COMSATS University Islamabad Campus


Synopsis for the degree of  M.S./M.Phil. Ph.D.


PART-1 (to be completed by the student)

Name of Student Waqas Saeed

Department Mathematics

Registration No. SP22-RMT-057

Name of (i) Research Supervisor Dr. Mansoor Shaukat Khan

(ii) Co - Supervisor

Members of Supervisory Committee (Applicable for PhD students only)

1.

2.

3.

Title of Research Proposal A Statistical Scheme to Monitor Process Location

Summary of the Research Proposal

Memory control charts, such as the exponentially Control graphs for modified shifting
average (EWMA) and the cumulative sum (CUSUM), is utilized a lot for tracking modest
to slight changes in the technique's site and variance. Enhanced version of EWMA
controls graph to track procedure site moves the heterogeneously weighted moving
average (HWMA) controls graph. Additionally, based on information secondary storage
controls graph effectively track shifts in the position of processes. The goal of this
research is to suggest a double HWMA based on auxiliary information, denoted by the

control chart , in order to improve the tracking of position movements for the

procedure. The controls graph is modelled merging the component dependent

on data HWMA charting statistical features with the other HWMA controlling graph.
Monte Carlo simulations are a computer method used for Numeric outcomes. A

suggested controls graph is compared with more controls graph, as well as

classical CUSUM and EWMA, mixed EWMA-CUSUM, Additional details predicated

EWMA ( ), Heterogeneously WMA, Additional details predicated

Heterogeneously WMA ( ), and double Heterogeneously WMA control charts,

using well-known performance evaluation like typical duration of a run (ARL), extra
quadratic loss, relative ARL, and index for assessment of efficiency. Comparative
analysis showed that the suggested controls graph worked better than alternative controls
graphs, particularly is minor the slight procedure position changes. Users and
practitioners are also given access to an application for the automotive braking system
that highlights the significance of the suggested study from a real-world standpoint.

Introduction
A type of control method known as statistical process control (SPC) makes use of
statistical techniques to monitor and regulate a procedure to make sure it runs regularly
and up to the required standards of quality. It includes gathering data in real-time,
analyzing it to see any changes or patterns that could point to a system issue, and then
taking measures to stop faults in happening. All manufacturing and non-manufacturing
procedures, regardless of their nature, require variation. There are two types of variations:
natural and unnatural. Natural variations cannot be prevented because they are risk-free
and part of every process. The artificial differences, however, weaken the sufficiency of
the final result and damage the procedure' efficacy. Shewhart developed memoryless
control charts, which are useful for keeping track of lengthy tasks [1]. whereas traditional
memory control charts like Roberts' exponentially weighted moving average (EWMA)
and cumulative sum (CUSUM) are utilized to successfully handle small to regulate sized
changes [2]. The basic structures of traditional memory control charts are regularly
modified and improved to increase process monitoring. A mixed method for traditional
memory control charts is one such improvement. The achievement of controls graphs is
enhanced by mixed method. For example, Abbas and co [3]. proposed the use of a
combined EWMA CUSUM (MEC) control chart to identify alterations in the exact spot
of the procedure. The combined CUSUM-EWMWA (MCE) control chart was also
suggested by Zaman eit all [4]. for the monitoring of the position and distribution
parameters [5]. For effectively tracking process location, Ali and Haqq advise using the
mixed CUSUM-Generally weighted moving average control chart [6]. Additionally,
Anwar et al expanded the MCE and MEC controls graphs and proposed two additional
based on data control charts for process location (referred to as MEC_AIB and MCE-
AIB) [7]. A hybrid EWMA (HEWMA) controls graphs was also created by Haqq so that
one EWMA statistic could be utilized as an entry for a different EWMA statistic [8].
Shama and Shama,[9] Haqq et all,[10] Adeioti and Maleela Majiika,[11] Zamaan eit
all,[12] Zamaan eit all,[13] Mehmood eit all,[14] Aslam, Anwar [15] and citations there
in provide more information and insights in this area. Hunter noted that the traditional
EWMA statistic had the problem of giving the earlier data a lower value [16]. A
heterogeneously weighted mean moving average (HWMA) controls graph was presented
by Abbas as a solution to this flaw. The HWMA controls graph evenly divides the unused
values among the prior data while assigning a specific weight to the current observation.
This enhances a control chart's ability to identify changes in comparison to rival control
charts. The HWMA control chart keeps track of single-factor reliability characteristics,
yet it’s useless if the level of characteristic and other supporting data is also present [17].
The DHWMA controls graph is crucial because the initial parameter is consistent and
devoid of additional data. When the initial parameter contains additional data, like a
supplemental factor, scholars, scientists, and operators may be passionate in using
DHWMA controls graph properties to increase process' efficacy. In that situation, the
DHWMA controls graph is useless.

Statement of the Problem

The paragraph addresses how control graphs are essential elements in statistical process
control (SPC) for identifying technique’s modifications and the significance of
differences in all types of manufacturing and non-manufacturing techniques. The chapter
discusses numerous traditional memory controls graph, including their improvements and
changes, comprising a moving average (EWMA) with exponential weights and a total
(CUSUM). The changes involve the combined method for traditional memory-controlled
charts, that improves the performance of control charts by combining auxiliary data with
the first parameter. Additionally, the heterogeneously weighted moving average
(HWMA) controls graphs is recommended as solution to the problems with traditional
EWMA statistics. The D Heterogeneously WMA controls graph is also mentioned in
paragraph, and it works best when the initial parameter is consistent and unambiguous.
Research Methodology

1. Define the process: First, define the process and choose the crucial variables that
must be watched. The dimensions of the product, the location of the tool. Or the
temperature and humidity of the production area, for instance, may be consider
the main characteristics if you were monitoring the location of a manufacturing
process.
2. Collect Data: Obtain data from the procedure on the important variables
identified in step 1. Sensors or other monitoring tools can be used to manually or
automatically collect this data. It is crucial to make sure the data is dependable,
consistent, and correct.
3. Calculate process mean and standard deviation: Determine the mean and standard
deviation for each key parameter from the data that was collected. The standard
deviation shows how variable the data is, whereas the mean represents the
parameter's typical value.
4. Establish control limits: Establish upper and lower control limits based on the
mean and standard deviation computed in step 3 of the process. The range of
allowed fluctuation for each important parameter is determined by the upper and
lower control limits.
5. Monitor the process: You can monitor the process by collecting data on the
crucial variables and plotting the data on a control chart. The control chart will
show if the process is operating within the control limits established in step 4. If
the data points fall within the control limits, the process is said to be in control. If
the data points go beyond the control parameters, the process is said to be out of
control.
6. Take corrective action: If the process is out of control, take corrective measures
to get it under control. This may include modifying the process's settings or
determining the root of the problem.
7. Continuous Improvement: Once the process has been brought under control,
continue to monitor it and look for ways to improve it. This can mean improving
the procedure by changing the procedure's specifications or using new
technology.
Classical CUSUM Control Chart

Graphical statistics is the traditional CUSUM are provided as

Where are initial values, is the IC procedure requires, . Where is

the initial value and the change's half-shift . The control limit of classical CUSUM
controls graph is presented as and is considered as stable. Now plot and

given in Equation against G. If any , above G, the procedure is OC state.

Classical EWMA Control Chart

Graphical statistics are provided as

HWMA Control Chart

The graphical statistics of HWMA control chart are provided as

Bibliography
1. Shewhart, W.A., 1980. Economic control of quality of manufactured product.
1931. Milwaukee (WI): Reprinted by ASQC.
2. Roberts SW. Control chart tests based on geometric moving averages.
Technometrics. 2000 Feb 1;42(1):97-101.
3. Abbas, Nasir, Muhammad Riaz, and Ronald JMM Does. "Mixed exponentially
weighted moving average–cumulative sum charts for process
monitoring." Quality and Reliability Engineering International 29, no. 3 (2013):
345-356.
4. Zaman, B., Riaz, M., Abbas, N., & Does, R. J. (2015). Mixed cumulative sum–
exponentially weighted moving average control charts: an efficient way of
monitoring process location. Quality and Reliability Engineering
International, 31(8), 1407-1421.
5. Zaman B, Abbas N, Riaz M, Lee MH. Mixed CUSUM-EWMA chart for
monitoring process dispersion. The International Journal of Advanced
Manufacturing Technology. 2016 Oct;86:3025-39.
6. Ali R, Haq A. A mixed GWMA–CUSUM control chart for monitoring the
process mean. Communications in Statistics-Theory and Methods. 2018 Aug
3;47(15):3779-801.
7. Anwar, Syed Masroor, Muhammad Aslam, Muhammad Riaz, and Babar Zaman.
"On mixed memory control charts based on auxiliary information for efficient
process monitoring." Quality and Reliability Engineering International 36, no. 6
(2020): 1949-1968.
8. Haq A. A new hybrid exponentially weighted moving average control chart for
monitoring process mean: discussion. Quality and Reliability Engineering
International. 2017 Nov;33(7):1629-31.
9. Shamma, S. E., & Shamma, A. K. (1992). Development and evaluation of control
charts using double exponentially weighted moving averages. International
Journal of Quality & Reliability Management.
10. Haq A, Gulzar R, Khoo MB. An efficient adaptive EWMA control chart for
monitoring the process mean. Quality and Reliability Engineering International.
2018 Jun;34(4):563-71.
11. Adeoti, O.A. and Malela-Majika, J.C., 2020. Double exponentially weighted
moving average control chart with supplementary runs-rules. Quality Technology
& Quantitative Management, 17(2), pp.149-172.
12. Zaman B, Lee MH, Riaz M, Abujiya MA. An adaptive approach to EWMA
dispersion chart using Huber and Tukey functions. Quality and Reliability
 Engineering International. 2019 Oct;35(6):1542-81.
13. Zaman, B., Lee, M. H., & Riaz, M. (2019). An adaptive EWMA chart with
CUSUM accumulate error-based shift estimator for efficient process dispersion
monitoring. Computers & Industrial Engineering, 135, 236-253.
14. Mehmood, R., Lee, M. H., Riaz, M., Zaman, B., & Ali, I. (2022). Hotelling T 2
control chart based on bivariate ranked set schemes. Communications in
Statistics-Simulation and Computation, 51(1), 1-28.
15. Aslam M, Anwar SM. An improved Bayesian Modified-EWMA location chart
and its applications in mechanical and sport industry. PLoS One. 2020 Feb
26;15(2):e0229422.
16. Hunter JS. The exponentially weighted moving average. Journal of quality
technology. 1986 Oct 1;18(4):203-10.
17. Abbas, N. (2018). Homogeneously weighted moving average control chart with
an application in substrate manufacturing process. Computers & Industrial
Engineering, 120, 460-470.
Tentative Time Table

Months/ Feb. to May. to Jul. to Sep. to Nov. to

Stages Apr. June. August. Oct. Dec.
2023 2023 2023 2023 2023
Literatur Literatur
e review e review
Problem Problem
formulat formulat
ion ion
Solving Solving
Problem Problem
Write up Write up
Submiss Submiss
ion of ion of
thesis thesis
PART II

Signature by the Student:

Name ------------------------------------ Signature----------------------- Date------------------------

Recommendation by the Research Supervisor

Name _________________________________ Signature _____________ Date___________

Signed by Supervisory Committee (Applicable for PhD students only)

S. # Name of Committee Member Designation Signature and Date

1
2
3

Approved by Departmental Advisory Committee

Certified that the synopsis has been seen by members of DAC and considered it suitable for
putting up to BASAR.

Secretary
Departmental Advisory Committee
Name: _______________________ ___________Signature:__________________________

Date:________________________

Chairman/HoD ____________________________________

Signature: ______________________________

Date ______________________________

PART III

Dean, Faculty of Sciences

________________ Approved for placement before BASAR.

________________ Not Approved on the basis of following reasons

Signature ______________________________ Date ________________

Secretary BASAR

________________ Approved from BASAR.

________________ Not Approved on the basis of the following reasons

Signature ______________________________ Date ________________

Dean Faculty of Sciences

Remarks:

___________________________________________________________________________

___________________________________________________________________________

___________________________________________________________________________

Signature:_____________________________ Date______________________

APO Dean Faculty of Sciences