Received 19 July 2023, accepted 30 August 2023, date of publication 5 September 2023, date of current version 13 September 2023.

Digital Object Identifier 10.1109/ACCESS.2023.3312226

Design of Risk-Based Univariate Control Charts

With Measurement Uncertainty
ATTILA I. KATONA 1, AAMIR SAGHIR 2, CSABA HEGEDŰS 3, AND ZSOLT T. KOSZTYÁN 1
1 Department of Quantitative Methods, University of Pannonia, 8200 Veszprém, Hungary
2 Department of Statistics, Mirpur University of Science and Technology (MUST), Mirpur, Azad Jammu and Kashmir 10250, Pakistan
3 Department of Supply Chain Management, University of Pannonia, 8200 Veszprém, Hungary

Corresponding author: Zsolt T. Kosztyán (kosztyan.zsolt@gtk.uni-pannon.hu)

This work was supported by the Ministry of Culture and Innovation of Hungary from the National Research, Development and Innovation
Fund, financed through the 2021 Thematic Excellence Programme Funding Scheme, under Project TKP2021-NVA-10. The work of Attila
I. Katona was supported by the ÚNKP-22-4 New National Excellence Program of the Ministry for Culture and Innovation from the Source
of the National Research, Development and Innovation Fund.

ABSTRACT Control charts are widely used tools of statistical process control (SPC). They are effective tools
for indicating process shifts, helping avoid an increased number of defective products in the manufacturing
environment. However, their effectiveness is strongly dependent on the performance of the measurement
system. In the scientific literature, numerous studies have examined the effect of measurement errors in
statistical process control, but only a few of them have focused on developing risk-based control charts
that consider the cost of decisions during process control; furthermore, they were developed only for the
case of multivariate and adaptive control charts such as T2 and VSSI (Variable sample Size and Sampling
Interval) X-bar charts. This paper fills this gap by using the proposed risk-based concept for commonly used
control charts such as X-bar, Moving Average (MA), Exponentially Weighted Moving Average (EWMA)
and Cumulative Sum (CUSUM) charts. The effectiveness of the developed risk-based control charts is
demonstrated by simulations and real-life datasets.

INDEX TERMS Control charts, consumer risk, risk-based approach, decision cost, EWMA chart.

I. INTRODUCTION Therefore, the development of new methods of decreasing the

Statistical quality control (SQC) is a widely used method number of incorrect decisions is an important area of SQC
in operation management. Conformity control and process research that requires further exploration.
control are important means of controlling a process or pro- As a solution to the abovementioned issues, a risk-based
duction method. The effectiveness of applied quality control method was designed by [8] considering all four decision
methods depends on the performance of the measurement outcomes to minimize the risk of false decisions under param-
system. Measurement uncertainty in a system can lead to eter uncertainty. Risk-based control charts use simulation
incorrect decisions such as unnecessary manufacturing stop- and optimization to select the optimal values of the chart
pages or missed interventions [1]. Therefore, the rate of parameters to minimize the risks arising from incorrect deci-
producer and customer risk is strongly impacted by the per- sions related to process control. Contrary to Shewhart control
formance of the measurement process [2]. charts, they consider the effect of measurement errors and
Traditional methods do not account for measurement count with the costs of each decision outcome during the
uncertainty [3], [4], [5]. Several researchers have examined control chart design. The results revealed that a risk-based
the effect of measurement uncertainty in conformity test- approach was highly effective in reducing decision costs,
ing [6], [7], [8], [9], [10] and statistical process control [11], even under a nonnormal measurement error distribution.
[12], [13], [14], [15], [16], [17], [18], [19], [20], [21], [22]. Later, [7] and [9] studied a risk-based approach in confor-
mity testing, and [23], [24], and [25] designed risk-based
The associate editor coordinating the review of this manuscript and control charts for statistical process control. Using industrial
approving it for publication was Amjad Mehmood . datasets, [5] validated the applicability of risk-based methods

This work is licensed under a Creative Commons Attribution 4.0 License. For more information, see https://creativecommons.org/licenses/by/4.0/
VOLUME 11, 2023 97567
 A. I. Katona et al.: Design of Risk-Based Univariate Control Charts

in conformity testing and for control charts. The results of of the traditional control chart parameters, (2) estimation
all these studies reveal the applicability of the risk-based of the decision costs, (3) calculation of the total cost and
approach in minimizing the decision cost. The aforemen- (4) optimization of the control lines.
tioned studies showed that risk-based control charts are
important tools for statistical process control since they could A. CALCULATION OF CONTROL CHART PARAMETERS
achieve a 5-13% total decision cost reduction in the case The first step is to calculate the traditional control chart
of process control and a nearly 5% increase in total profit parameters, such as the centerline, control lines and plotted
during industrial conformity control. Their importance is also data points.
supported by the ability to greatly reduce prestige loss due to X̄ chart
the elimination of error type II. In the case of the X̄ chart, the sample statistics for the ith
Although the results of these studies provide great poten- sample can be calculated as follows [32]:
tial for significant cost and risk reduction in manufacturing n
1X
processes, they focus on complex process structures only, X̄i = Xij (1)
such as multivariate and adaptive control charts. Never- n
j=1
theless, the need for monitoring univariate processes with
where X̄i is the sample mean, n is the sample size and Xij
fixed sample sizes and intervals, such as the application
denotes the jth observation of sample i. The control lines can
of Shewhart X̄ , MA, EWMA and CUSUM charts, is com-
be expressed as follows:
mon. Several studies have noted the common usage of She-
whart and memory-based univariate charts in multiple fields, kx σ
UCLX̄ = µ + √ (2)
e.g., healthcare [26], high-quality processes [27], wastew- n
ater treatment [28], electronics [29] and financial applica- kx σ
LCLX̄ = µ − √ (3)
tions [30]. Although [31] proposed a joint optimization of n
control charts and preventive maintenance policies, this solu- where µ is the process mean (expected value), σ is the
tion does not address the effect of measurement errors during process standard deviation, kx σ is the shift regarding the
the optimizations. Therefore, the lack of a univariate ver- process mean expressed in standard deviation units and
sion of the risk-based control charts is a huge gap since it UCLX̄ and LCLX̄ denote the upper and lower control lines,
strongly weakens the practical applicability of the developed respectively.
risk-based approach and prevents many practitioners from MA chart
benefiting from the cost minimization potential. The MA chart statistic can be calculated as follows [33]:
This paper fills this gap by extending the risk-based  Pi
approach to the commonly used univariate control charts. X̄j
 j=1 , for i < w


There are two aims of this article: MAi = Pi i (4)
X̄
A1 to design Shewhart X̄ , MA, EWMA and CUSUM  j=i−w+1 j , for i ≥ w


charts in the presence of measurement uncertainty w
for monitoring the stability of a process in terms of where X̄j (i, j = 1, 2, . . . , m) is the average of the jth observa-
the average (µ) and tion, m is the number of observations, w is the span parameter
A2 to validate the results via simulation and real-life and MAi is the (moving) average of the ith sample as a control
data. chart statistic. The control lines can be calculated based on the
The main contributions of this paper can be determined as following equations:
follows: km σ

 µ + √ , for i < w


C1 The practical applicability of the risk-based concept ni
UCLMA = (5)
is greatly increased due to the extension with uni- km σ
 µ + √ , for i ≥ w


variate cases nw
C2 The cost consequences of decisions of traditional k σ

m
 µ − √ , for i < w

and risk-based charts are studied 
ni
C3 The proposed risk-based control charts are vali- LCLMA = (6)
km σ
 µ − √ , for i ≥ w

dated using simulated and real-life examples. 
nw
The rest of this article is organized as follows: Section
where UCLMA and LCLMA are the upper and lower control
II describes the methods designed in this work. The results
lines of the MA chart, respectively.
are presented in Section III. Finally, the article ends with a
EWMA chart
summary and conclusion (Section IV).
In the case of the EWMA chart, the exponentially
smoothed moving average statistic can be calculated as fol-
II. METHODS
lows [34], [35]:
Following the process proposed by Ref. [25], risk-based con-
trol charts can be designed based on four steps: (1) calculation Ei = λX̄i + (1 − λ)Ei−1 (7)

97568 VOLUME 11, 2023

A. I. Katona et al.: Design of Risk-Based Univariate Control Charts

where Ei is the sample statistic for sample i and λ, description of the decision cost estimation, the reader is
0 < λ ≤ 1 is the smoothing parameter. The control lines can referred to [24], where the main components are described.
be determined in the following way: In the following, a more detailed description of each cost
s component is also provided based on [24] to demonstrate the
λ estimation related to the costs of the decision outcomes.
UCLEWMA = µ + ke σ [1 − (1 − λ)2i ] (8)
n(2 − λ) Cost of correct acceptance (c11 ): The cost of correct accep-
tance consists of all the production and inspection costs. The
s
λ
LCLEWMA = µ − ke σ [1 − (1 − λ)2i ] (9) production costs include labor, material costs, rent, operating
n(2 − λ) and indirect costs as well. The inspection costs are the operat-
CUSUM chart ing cost of the measurement device, labor and sampling costs.
In the CUSUM scheme, the positive and negative changes If c11 is considered the cost of the default case, its value is
in the process mean are calculated as follows [36]: usually set to 1. In this way, the other three decision costs can
be relative to c11 , which makes the interpretation easier.
Ci+ = max{0, Ci−1
+
+ Xi − µ − zσ } (10) Cost of correct control (c00 ): The two main costs arising in
Ci− = −
min{0, Ci−1 + Xi − µ + zσ } (11) this case are the cost of control and the cost of restart. In this
case, the cost of control also includes the cost of root cause
where Ci+and Ci−
denote the upper and lower cumula- search and the maintenance cost of the manufacturing device.
tive sums, respectively, and z represents the detectable shift These can be estimated considering the mean stoppage- and
expressed relative to the process standard deviation. The maintenance times.
upper and lower control lines are described in the following Cost of incorrect control (c10 ): The cost of incorrect control
forms: contains the same cost components as it is observable by c00 .
UCLCUSUM = hc σ (12) However, in this case the overregulation of the process can
LCLCUSUM = −hc σ (13) be considered an additional cost optionally to assign more
weight to the type I error.
where hc is the number of standard deviations considered Cost of missed control (c01 ): If the control chart does not
during the control line calculation; its value is usually detect the existing process shift (type II error), the manufac-
set to 4 or 5. turing device produces scraps until the next stoppage. In this
case, the mean cost of scraps should be estimated and added
B. ESTIMATION OF DECISION COSTS to the production and inspection costs.
To determine the cost of decisions, the decision outcomes As mentioned previously, the estimation of the different
must first be interpreted. Since the real value of the moni- cost components, for example, the mean cost of scraps or the
tored product characteristic (x) is distorted by measurement mean cost of root cause search, can be supported by the usage
error (e), each observed value (y) can be expressed as follows: of an ERP system.
yi = xi + ei (14)
C. TOTAL COST CALCULATION
Due to the presence of measurement error, four different During risk-based control chart design, the total cost of
decision outcomes can be defined during process control: decisions (TC) is the output of the objective function to
(1) correct acceptance, (2) type I error, (3) type II error and be minimized. Opposed to the economic design of control
(4) correct control. Table 1 demonstrates the structure of charts [37] here, all the four states are considered in the
decision outcomes according to the observed and real product total cost: letting the controlled process run, investigating for
characteristics, where c11 is the cost of correct acceptance, c10 assignable cause when the process gets out of control and the
is the cost associated with type I error or incorrect control, two – type I and type II – decision errors. Based on [24], the
c01 is the cost of type II error or incorrect acceptance, and c00 total decision cost can be calculated by aggregating the cost
denotes the cost of correct control when the out-of-control of each decision outcome:
state is correctly detected.
TC = C11 + C10 + C01 + C00 = q11 c11
TABLE 1. Decision outcomes of statistical process control (based on [9]). + q10 c10 + q01 c01 + q00 c00 (15)

where Cab is the aggregate cost of a specific decision outcome

and qab is the number of times each case occurred during
control. a = 1 if the real product characteristic is in control
and 0 otherwise, and b = 1 if the observed product charac-
teristic is in control and 0 otherwise.
Each decision cost can be estimated by considering the Type I. and type II. errors are identified by comparing
cost components from the ERP system (Enterprise Resource the plotted control chart statistics (observed and real) to the
Planning system) of the manufacturer. For a more detailed control lines calculated considering the real and measurement

VOLUME 11, 2023 97569

 A. I. Katona et al.: Design of Risk-Based Univariate Control Charts

error distorted processes. The cost of the decision outcome is

calculated for each plotted control chart statistic.

D. OPTIMIZATION OF THE CONTROL LINES

To modify the control lines, the total cost of decisions (TC)
is minimized by optimizing the control coefficient by each
control chart. Let k represent a one-dimensional vector con-
taining the correction coefficient for each control chart and
k = [kx , km , ke , hc ]. Then, the optimization problem is for-
mulated as follows:
For each k(i):
minimize TC(k(i))
k(i)
subject to k(i) > 0
∀k(i) ∈ R (16)
where k(i) represents the ith element of vector k. The FIGURE 1. The steps conducted in each iteration during the simulation.
Nelder–Mead simplex search is applied to solve the optimiza-
TABLE 2. Simulation parameters.
tion problem. Only the control lines are optimized in this
paper but as shown in [25], the method can be combined with
the optimal setting of the sample size and sampling interval,
as is done in economic design.

III. RESULTS
In this section, first, the simulation results are presented to
examine the applicability of the risk-based approach related
to the aforementioned additional univariate control chart
types. The optimal parameter of any risk-based control chart
has been determined using equation (16) via the Nelder-Mead
simplex method. In this optimization, two different scenarios
have been considered. First, a simulations study is conducted
to obtain the optimal parameter of the risk-based univariate
control chart in subsection (A). Second, real-life data are used software such as Minitab [38]. The EWMA weighting param-
to calculate the optimal parameter of the risk-based univariate eter λ is set to 0.2 based on the recommendation given by [39],
control chart in subsection (B). namely that the EWMA chart is effective in the detection of
small shifts when 0.05 < λ < 0.25. The minimum detectable
A. SIMULATION RESULTS mean shift (z) by the CUSUM chart is set to 1 based on the
The simulation was conducted starting with random number recommendation of [40]. Regarding the decision costs (c11 ,
generation. Figure 1 demonstrates the steps of one iteration c10 , c01 , c00 ), the cost structure proposed by [24] was applied
during the simulation process. The equations used in each in the simulation.
step are typed in red. The results were visualized with boxplot diagrams for each
A process consisting of 1000 samplings with a sample size control chart type, as shown in Fig. 2. The vertical axes denote
n = 1 was generated following a normal distribution with the distribution and median of the total decision cost values
expected value µx = 10 and standard deviation σx = 0.5. for the generated processes before and after optimization
Then, measurement errors were added to the process follow- (i.e., for the original and risk-based control charts).
ing a normal distribution with expected value µe = 0 and As Fig. 2 shows, a decrease in the total decision cost is
standard deviation σe = 0.05 (i.e., two vectors are gener- achievable for all control chart types by applying the pro-
ated, one for the process and one for measurement errors.). posed optimization. Notably, higher total costs are observable
A control chart was designed for the generated process, and in the case of the CUSUM control chart since to generate out-
the control lines were optimized to achieve the minimal total of-control situations, an increasing trend must be added to the
decision cost. The process and measurement error genera- process (the shift size is denoted by sCUSUM ). Therefore, the
tion, control chart design and optimization were iterated one increasing trend causes more frequent out-of-control states,
hundred times for each control chart type. Table 2 shows a resulting in higher decision costs.
detailed list of the input parameters of the simulation. Table 3 shows the control chart schemes before and after
As Table 2 shows, the lag parameter of the MA chart (w) the optimization of the control coefficients. Since the pro-
was set to w=3 as the default value of widely used statistical cess generation and chart optimization were conducted one

97570 VOLUME 11, 2023

A. I. Katona et al.: Design of Risk-Based Univariate Control Charts

TABLE 4. Decision outcomes and total decision cost in the simulation.

FIGURE 2. Simulation results.

hundred times, UCL, LCL, CL and the coefficient parameters

show average values. In the case of the other control chart
parameters such as n, λ, z and w, the values remained constant
during the simulation.

TABLE 3. Control chart schemes in the simulation.

FIGURE 3. Control charts reflecting the real and observed processes.

The applicability of the proposed risk-based control charts

was also tested on a real-life dataset.

B. RESULTS ON A REAL DATASET

As Table 3 shows, the average values of the chart coef- The proposed risk-based univariate control charts were tested
ficients decreased after optimization for all control chart on a real dataset constructed by [5]. In the dataset, the cut-
types, except for the CUSUM chart. The reason for this is ting length of brake cylinders was monitored as a product
as follows: in the case of the X-bar, MA and EWMA control characteristic. The dataset contains 50 measured values. Each
charts, the average numbers of type I and type II errors are part was measured twice, first with a manual height gauge
balanced, and since type II errors have significantly more (used in production) and then by a 3D high-precision device
serious consequences (meaning significantly higher decision to obtain measurements representing the observed and ‘‘real’’
costs), the method narrows the control region to avoid incor- product characteristic values, respectively. Due to the short
rect acceptance. On the other hand, since an increasing trend length of this process, a sample size n = 1 was applied in the
was added to the process in the CUSUM chart case, the case of all four charts, and the control lines were estimated
number of out-of-control samples also increased, indicating taking the first 25 observations into consideration, since sev-
a higher probability of committing a type I error. Therefore, eral software packages use 25 as the default value. Fig. 3
the proposed method broadens the control interval to avoid shows the control charts without optimization. The black
type I errors. Table 4 shows the average count of each decision lines represent the ‘‘real’’ process, and the red lines denote
outcome and the average total decision costs before and after the ‘‘observed’’ sample values. The dashed lines denote the
optimization. control lines based on the real and observed data. Finally, the
As Table 4 shows, the risk-based control charts outperform occurrence of type I errors is indicated with blue dots.
the traditional charts in terms of total decision cost. The As Fig. 3 shows, the application of a manual height
numbers of type I and type II errors decreased after the gauge indicates measurement errors, and type I errors can
optimization in every case. be observed during the process monitoring in all four cases.

VOLUME 11, 2023 97571

 A. I. Katona et al.: Design of Risk-Based Univariate Control Charts

Table 5 shows the applied control chart schemes before and in complex cases, such as multivariate and variable-sample-
after the optimization. size control charts but also in simple univariate cases.
If the costs of decision errors, such as incorrect accep-
TABLE 5. Control chart schemes on real dataset. tance and incorrect rejection, are enumerated or at least
their proportions are specified, the total decision cost can
be significantly reduced (A2 ) for both memory-based and
nonmemory-based control charts.
The proposed extensions work not only in simulated but
also in real-life situations (A2 ). Therefore, the optimization
of the risk-based univariate control chart is determined via
simulation as well as using a real life dataset. Moreover,
we believe that the reduction in total decision costs easily
compensates for the cost of the implementation of risk-based
control charts.
The contributions of this paper can be summarized as
As Table 5 demonstrates, the control coefficients are follows. First, the applicability of the proposed risk-based
increased after optimization in all cases. In this practical approach is significantly increased by the development of
example, the measurement gauge usually overmeasures the univariate RB control charts (C1 ) The proposed univariate
real product characteristic, which leads to type I errors. control charts opens new room for software development
Therefore, the risk-based chart relaxes the control interval since they can be implemented in new software packages
(through the increase in the coefficient) to eliminate type I written in program languages such as R or Python. Both plat-
errors. For the parameters n, λ, z and w, the same values forms offer easily accessible and free packages, which makes
were used as in the simulation. Table 6 depicts the decision the application easier for practitioners. Second, demonstrat-
outcomes and total decision cost on the real dataset. ing that the total decision cost can be decreased signifi-
cantly in the case of the average control charts indicates
TABLE 6. Decision outcomes and total decision cost on real dataset. that manufacturers working with univariate processes (and
fixed samples) can benefit from the risk- and total decision
cost reduction (C2 ). Third, the validation of the proposed
risk-based charts on a real dataset supports the necessity of
the practical application of the proposed approach (C3 ).
As a next step, we plan to implement the proposed method-
ology as an R package that provides all the functions needed
to design risk-based control charts.

REFERENCES
[1] L. R. Pendrill, ‘‘Operating ‘cost’ characteristics in sampling by vari-
able and attribute,’’ Accreditation Qual. Assurance, vol. 13, no. 11,
pp. 619–631, Nov. 2008.
[2] R. S. Peruchi, P. P. Balestrassi, A. P. de Paiva, J. R. Ferreira, and
All the risk-based univariate control charts were able to M. D. S. Carmelossi, ‘‘A new multivariate gage R&R method for corre-
eliminate type I errors by increasing the control interval lated characteristics,’’ Int. J. Prod. Econ., vol. 144, no. 1, pp. 301–315,
Jul. 2013.
(indicated by the increase of the control chart coefficients)
[3] B. C. Giri, ‘‘Managing inventory with two suppliers under yield uncer-
in this real-life example. Furthermore, the total decision cost tainty and risk aversion,’’ Int. J. Prod. Econ., vol. 133, no. 1, pp. 80–85,
decreased in all cases. Sep. 2011.
[4] I. Lira, ‘‘A Bayesian approach to the consumer’s and producer’s risks in
measurement,’’ Metrologia, vol. 36, no. 5, pp. 397–402, Oct. 1999.
IV. SUMMARY AND CONCLUSION [5] A. I. Katona, ‘‘Validation of risk-based quality control techniques: A case
In an increasingly industrialized environment, the accuracy study from the automotive industry,’’ J. Appl. Statist., vol. 49, no. 12,
of measurements has become a key issue. It is important pp. 1–20, Jun. 2021.
[6] A. B. Forbes, ‘‘Measurement uncertainty and optimized conformance
to detect and account for measurement uncertainty when assessment,’’ Measurement, vol. 39, no. 9, pp. 808–814, Nov. 2006.
making decisions. Therefore, it is very important to develop [7] C. Hegedűs and Z. T. Kosztyán, ‘‘The consideration of measurement uncer-
methods in process control that account for measurement tainty in forecast and maintenance related decisions,’’ Problems Manage.
21st Century, vol. 1, no. 1, pp. 46–59, May 2011.
uncertainty and the consequences of decisions. The main goal
[8] Z. T. KosztyÁN, T. Csizmadia, C. HegedŰs, and Z. KovÁCs, ‘‘Treat-
of this study was to propose a new family of univariate control ing measurement uncertainty in complete conformity control system,’’ in
charts. To do this, three goals were pursued and achieved. Innovations and Advances in Computer Sciences and Engineering. Cham,
A family of univariate risk-based control charts was pro- Switzerland: Springer, 2010, pp. 79–84.
[9] Z. T. Kosztyán, C. Hegedűs, and A. Katona, ‘‘Treating measurement
posed to handle measurement uncertainty (A1 ). It was impor- uncertainty in industrial conformity control,’’ Central Eur. J. Oper. Res.,
tant to demonstrate that risk-based extension works not only vol. 25, no. 4, pp. 907–928, Dec. 2017.

97572 VOLUME 11, 2023

A. I. Katona et al.: Design of Risk-Based Univariate Control Charts

[10] L. R. Pendrill, ‘‘Using measurement uncertainty in decision-making [36] R. B. Crosier, ‘‘A new two-sided cumulative sum quality control scheme,’’
and conformity assessment,’’ Metrologia, vol. 51, no. 4, pp. 206–218, Technometrics, vol. 28, no. 3, pp. 187–194, Aug. 1986.
Aug. 2014. [37] V. Vommi, Murty, and S. N. Seetala, ‘‘A simple approach for robust
[11] B. Abraham, ‘‘Control charts and measurement error,’’ in Proc. Annu. Tech. economic design of control charts,’’ Comput. Oper. Res., vol. 34, no. 7,
Conf. Amer. Soc. Quality Control, vol. 31, 1977, pp. 370–374. pp. 2001–2009, Jul. 2007.
[12] F. Asif, S. Khan, and M. Noor-Ul-Amin, ‘‘Hybrid exponentially weighted [38] Minitab Statistical Software. Accessed: Aug. 25, 2023. [Online]. Avail-
moving average control chart with measurement error,’’ Iranian J. Sci. able: https://www.minitab.com/en-us/products/minitab/
Technol., Trans. A, Sci., vol. 44, no. 3, pp. 801–811, Jun. 2020. [39] J. M. Lucas and M. S. Saccucci, ‘‘Exponentially weighted moving average
[13] X. Hu, P. Castagliola, J. Sun, and M. B. C. Khoo, ‘‘The performance of control schemes: Properties and enhancements,’’ Technometrics, vol. 32,
variable sample size chart with measurement errors,’’ Quality Rel. Eng. no. 1, pp. 1–12, Feb. 1990.
Int., vol. 32, no. 3, pp. 969–983, Apr. 2016. [40] V. V. Koshti, ‘‘Cumulative sum control chart,’’ Int. J. Phys. Math. Sci.,
[14] K. W. Linna, W. H. Woodall, and K. L. Busby, ‘‘The performance of vol. 1, no. 1, pp. 28–32, Oct. 2011.
multivariate control charts in the presence of measurement error,’’ J. Qual.
Technol., vol. 33, no. 3, pp. 349–355, Jul. 2001.
[15] P. E. Maravelakis, ‘‘Measurement error effect on the CUSUM control ATTILA I. KATONA is currently a Researcher
chart,’’ J. Appl. Statist., vol. 39, no. 2, pp. 323–336, Feb. 2012. with the Department of Quantitative Methods,
[16] H. Sabahno, A. Amiri, and P. Castagliola, ‘‘Performance of the variable Faculty of Business and Economics, University
parameters X control chart in presence of measurement errors,’’ J. Test. of Pannonia, and a former Data Scientist with
Eval., vol. 47, no. 1, Jan. 2019, Art. no. 20170527. IBM. His experiences associate with manage-
[17] A. Saghaei, S. M. T. F. Ghomi, and S. Jaberi, ‘‘Economic design of ment science and operations research and big
exponentially weighted moving average control chart based on measure- data approaches. He has valuable publications in
ment error using genetic algorithm,’’ Qual. Rel. Eng. Int., vol. 30, no. 8, Expert Systems With Applications and Computers
pp. 1153–1163, Dec. 2014. and Industrial Engineering journals. His research
[18] K. P. Tran, P. Castagliola, and G. Celano, ‘‘The performance of the interests include the treatment of decision risks
shewhart-RZ control chart in the presence of measurement error,’’ Int. arising from measurement uncertainty in quality control, and statistical
J. Prod. Res., vol. 54, no. 24, pp. 7504–7522, Dec. 2016. control chart design. He was a recipient of the Pro Scientia Golden Medal
[19] F. S. Zaidi, P. Castagliola, K. P. Tran, and M. B. C. Khoo, ‘‘Performance of from the Hungarian Academy of Sciences.
the MEWMA-CoDa control chart in the presence of measurement errors,’’
Qual. Rel. Eng. Int., vol. 36, no. 7, pp. 2411–2440, Nov. 2020.
[20] L.-P. Chen and S.-F. Yang, ‘‘A new p-control chart with measurement error AAMIR SAGHIR received the Ph.D. degree from
correction,’’ Qual. Rel. Eng. Int., vol. 39, no. 1, pp. 81–98, Feb. 2023. Zhejiang University, China, in 2014. He is cur-
[21] M. R. Maleki, A. Salmasnia, and S. Y. Saber, ‘‘The performance of triple rently an Associate Professor with the Department
sampling X control chart with measurement errors,’’ Qual. Technol. Quant. of Statistics, Faculty of Natural and Applied Sci-
Manage., vol. 19, no. 5, pp. 587–604, Sep. 2022. ences, Mirpur University of Science and Technol-
[22] M. R. Maleki, B. Shamseddin, H. Eghbali, and A. Bazdar, ‘‘The effect of ogy (MUST), Pakistan. He is also a member of the
gauge measurement errors on double sampling X̄ control chart,’’ Commun. Department of Quantitative Methods, Faculty of
Statist. Theory Methods, vol. 52, no. 8, pp. 2702–2717, Apr. 2023. Business and Economics, University of Pannonia,
[23] C. Hegedüs, Z. T. Kosztyán, and A. Katona, ‘‘Parameter drift in risk-based Hungary. His research interests include statistical
statistical control charts,’’ Global J. Technol., vol. 3, no. 1, pp. 1360–1366, process control methods, probability distributions,
2013.
and the analysis of big data.
[24] Z. T. Kosztyán and A. I. Katona, ‘‘Risk-based multivariate control chart,’’
Expert Syst. Appl., vol. 62, pp. 250–262, Nov. 2016.
[25] Z. T. Kosztyán and A. I. Katona, ‘‘Risk-based X-bar chart with vari- CSABA HEGEDŰS is currently an Associate Pro-
able sample size and sampling interval,’’ Comput. Ind. Eng., vol. 120, fessor and the Head of the Department of Sup-
pp. 308–319, Jun. 2018. ply Chain Management, University of Pannonia.
[26] G. Suman and D. Prajapati, ‘‘Control chart applications in healthcare: He has published papers in Expert System with
A literature review,’’ Int. J. Metrol. Qual. Eng., vol. 9, p. 5, May 2018. Applications and Operations Research Perspec-
[27] S. Ali, A. Pievatolo, and R. Göb, ‘‘An overview of control charts for high- tives. He has carried out research, development,
quality processes,’’ Qual. Rel. Eng. Int., vol. 32, no. 7, pp. 2171–2189, and innovation projects with members of the
Nov. 2016. industry. His research interest includes quantita-
[28] H. L. Marais, V. Zaccaria, and M. Odlare, ‘‘Comparing statistical process
tive methods for risk-oriented decisions in oper-
control charts for fault detection in wastewater treatment,’’ Water Sci.
ations management, including statistical process
Technol., vol. 85, no. 4, pp. 1250–1262, Feb. 2022.
control, maintenance, and project management. He is a fellow of the Decision
[29] D. R. Lakus, M. Pizzolato, F. D. M. Albano, and P. L. Menin, ‘‘Shewhart,
CUSUM and EWMA control charts: A comparative study on intermediate Sciences Institute.
check of balances,’’ MAPAN, vol. 37, no. 2, pp. 453–464, Jun. 2022.
[30] K. Bisiotis, S. Psarakis, and A. N. Yannacopoulos, ‘‘Control charts in ZSOLT T. KOSZTYÁN is currently a Full Professor
financial applications: An overview,’’ Qual. Rel. Eng. Int., vol. 38, no. 3, and the Head of the Department of Quantitative
pp. 1441–1462, Apr. 2022. Methods, University of Pannonia. He was a Senior
[31] Y. Xiang, ‘‘Joint optimization of X̄ control chart and preventive mainte- Research Fellow with the Institute of Advanced
nance policies: A discrete-time Markov chain approach,’’ Eur. J. Oper.
Studies, KRAFT Social Innovation Laboratory,
Res., vol. 229, no. 2, pp. 382–390, Sep. 2013.
and the MTA-PE (Hungarian Academy of Sci-
[32] Z. Wu, ‘‘Asymmetric control limits of the x-bar chart for skewed pro-
ences University of Pannonia) Budapest Ranking
cess distributions,’’ Int. J. Qual. Rel. Manage., vol. 13, no. 9, pp. 49–60,
Dec. 1996. Research Group. He was a single or corresponding
[33] V. Alevizakos, K. Chatterjee, and C. Koukouvinos, ‘‘The triple moving author of more than 200 publications and three
average control chart,’’ J. Comput. Appl. Math., vol. 384, Mar. 2021, books. His research interest includes the devel-
Art. no. 113171. opment of methodologies to manage complex management problems and
[34] N. A. Saleh, M. A. Mahmoud, L. A. Jones-Farmer, I. Zwetsloot, and systems relating to mathematical models and algorithms of project manage-
W. H. Woodall, ‘‘Another look at the EWMA control chart with estimated ment, production, maintenance, and network science. This research area is on
parameters,’’ J. Qual. Technol., vol. 47, no. 4, pp. 363–382, Oct. 2015. the frontier between management science, applied informatics, and applied
[35] S.-F. Yang, J.-S. Lin, and S. W. Cheng, ‘‘A new nonparametric EWMA sign network science.
control chart,’’ Expert Syst. Appl., vol. 38, no. 5, pp. 6239–6243, May 2011.

VOLUME 11, 2023 97573