝑥 ∈ 𝐼𝑅; 𝜇𝑁 (𝑥) ≥ 𝛼}
𝑁𝛼𝑈 = 𝑠𝑢𝑝{𝑥 ∈ 𝐼𝑅; 𝜇𝑁 (𝑥) ≥ 𝛼}
Definition 4: Due to the uncertainty in 𝑙𝑖 ’s values we substitute 𝑙𝑖 , a fuzzy number, for each
𝑙𝑖 and assume that 0< 𝑙𝑖 <1 all i . Then X together with the 𝑙𝑖 value is a discrete fuzzy probability
distribution. We write 𝑝 for fuzzy P and we have 𝑃 𝑥𝑖 = 𝑙𝑖
Let A= {𝑥1 , 𝑥2 , … . 𝑥𝑙 } be subset of X. Then define:

𝑙
𝑖=1 𝑙 𝑖
𝑃 𝐴 𝛼 = (1)
𝑠

For 0<α< 1, where stands for the statement “𝑙𝑖 ∈ 𝑘𝑖 [α], 1<i<n, 𝑙𝑖=1 𝑙𝑖 = 1”
This is our fuzzy arithmetic.
Definition 5: Let x be a random variable having the Poisson mass function. If P(x) stands for the
probability that X= x, then
𝑒 −𝜆 𝜆 𝑥
P(x)= (2)
𝑥!
For x=0, 1, 2,….. & λ> 0.

Volume XII, Issue I, January/2020 Page No:339

The International journal of analytical and experimental modal analysis ISSN NO: 0886-9367

Now substitute fuzzy number 𝜆 >0 for λ to produce the fuzzy Poisson probability mass function.
Let P(x) to be the fuzzy probability that X= x. Then α- cut of this fuzzy number as,
𝑒 −𝜆 𝜆 𝑥
𝑃(x)[α]= 𝜆 ∈ 𝜆[α] (3)
𝑥!
For all α ∈ [0,1]. Let X be a random variable having the fuzzy binomial distribution and 𝑃 in
the definition 4 are small. That is all are p ∈ 𝑝 sufficiently small. Then 𝑃[a,b][α] using the fuzzy
poisson approximation. Then,
𝑏 𝑒 −𝜆 𝜆 𝑥
𝑃 [a,b][α]= 𝑥=𝑎 𝑥! (4)

QSSDSS OC Band with Fuzzy Parameter:

Based on the OC function derived by Romboski (1969) the Operating characteristic

function of QSSDSS is,
𝑃𝑇
𝑃𝑎 = (5)
1−𝑃𝑁 +𝑃𝑇

where 𝑃𝑁 =Probability of acceptance under normal sampling plan

𝑃𝑇 = Probability of acceptance under tightened sampling plan
𝑃𝑎 =Probability of acceptance

Quick Switching Single Double Sampling Plan with Fuzzy Parameter:

In a production process, the products have to be inspected using lots with the large size of
„N’, such that the proportion of damaged items is not known precisely. In this situation the
parameter is to be represented with a fuzzy number 𝑝 as follows:
𝑝 = 𝑎1 , 𝑎2 , 𝑎3 , 𝑝 ∈ 𝑝 1 , 𝑞 ∈ 𝑞 1 𝑝 + 𝑞 = 1.
𝜆 = 𝑛𝑝 = (n𝑙𝑖 , na2+n𝑙𝑖 , na3+n𝑙𝑖 )
with which variation of 𝑙𝑖 in the domain of [0, 1-a3], the OC band is plotted according to the
calculation of fuzzy probability.
A single normal and double tightened sampling plan with a fuzzy parameter if defined by
the sample size „n’, and acceptance number c, 𝑐2 , 𝑐4 and if the number of observation defective
product is less than or equal to „c’, the lot will be acceptance.

The implementation of the Normal plan with Reference plan as Single Sampling Plan:
If „N’ is a large number, if 𝑝 is a small, then random variable „d’ has a fuzzy Poisson
distribution with parameter 𝜆 = 𝑛𝑝 So the fuzzy probability for the number of defective items in
a sample size that is exactly equal to „d’
𝑐
𝑒 −𝜆 𝜆 𝑑
𝑃𝑎 = │𝜆 ∈ 𝜆[𝛼] = [𝑃𝐿 𝛼 , 𝑃𝑈 𝛼 ] (6)
𝑑=0 𝑑!
𝑐
𝑒 −𝜆 𝜆 𝑑
𝑃𝐿 𝛼 = 𝑚𝑖𝑛 │𝜆 ∈ 𝜆[𝛼] , (7)
𝑑=0 𝑑!

Volume XII, Issue I, January/2020 Page No:340

The International journal of analytical and experimental modal analysis ISSN NO: 0886-9367

𝑐
𝑒 −𝜆 𝜆 𝑑
𝑃𝑈 𝛼 = 𝑚𝑎𝑥 │𝜆 ∈ 𝜆[𝛼] (8)
𝑑=0 𝑑!

The implementation of the Tightened plan with reference plan as Double Sampling Plan:
The parameters of the Double Sampling Plan is (𝑛1 ,p) and (𝑛2 p) in which p indicates the
fraction of nonconforming items of the lot, however, if the size of the sample be large and p is
small then the random variable 𝐷1 and 𝐷2 has a Poisson approximation distribution with
parameter 𝜆1 = 𝑛1 𝑝 and 𝜆2 = 𝑛2 𝑝. The Probability of acceptance for the Double sampling
plan is the Probability of acceptance on combined samples with 𝑝𝛼 and then the probability of
the lot‟s acceptance in first and second samples by 𝑝𝑎𝐼 , 𝑝𝑎𝐼𝐼 , respectively , then,
𝑝𝑎 = 𝑝𝑎𝐼 + 𝑝𝑎𝐼𝐼 (9)
𝐼
Where 𝑝𝑎 indicates the probability of the event D1 = c1
Thus,
𝑐1
𝑒 −𝜆 1 𝜆 1 𝑑 1
𝑝𝑎𝐼 = │𝜆 ∈ 𝜆[𝛼] 𝑝𝑎𝐼𝐼 = p(𝐷1 + 𝐷2 < 𝑐2 , 𝑐1 < 𝐷1 < 𝑐2 ) (10)
𝑑 1 =0 𝑑1 !

The Fuzzy Probability of Acceptance of the lot in the first and second samples, 𝑝𝑎𝐼 , 𝑝𝑎𝐼𝐼 ,
respectively and
𝑝𝑎 = 𝑝𝑎𝐼 + 𝑝𝑎𝐼𝐼 (11)
OC – Band with Fuzzy Parameter
By the Operating Characteristic curve, one could determine the probability of acceptance
or rejection of a lot having some specific defective items. One can understand the performance of
the acceptance sampling plans and systems. OC curve can be drawn by plotting the probability of
acceptance a lot versus its production quality, which is expressed by the proportion of non-
conforming items in the lot. OC curve gives the clear vision about selection of plans that are
effective in reducing risk and indicates discriminating power of the system.
The fuzzy probability of acceptance a lot in terms of fuzzy fraction of defective items
would be as a band with upper and lower bounds under normal and tightened plans with varied
sample sizes,
𝑐
𝑒 −𝜆 𝜆 𝑑
𝑃𝐿 𝛼 = 𝑚𝑖𝑛 │𝜆 ∈ 𝜆[𝛼] , (12)
𝑑=0 𝑑!

𝑐
𝑒 −𝜆 𝜆 𝑑
𝑃𝑈 𝛼 = 𝑚𝑎𝑥 │𝜆 ∈ 𝜆[𝛼] (13)
𝑑=0 𝑑!

The uncertainty degree of proportion can be identified by the bandwidth of the OC curve.
The less uncertainty value results in less bandwidth, and if proportion parameter gets a crisp
value, lower and upper bounds will become equal, which that OC curve is in classic state.

Volume XII, Issue I, January/2020 Page No:341

The International journal of analytical and experimental modal analysis ISSN NO: 0886-9367

Table 1: Probability of Acceptance for Fuzzy QSSDSS

n = 80, 𝑪𝑵 =1 & 𝑪𝑻𝟏 = 2; 𝑪𝑻𝟐 =4

P 𝑷𝑵 𝑷𝑻 P(a)p
[0,0.01] [1,0.9986] [1,0.9898] [1,0.9986]
[0.01,0.02] [0.9986,0.9763] [0.9898,0.8846] [0.9986,0.9739]
[0.02,0.03] [0.9763,0.9041] [0.8846,0.6729] [0.9739,0.8753]
[0.03,0.04] [0.9041,0.7806] [0.6729,0.4485] [0.8753,0.6715]
[0.04,0.05] [0.7806,0.6288] [0.4485,0.2739] [0.6715,0.4246]
[0.05,0.06] [0.6288,0.4763] [0.2739,0.1585] [0.4246,0.2323]
[0.06,0.07] [0.4763,0.3422] [0.1585,0.0887] [0.2323,0.1188]
[0.07,0.08] [0.3422,0.2351] [0.0887,0.0486] [0.1188,0.0597]
[0.08,0.09] [0.2351,0.1555] [0.0486,0.0263] [0.0597,0.0302]
[0.09,0.1] [0.1555,0.0996] [0.0263,0.0140] [0.0302,0.0153]
[0.1,0.11] [0.0996,0.0621] [0.0140,0.0074] [0.0153,0.0078]

Figure 1: Operating Characteristics curve for Fuzzy QSSDSS

n = 80, 𝑪𝑵 =1 & 𝑪𝑻𝟏 = 2; 𝑪𝑻𝟐 =4

0.8

0.6

0.4

0.2

0
0 0.02 0.04 0.06 0.08 0.1

QSSDSS(Upper) QSSDSS(Lower)

From the above OC band of fuzzy QSSDSS we can identify the state of uncertainty. This
figure represents that when the process quality decrease from a very good state to a moderate
state, then the OC band will be wider. Less bandwidth shows the less uncertainty and the wider
bandwidth shows the high uncertainty. In the Fuzzy Logic, one can ascertain the quality range of
the product in both consumer and producer view.

Volume XII, Issue I, January/2020 Page No:342

The International journal of analytical and experimental modal analysis ISSN NO: 0886-9367

Conclusion:
In this article, construction and designing of construction and designing of Quick
Switching Single Double Sampling System with reference to single sampling plan as Normal
plan and Double Sampling plan as Tightened plan using Fuzzy Poisson Distribution for various
Fuzzy Quality Characteristics are studied. These systems are well defined and then if the fraction
of defective items is crisp they reduce to classical plans. The uncertainty degree of a proportion
parameter is one of the factors of the width of the OC band. The less uncertainty value results in
less bandwidth, and greater uncertainty values results in wider bandwidth.
From this it is suggested that, can adopt this system to predict the uncertainty level.
Based on Fuzzy QSSDS System, one can achieve better outcome with minimum sampling cost
and time. Fuzzy QSSDSS helps to increase the protection level without increasing costs. Fuzzy
QSSDSS is a versatile and valuable new tool for one‟s inspection program in shop floor
situations.
Reference:

[1] AMERICAN NATIONAL STANDARDS, Institute/American Society for Quality Control

(ANSI/ASQC) STANDARD A2 (1987). “Terms, symbols and Definition for Acceptance
Sampling”, American Society for Quality Control, Milwaukee, Wisconsin. USA.
[2] Bahram Sadeghpour-Gildeh, Gholamhossein Yari, Ezzatallah Balouli Jamkhaneh (2008);
Acceptance double sampling plan with fuzzy parameter, Proceedings of the 11th Joint
Conference on Information Sciences.
[3]Cameron J.M., (1952). Tables for constructing and computing the operating characteristics of
single sampling plan, Industrial Quality Control, Vol.9, N0.1, pp. 37-39.
[4] Devaraj Arumainayagam S and Uma G (2009). Construction and selection of quick switching
multiple sampling system-sample size tightening, Indian Association Productivity and Quality
Journal, 2009, vol.34, No.2.
[5] Devaraj Arumainayagam S and Uma G (2013). Construction and selection of quick switching
system using weighted Poisson distribution-sample size tightening, Indian Association
Productivity and Quality Journal, 2009, vol.34, No.2.
[6] Arumainayagam, S. Devaraj; Uma, G., (2009). Construction and Selection of Quick
Switching System Using Weighted Poisson Distribution, International Journal of Statistics &
Systems, Vol. 4, Issue 2, pp127.
[7]Devaraj Arumainayagam, S. (1991). Contributions to the study of Quick Switching System
(QSS) and its Applications. Doctoral Dissertation, Bharathiar University, Coimbatore, Tamil
Nadu, India.
[8] D. C. Montgomery (1991): Introduction to Statistical Quality Control, Wiley, New York
[9] Divya. P. R, (2012): Quality Interval Acceptance Single Sampling Plan with fuzzy parameter
using Poisson Distribution, International Journal of Advancements in Research and Technology,
1(3).
[10] Dodge, H.F. and H.G. Romig, (1956), Sampling inspection tables, John Wiley Co., New
York, U.S.A.
[11] Dodge, H.F. (1967). A new dual system of acceptance sampling, Technical Report 16 (New
Brunwick, NJ, The statistics center, Rutgers – The State University).

Volume XII, Issue I, January/2020 Page No:343

The International journal of analytical and experimental modal analysis ISSN NO: 0886-9367

[12] Duncan, A.J., (1986). Quality Control and Industrial Statistics, 5th Edition, Homewood,
IIIinois, Richard D. Irwin.
[13] E. G. Schiling (1982): Acceptance Sampling Quality Control, Dekker, New York.
[14] F.Tamaki, A.Kanagawa, Ohta.H, (1991), A fuzzy design of sampling inspection plans by
attributes, Japanese journal of fuzzy theory and systems, 3, 315 – 327.
[15] Ezzatallah Balouli Jamkhaneh, Bahram Sadeghpour-Gildeh, Gholamhossein Yari (2009);
Acceptance single sampling plan with fuzzy parameter with the using of poisson distribution,
World Academy of Science, Engineering and Technology, 3, 01-20.
[16] Hamaker, H.C (1950). The theory of sampling inspection plans, Philips Technical Review,
11, pp.260-270.
[17] A. Kanagawa and H. Ohta (1990), A design for single sampling attribute plan based on
fuzzy sets theory, Fuzzy Sets and Systems, 37, 173-181.
[18] Nandhinidevi.R and Uma.G (2018), An impact of fuzzy on Quick Switching System by
attributes under Poisson Distribution, International Journal of Pure and Applied Mathematics,
118, No.24
[19] T. K. Chakraborty (1992), A class of single sampling plans based on fuzzy optimization,
Opsearch, 29(1), 11-20.
[20]T. K. Chakraborty (1994a): Possibilistic parameter single sampling inspection plans,
Opsearch, 31(2), 108-126.
[21]Ohta. H and H. Ichihashi, (1988), Determination of single sampling-attributes plans based on
membership functions, International Journal of Production Research, 26(9), 1477 – 1485
[22] Romboski, L.D. (1969). An investigation of quick switching acceptance sampling systems,
Doctoral Dissertation, (New Brunwick, NJ, The statistics center, Rutgers – The State University)
[23] Soundararajan V & Arumainayagam S D (1995). Construction and Selection of quick
switching system sample size tightening, Journal of applied Statistics, Vol.22, No.2, 1995, pp.
105-119.
[24] Uma, G.; Arumainayagam, S. Deveraj (2008). Construction of Matched Quick Switching
Single Double and Multiple Sampling Systems, International Journal of Statistics & Systems;
Vol. 3,Issue2,pp.147.
[25] Vennila J and Devaraj Arumainayagam (2017). Construction and Selection of Quick
Switching System Using Normal Single Sampling Plan and Tightened Double Sampling Plan,
International Journal of Applied Mathematics and Statistics, Vol.56, Iss. No.6, 2017, pp. 124-
137.

Volume XII, Issue I, January/2020 Page No:344