Purdue University

Purdue e-Pubs

Biotechnology Innovation and Regulatory

BIRS Africa Technical Reports Science Center Global Community

12-9-2021

Defects Analysis and Root Cause Robustness of Product Labels

of an Acaricide Product in East Africa
Moses Okal Ogodo
Purdue University, [email protected]

Z Ekeocha
Purdue University, [email protected]

S Byrn
Purdue University, [email protected]

K Clase
Purdue University, [email protected]

Follow this and additional works at: https://docs.lib.purdue.edu/birsafricatr

Recommended Citation
Ogodo, Moses Okal; Ekeocha, Z; Byrn, S; and Clase, K, "Defects Analysis and Root Cause Robustness of
Product Labels of an Acaricide Product in East Africa" (2021). BIRS Africa Technical Reports. Paper 9.
http://dx.doi.org/10.5703/1288284317441

This document has been made available through Purdue e-Pubs, a service of the Purdue University Libraries.
Please contact [email protected] for additional information.
1

Defects Analysis and Root Cause Robustness of Product Labels of an Acaricide Product in
East Africa
M. Okal Ogodo 1, Z. Ekeocha 2, S. Byrn 3, K. Clase 4

ABSTRACT
This project aimed at investigating online damages of packaging materials generated during production because
of imperfectness of production processes. These online damages may lead to upsurge in production cost and/or
market returns, causing the company to experience losses and even damage its reputation. Data on online
damages were collected for 20ml, 40ml and 100ml labels of an acaricide product for the period between July 2018
and June 2019 and statistically analyzed. Investigation was done on the causes of online damages, then corrective
and preventive actions carried out. Analysis of online damages of labels of the three pack sizes revealed a loss of
1.01% on labels alone during the year under analysis. After implementing corrective and preventive actions, there
was a reduction of online damages of labels.

Keywords: zero defects policy, defective items, imperfect production, root cause robustness, analysis of defects,
IN state, OUT state, in-control, out-of-control, just-in-time

1. INTRODUCTION parameter values for wrongly accepting a

nonconforming product and rejecting a conforming
This paper extends previous models of production product as infinity (Capdevilla et al., 2020; Kancherla et
process in relation to packaging materials that were al., 2017; Shih et al., 2018; Li, et al., 2015) while
issued for packaging processes. One product line with investigating by jointly considering product
several stock keeping units (SKUs) was selected for the deterioration and a deteriorating production system,
study where labels of three SKUs of 20ml, 40ml and state that, not only does the machine produce defective
100ml were considered. An assumption was made in product, but also the machine is subjected to quality
the production process design that production deterioration (Li et al., 2015). It was noted that while
processes are imperfect and all material issued to producing, defective products occurred because of
production are used, without rejection. Shih and Wang labeling nonconformance at different time intervals
stated that products are produced by an imperfect during the labeling process which are isolated then kept
process that may shift randomly from the IN state aside till end of normal production process. The
control to the OUT state control (Shih & Wang, 2013). defective products because of labeling nonconformance
Further, they stated that when the process is in the OUT are then assessed and a decision is made whether to
state, it had a higher probability of producing a rework or reject. The production system is assumed to
nonconforming product compared to when it is in the IN produce some time-varying proportion of defective parts
state (Shih & Wang, 2013). Some companies have, which can be repaired at some unit cost. (European
therefore, adopted a “zero defects” policy, to reduce the Journal of General Practice, 2020). When reworked,
many online production defects that may otherwise extra cost is incurred so as to restore their quality to
increase production cost. Accordingly, several authors compliance to specifications and customer expectation.
have alluded to the fact that a “zero defects” policy can Datta, in his journal An Inventory Model with Price and
be easily achieved by a model by setting both cost Quality Dependent Demand Where Some Items

1
[email protected]; Biotechnology Innovation and Regulatory Science (BIRS) Center; Agricultural and Biological
Engineering, Purdue University
2
[email protected]; Medical Missionaries of Mary; Biotechnology Innovation and Regulatory Science (BIRS) Center,
Purdue University
3
[email protected]; Biotechnology Innovation and Regulatory Science (BIRS) Center; Industrial and Physical Pharmacy,
Purdue University
4
[email protected]; Biotechnology Innovation and Regulatory Science (BIRS) Center; Agricultural and Biological
Engineering, Purdue University
2

Produced Are Defective, analyzed an inventory system and ensure that delays in delivering products to
for joint determination of product quality and selling customers that may be attributed to reworks do not
price where he stated that a fraction of items produced occur.
were defective and that only a fraction of the defective
items can be repaired or reworked (Datta, 2013). As for
First, in this study, all defects were considered of 20ml,
the 20ml, 40ml and 100ml labels of acaricide product
40ml and 100ml labels that were attributed to online
that were under review, it was noted that most online
inspection were analyzed. Then the Pareto Analysis tool
damages could not be reworked or repaired at all,
was used to establish the top three probable causes
leading to outright rejection. The cost of rejection was
(Giraldo-Londono et al., 2020; Hajabdollahi et al., 2020;
therefore incurred because these labels ended up being
Liu et al., 2020; Vo-Duy et al., 2020). Second, corrective
disposed. To complete production, additional labels had
and preventive actions were carried out. Essentially, key
to be obtained, hence, causing inflated cost of
strategies and improvements implemented in the Henry
production. He et al.(2015), in the journal, Optimal
Ford Production System during a period of 1 year that
Production Planning for Manufacturing Systems with
were responsible for the success in markedly reducing
Instantaneous Stock-Dependent Demand and
waste and rework in the labelling process of the three
Imperfect Yields, state that during production process,
SKUs were shared (Hodgson et al., n.d.). To measure
the system can evolve from in-control state into out-of-
progress in this effort, comparisons were made of the
control state at any random time. They further go on to
baseline pre-improvement with post-improvement rates
state afterwards, the defective items will be generated
and types of process defects encountered in the
and likely cause quantity loss (He et al., 2015). This is
process. The continuous quality and process
mostly the case when large numbers of packaging, like
improvements were accomplished by empowered
labels, are issued to production. As such, it was noted
workers in a blame-free environment using innovative
that production labeling machines required accurate
tools for real-time data collection (Hodgson et al., n.d.).
setup at the initial stage, along with continuous
monitoring to maintain the system in an in-control state An additional aspect of the study considered process
constant. inspection during normal production for a period of six
months. During this time the process was documented
Process yield target and production costs are based on
to be IN state control and/or OUT state control, as per
the assumption of process imperfectness. The
Shih, et al. (2018) model. The study was extended by
imperfectness in production process result in the
adopting quality improvement investment as per Salari
projected production yield assuming that some
and Makis (2020) model.
product’s inputs are lost or damaged. This result results
into a relatively low target yield which eventually may
lead to high cost of production. From the this study, data 2. METHODS
analysis of defects of 20ml, 40ml and 100ml Labels of
the acaricide product revealed this through labels being Data on 20ml, 40ml and 100ml online labels damages
damaged due to misalignments in the labeling machine were collected from packing batch records for all the
and improper machine setting by the operators. During batches manufactured in the period between July 2018
the period between July 2018 and June 2019, out of and June 2019. Tabulation of data was done per month
1,229,966 pieces of labels for the three pack sizes for all the labels issued against all that were damaged
issued to production, 1.01% was damaged online, online. Sum and mean values of online damages and
prompting additions for completion of production total labels issued per month were calculated and
batches. 1.82% of the damages were for 20ml labels, tabulated in Table 1.
1.12% for 40ml and 0.58% for 100ml. Graphical presentation was done to show the general
This meant that online quality control checks were distribution per pack size (or SKU) per month as shown
paramount at regular intervals. The machine operators in Figure 1. Investigation on root cause with the help of
ensured regular adjustment of labeling machine in Ishikawa Fish- Borne diagram was carried out to help
cases of misalignment or variations. establish the causes of damages to inform actions to
be taken to mitigate problem of damages. Pareto
In a scenario in which a company adopts a “just-in-time” analysis was carried on three causes of online
(JIT) Policy, Salari and Makis in the journal Joint damages. These three causes were selected for
maintenance and just-in-time spare parts provisioning mitigation because of their potential to significantly
policy for a multi-unit production system, explain that impact the reduction of damages.
quality improvement investment is a function of defects
Corrective actions and preventive actions (CAPA)
rate (Salari & Makis, 2020). The strategy for quality
were proposed and documented for the three selected
improvement should commence at the point of receiving
causes. Proposed actions were implemented
materials and continue all through to production. This is
immediately after CAPA implementation. Fresh data
to ensure proper inventory management of materials
3

was collected for 20ml, 40ml and 100ml labels in July Sum and mean values of online damages and total
2019 and tabulated per month against all that were labels issued per month were calculated and tabulated
damaged online. in Table 2. Graphical presentation was done to show
the general distribution per pack size (or SKU) per
month as shown in Figure 2.

Table 1: Quantities of Labels’ Damages by Month (July 2018 to June 2019)

MONTH 20ML 40ML 100ML
Total Qty % Total Qty % Total Qty %
Damages Issued Damages Damages Issued Damages Damages Issued Damages

Jul 2018 0 0 0 532 49128 1.08 121 14049 0.86

Aug 2018 260 2520 10.32 98 16800 0.58 282 21000 1.34
Sep 2018 208 24000 0.87 404 31200 1.29 35 15540 0.23
Oct 2018 283 9600 2.95 770 32400 2.38 55 5334 1.03
Nov 2018 105 21600 0.49 1853 110640 1.67 93 28539 0.33
Dec 2018 134 12005 1.12 423 71976 0.59 82 8400 0.98
Jan 2019 207 51197 0.4 1037 124368 0.83 30 51662 0.06
Feb 2019 47 8400 0.56 895 45600 1.96 41 10500 0.39
Mar 2019 30 13200 0.23 327 36880 0.89 43 18900 0.23
Apr 2019 518 79896 0.65 403 55584 0.73 177 38094 0.46
May 2019 1529 36000 4.25 1166 114456 1.02 130 27930 0.47
Jun 2019 6 9784 0.06 101 28584 0.35 23 4200 0.55
Total 3327 268202 0.06 8009 717616 1.12 1112 244148 0.58

Table 2: Summary of Damages

SKU Total Qty of Damages Total Qty Issued

20ML 3327 268202

40ML 8009 717616

100ML 1112 244148

Total Damages 12448 1229966

% Damages(Average) 1.01%
 Figure 1: Monthly distribution of online damages of 20ml, 40ml
and 100ml labels of the acaricide product between July 2018 and
June 2019
Number of damages in pieces

2000
1800
1600
1400
1200
1000
800

I. I.I .1_ .I. ...I. I _ _I_ J _ 11.

600
400
200
0 I •-
2018 2018 2018 2018 2018 2018 2019 2019 2019 2019 2019 2019
Jul Aug Sep Oct Nov Dec Jan Feb Mar Apr May Jun
Month

■ Online Damages of 20ml ■ Online Damages of 40ml ■ Online Damages of 100ml

Table 3: Quantities of Labels’ Damages by Month (July 2019 to December 2019)

MONTH 20ML 40ML 100ML
Total Qty % Total Qty % Total Qty %
Damages Issued Damages Damages Issued Damages Damages Issued Damages
I
Jul 2019 0 0 0 651 48984 1.33 0 0 0
Aug 2019 320 69674 0.46 686 60386 1.14 12 6301 0.19
Sep 2019 179 36152 0.50 0 0 0.00 88 21211 0.41
Oct 2019 320 56401 0.57 557 38760 1.44 142 20581 0.69
Nov 2019 32 24048 0.13 121 36985 0.33 4 4830 0.08
Dec 2019 252 49992 0.50 830 57985 1.43 82 13923 0.59
Total 1103 236267 0.47 2845 243100 1.17 328 66846 0.49

Table 4: Summary of damages

SKU Total Qty of Damages Total Qty Issued

20ML 1103 236267

40ML 2845 243100

100ML 328 66846

Total Damages 4276 546213

% Damages (Average) 0.78%

Figure 2: Monthly distribution of damages of labels of the acaricide product
between July 2019 and December 2019)
900
Number of damages in pieces

675

450

225

0
2019-Jul 2019-Aug 2019-Sep 2019-Oct 2019-Nov 2019-Dec
Month
■ 20ml ■ 40ml ■ 100ml

introduced as a new pack size towards the end of the

3. RESULTS AND DISCUSSION year in 2018, so production volumes were still not as
much as 40ml which was the fastest moving SKU.
Table 1 above is quantity of all the labels for 20ml, However, damages were still being experienced with
40ml and 100ml of the acaricide product that were
20ml SKU because the labels quality specifications
issued monthly to production during the period
such as dimensions and grammages were still not
between July 2018 and June 2019. The table also
fully aligned with the dimensions of the labeling
has quantities of labels that got damaged during
machine.
packaging process in production (i.e., online
damages). Table 2 is a summary of total quantities Figure 1 shows distribution of online damages of
issued and online damages per SKU. A total of 20ml, 40ml and 100ml labels by month between July
1,229,966 pieces of labels were issued to production 2018 and June 2019. This was before
for 20ml, 40ml and 100ml SKUs at different months implementation of proposed corrective and
during the period. Out of that was issued, 12,448 preventive actions in Table 5. Online damages were
pieces were recorded as online damages during highest in October to December 2018 due to high
labeling. Percentage online damages was 1.012% of personnel turnover that was experienced, leading to
the total labels issued during the period for the three inadequately trained operators handling the
SKUs. Out of the total recorded online damages, machine. More experienced machine operators left
3,327 pieces were for 20ml, 8,009 pieces for 40ml the company, leaving the organization with
and 1,112 pieces for 100ml. The total online inexperienced operators to handle the machines.
damages of labels amounted to an equivalent cost of This led to regular machine interruptions due to setup
USD 52 for the period between July 2018 and June issues. Many cases of label misalignments were
2019, based on the unit price of the labels alone. being reported and would lead to stoppages. By the
That is, USD 22 for 20ml, USD 14.5 for 40ml and time mitigation measures were taken, several labels
USD 15.5 for 100ml. Unit price of labels for each SKU were already damaged.
was multiplied by the total online damages for that
SKU. The totals for each SKU were then summed up
to obtain the total of USD 52. This cost, however, did Root Cause Analysis:
not include the man and machine hours during
reworking of the affected units. a. The outcome of data analysis of Tables 1 and
2 prompted investigation which led root cause
analysis by use of Ishikawa diagram
From Table 2 it can be seen that 40ml labels (Chokkalingam et al., 2017; Liliana Luca et al.,
contributed to the highest number of the total online 2017; Petr Kedaj & Josef Pavlíček, 2015;
damages. This was because 40ml labels were the Stefanovic, Kiss, Stanojevic, & Janjic, 2014).
highest used, as a total of 717,616 pieces were Investigation report identified the following as
issued to production. The 20ml SKU had just been probable causes of online damages of labels
 of the acaricide product: Material: 20ml and d. Machine: Though preventive maintenance
40ml labels’ specifications of grammage and schedule was in place; it was not being
dimensions were not compatible with the followed as per frequency defined. In addition,
labeling machine’s drumhead, making labels the scope of parameters that were being
picking not symmetrically automatic, as was maintained was not adequate.
expected.
e. Management:
b. Man:
The machine was old and therefore deteriorated due
i. Personnel operating the labeling machine to wear and tear. The spares were not available,
were relatively new and so were not even with the manufacturer, because of the model
experienced in identifying and correcting the being obsolete, forcing the maintenance team to
problem during operation. regularly fabricate the parts. This introduced a lot of
variations, including not being able to set and/or
ii. Operators were not trained on the
determine the speed. Pareto Analysis of Root
autonomous maintenance of the machine,
making it difficult for them to adjust in case Causes and Corrective and preventive
of misalignment or variation after initial action (CAPA).
setup. Pareto analysis (Bajaj et al., 2018; Davis, 1981;
c. Method: Galloway, 2014; X. He & Khouja, 2011) tool was
used to isolate main causes with significant effect.
i. Procedure for receiving packaging materials These were reduced to the following three causes:
(including labels) was inadequate in that it
did not have a sampling plan that ensured i. Man
that labels sampled were representative. ii. Method
The scope of quality control checks was
inadequate with respect to parameters that iii. Material
were being checked at receipt. Corrective and Preventive Actions done are shown in
ii. Online quality control checks were Table 5 below.
inadequate.
Table 5: Corrective and Preventive Actions
Root Observation CAPA
Cause

Man i. Personnel operating the labeling Management increased the terms of contract of
machine were relatively new and so machine operators from short to long term with
were not experienced in identifying and increased remuneration.
correcting the problem during
operation. This was due to high
personnel turnover because of short
term contract.

ii. Operators were not trained on the Operators were trained on autonomous
autonomous maintenance of the maintenance by a qualified external party.
machine, making it difficult for them to
adjust in case of misalignment or
variation after initial setup

Method i. Procedure for receiving packaging Procedure for receiving packaging materials was
materials (including labels) was reviewed to include a robust sampling plan and
inadequate in that it did not have a scope of quality checks at receipt increased.
sampling plan that ensured that labels
sampled were representative. Besides,
the scope of quality control checks was
inadequate with respect to parameters
that were being checked at receipt.

ii. Online quality control checks were Batch packing records were reviewed to increase
inadequate frequency of in-process checks in production.

Material 20ml and 40ml labels’ specifications of Specifications of 20ml and 40ml labels were
grammage and dimensions were not reviewed to align with labeling machine and with
compatible with the labeling machine’s bottles.
drumhead, making labels picking not
symmetrically automatic as was expected

Machine Though preventive maintenance schedule was Scope of parameters for preventive maintenance
in place, it was not being followed as per defined and frequency of maintenance enforced as
frequency defined. In addition, the scope of per standard operating procedure.
parameters that were being maintained was not
adequate

The CAPA implementation, as per Table 5 above, caused by the old state of the machine. One of the
was done gradually between December 2018 and machine parts had to be fabricated because the
March 2019. manufacturer did not have the part since the machine
From Table 1, labels damages were unusually low model had been rendered obsolete. The matter was
in the months of March and April 2019 for 20ml reported to management for a consideration to
labels. This was since a review of labels procure modern equipment which would be
specifications of grammage and dimensions had just considered a current technology.
been done, leading to improved quality of labels Reviewing of batch packing records to include
being received from the supplier. Besides, the additional online quality checks, training of personnel
volumes of 20ml SKU were relatively low during on autonomous maintenance and machine setup
these two months. This was not so in the month of were conducted in June and July 2019. Additional
May 2019 for 20ml and 40ml labels due to the personnel were included in the training schedule to
breakdown of the labeling machine, which was increase the number of operators that were able to
operate the machine. Training requirements were not 4. CONCLUSION
only a need for the personnel on the labeling line for
these three SKUs, but for all the lines in production.
Machine setup, packaging mmaterials specification
and personnel competence is key to reduction of
From Table 3, data collected for a period of six (6) online damages. This is more so with respect to
months from July to December 2019 after (and as) compatibility of label’ specification with labeling
changes were (being) made revealed that a total of machine which, with use of competent machine
546,213 pieces had been issued so far for 20ml, operators, significantly improve efficiency. Most
40ml and 100ml production. Out of this, 1,103 pieces misalignment of labels during labeling occur because
of 20ml labels were damaged online against 236,267 of improper initial setup. Review of labels Quality
pieces that were issued, 2,845 pieces of 40ml labels Control specifications also impacted positively on the
were damaged online against 243,100 pieces that reduction of online damages for 20ml SKU of the
were issued while 328 pieces of 100ml labels were acaricide product.
damaged online against 66,846 pieces that were
issued. This represented percentage online
damages of 0.36%, 0.94% and 0.33% for 20ml, 40ml Online damages also contribute to total cost of
and 100ml labels, respectively. In terms of cost, production. This is because they lead to actual
these online damages amounted to an equivalent of losses by reducing the profit margins. Any reworks of
USD 16.8 for the period (i.e., USD 7.21 for 20ml, unit packs amount to additional cost due to use of
USD 5.04 for 40ml and USD 4.55 for 100ml). ). This extra materials, man and machine hours to complete
cost did not include the man and machine hours, and the production lot.
production downtime. The cost was only based on There was significant reduction of online damages of
the cost price of each label. This was too early to 20ml, 40ml and 100ml labels of the acaricide product
conclude, though, as it was only a half-year period. after implementation of proposed actions. This
However, between July to December 2019, online means that the mitigation measures that were put in
damages of 20ml labels reduced from 1.8% in Table place were effective.
1 to 0.36% in Table 3. For 40ml labels, reduction of
online damages was from 1.12% in Table 1 to 0.94%
5. RECOMMENDATIONS FOR NEXT STEPS
in Table 3 while that of 100ml labels was from 0.58%
in Table 1 to 0.33% in Table 3. Reduction in damages
was overall reduced from 1.01% to 0.5%, though this a) Monitoring of online damages to continue for
was for a half-year period. the next six months to conclusively have a trend
From Figure 2, online label damages generally on the impact of the CAPA.
reduced during July to December 2019. This was at
a time when the proposed CAPA had generally been b) Online damages of other packaging materials
implemented, meaning that the actions taken yielded such as bottles, unit cartons, etc. of the
results. acaricide product to be monitored to give overall
The projected cost for the whole year for the period cost of production defects of the Product.
starting July 2019 would be USD 33.6, implying that
cost would reduce by 35.4% i.e., USD 18.4. This
though was too early to conclude. c) Online damages of other packaging materials of
other products to be monitored so as to give
During the period between July and December 2019, overall cost of production
the management also reviewed contract of short-
term employees to three (3) years, and this
motivated staff to not only be more focused but also
to remain with the company.
REFERENCES He, X., & Khouja, M. (2011). Pareto analysis of
supply chain contracts under satisficing
Anonymous. (2020). General Practice and the objectives. European Journal of Operational
Community: Research on health service, Research, 214(1), 53–66.
quality improvements and training. Selected https://doi.org/10.1016/j.ejor.2011.03.047
abstracts from the EGPRN Meeting in Vigo, Hodgson, A. J., Shang, Y. M., Boulianne, P.,
Spain, 17-20 October 2019 Abstracts. Downes, M. R., Hwang, D., & Slodkowska,
European Journal of General Practice, E. (n.d.). A Practical Approach to
26(1), 42–50. Investigating Cross-Contaminants in the
https://doi.org/10.1080/13814788.2020.171 Anatomical Pathology Laboratory.
9994 International Journal of Surgical Pathology,
Bajaj, S., Garg, R., & Sethi, M. (2018). Total quality 1066896920913110.
management: A critical literature review https://doi.org/10.1177/1066896920913110
using Pareto analysis. International Journal Kancherla, V., Walani, S. R., Weakland, A. P.,
of Productivity and Performance Bauwens, L., Oakley, G. P., & Warf, B. C.
Management, 67(1), 128–154. (2017). Scorecard for spina bifida research,
https://doi.org/10.1108/IJPPM-07-2016- prevention, and policy—A development
0146 process. Preventive Medicine, 99, 13–20.
Chokkalingam B, Raja V, Anburaj J, Immanual R, & https://doi.org/10.1016/j.ypmed.2017.01.02
Dhineshkumar M. (2017). Investigation of 2
Shrinkage Defect in Castings by Quantitative Li, N., Chan, F. T. S., Chung, S. H., & Tai, A. H.
Ishikawa Diagram. Archives of Foundry (2015). An EPQ Model for Deteriorating
Engineering, 17(1), 174–178. Production System and Items with Rework.
https://doi.org/10.1515/afe-2017-0032 Mathematical Problems in Engineering,
Datta, T. K. (2013). An Inventory Model with Price 957970.
and Quality Dependent Demand Where https://doi.org/10.1155/2015/957970
Some Items Produced Are Defective. Liliana Luca, Minodora Pasare, & Alin Stancioiu.
Advances in Operations Research, UNSP (2017). STUDY TO DETERMINE A NEW
795078. MODEL OF THE ISHIKAWA DIAGRAM
https://doi.org/10.1155/2013/795078 FOR QUALITY IMPROVEMENT. Fiabilitate
Davis, D. (1981). Pareto analysis. Building Services, Şi Durabilitate, 1(19), 249–254.
3, 49–50. Liu, G., Zhang, B., Zhang, Y., & Guo, C. (2020).
Galloway, S. (2014). Finding Focus: A Modeling of double-layer triangular
Transformational Pareto Analysis. microchannel heat sink based on thermal
Professional Safety, 59(7), 52–53. resistance network and multivariate
Giraldo-Londono, O., Mirabella, L., Dalloro, L., & structural optimization using firefly algorithm.
Paulino, G. H. (2020). Multi-material Numerical Heat Transfer Part B-
thermomechanical topology optimization Fundamentals, 77(5), 417–428.
with applications to additive manufacturing: https://doi.org/10.1080/10407790.2020.171
Design of main composite part and its 7834
support structure. Computer Methods in Petr Kedaj, & Josef Pavlíček. (2015). Modified
Applied Mechanics and Engineering, 363, Ishikawa Diagram as a Tool for Knowledge-
112812. Mapping of Agronomic Practices. Acta
https://doi.org/10.1016/j.cma.2019.112812 Informatica Pragensia, 4(2), 174–181.
Hajabdollahi, Z., Hajabdollahi, H., & Kim, K. C. https://doi.org/10.18267/j.aip.68
(2020). Multi-objective optimization of solar Salari, N., & Makis, V. (2020). Joint maintenance and
collector using water-based nanofluids with just-in-time spare parts provisioning policy
different types of nanoparticles. Journal of for a multi-unit production system. Annals of
Thermal Analysis and Calorimetry, 140(3), Operations Research, 287(1), 351–377.
991–1002. https://doi.org/10.1007/s10973- https://doi.org/10.1007/s10479-019-03371-3
019-08444-w Shih, N.-H., Liao, Y.-S., & Wang, C.-H. (2018).
He, L., Peng, H., Niu, Z., Lu, H., & Xie, X. (2015). Determining an economic production
Optimal Production Planning for quantity under a zero defects policy. Journal
Manufacturing Systems with Instantaneous of the Operational Research Society, 69(8),
Stock-Dependent Demand and Imperfect 1309–1317.
Yields. Mathematical Problems in https://doi.org/10.1080/01605682.2017.139
Engineering, 271902. 2405
https://doi.org/10.1155/2015/271902 Shih, N.-H., & Wang, C.-H. (2013). An Extended
Production and Inspection Model with
 Nonrigid Demand. Scientific World Journal,
623075.
https://doi.org/10.1155/2013/623075
Stefanovic, S., Kiss, I., Stanojevic, D., & Janjic, N.
(2014). ANALYSIS OF TECHNOLOGICAL
PROCESS OF CUTTING LOGS USING
ISHIKAWA DIAGRAM. Acta Technica
Corviniensis - Bulletin of Engineering, 7(4),
93–98.
Vo-Duy, T., Duong-Gia, D., Ho-Huu, V., & Nguyen-
Thoi, T. (2020). An Effective Couple Method
for Reliability-Based Multi-Objective
Optimization of Truss Structures with Static
and Dynamic Constraints. International
Journal of Computational Methods, 17(6),
1950016.
https://doi.org/10.1142/S021987621950016
6

ACKNOWLEDGEMENTS
I would like to thank all of the BIRS guest faculty from
global industry and regulatory organizations for
generously sharing their professional expertise and
providing donated, in-kind time towards building the
professional skills and technical capabilities of the
students within the BIRS program. I would also like
to thank my fellow peers in the BIRS MS student
cohort for providing guidance and constructive
feedback during the classroom group work and
interactive sessions; Abigail Ekeigwe and Mercy
Okezue, Purdue ABE BIRS PhD candidates, for their
mentorship and input throughout the project;
Professor Fran Eckenrode for providing content
expertise throughout the review process on this
paper; and Lauren Terruso, operations manager for
BIRS Center, for all of her efforts on editing multiple
iterations of the technical paper draft in preparation
for publication. The international component of the
Purdue BIRS program was initiated through
educational support provided by the Merck
Foundation and most recently through a capacity
building effort funded by the Bill and Melinda Gates
foundation, grant # 41000460.

Special thanks go to my employer Cooper K-Brands

who, not only allowed me to be part of the program,
but also supported me financially in all my travel and
hotel expenses to and from academic site.