=Paper=
{{Paper
|id=Vol-2786/Paper17
|storemode=property
|title=RandomForest Enabled Collaborative COVID-19 Product Manufacturing/Fabrications
|pdfUrl=https://ceur-ws.org/Vol-2786/Paper17.pdf
|volume=Vol-2786
|authors=Shajulin Benedict
|dblpUrl=https://dblp.org/rec/conf/isic2/Benedict21
}}
==RandomForest Enabled Collaborative COVID-19 Product Manufacturing/Fabrications==
https://ceur-ws.org/Vol-2786/Paper17.pdf
99
RandomForest Enabled Collaborative
COVID-19 Product Manufacturing/Fabrications
Shajulin Benedict
Indian Institute of Information Technology Kottayam,
Valavoor P.O., Kottayam, Kerala, India – 686635.
Abstract
Manufacturing COVID-19-related products such as face masks, shields, ventilators, shoe covers, gowns,
and so forth, rapidly increased in recent months as the virus pandemic surges across the globe. Govern-
ments and Industrialists are keen to formulate quick decisions and are open to decentralize productions
within a prescribed time limit using smart techniques. Recommendations on the volume of productions
and the producer assignments are remaining a sole concern for policymakers or smart city authori-
ties due to the unforeseen or unpredictable nature of the raging pandemic. This article introduces a
RandomForest-assisted Collaborative COVID-19 Product Manufacturing (RFCCPM) framework. It col-
laboratively decides on producing COVID-19 preventive kits in a cost-efficient manner. The approach
was experimented at the IoT cloud research laboratory; it achieved a manufacturing cost efficiency of
66 percent when Threshold Accepting (TA) algorithm was incorporated in the framework.
Keywords
COVID-19, RandomForest, Fabrication, Manufacturing, Smart Decisions
1. Introduction to the ongoing lockdown and minimal em-
ployee situations in production units. High-
One of the great health-related pandemic that quality machines have become non- opera-
has clouted all growth sectors, including man- tional for months since the eruption of the
ufacturing and the world finance sector, and pandemic across the globe. Many manufac-
has unsolved challenges is the lethality due turing companies have almost closed their op-
to the COVID-19 public crisis. The virus has erations due to reduced workforce and erupt-
predominantly led nations to unwelcoming ing supply chain disruptions. Even the most
social distancing practices, ineffective com- urgent production of COVID-19-related pre-
munications, sweeping economies, discrimi- ventive products such as masks, face shields,
nations in certain locations, distrustful rela- hood caps, shoe covers, and so forth, has wit-
tionships, and so forth, at large. nessed a catastrophe which could adversely
The Manufacturing/Fabrication sector has reiterate until innovative solutions are framed
seen a disrupted shift in productions which in the manufacturing/fabrication sectors. A
remains as an unpredictable realization by man-single manufacturer of a region could not pro-
ufacturers in order to revamp productions due duce/supply all required COVID-19 products
with utmost satisfaction in a short period in
ISIC 2021: International Semantic Intelligence Conference, the midst of the exponential growth of COVID-
February 25–27, 2021, New Delhi, India. 19 cases.
" EMAIL: shajulin@iiitkottayam.ac.in (S. Benedict)
The manufacturing sector, typically, attracts
~ URL:http://www.sbenedictglobal.com (S. Benedict)
ORCID: 0000-0002-2543-2710 (S. Benedict) a major portion of revenue in various coun-
© 2021 Copyright for this paper by its authors. Use permitted under
Creative Commons License Attribution 4.0 International (CC BY 4.0).
tries when compared to services or retail sec-
CEUR Workshop Proceedings (CEUR- tors. Accordingly, governments and manu-
WS.org)
100
facturers are keen to provide solutions and dressed for improving the voluminous pro-
upsurge productions at about 3 to 5 percent- ductions: i) Which company needs to be per-
age in order to combat the anticipated eco- mitted to develop COVID-19 products? and
nomic consequences. However, the most cru- ii) How much quantity of COVID-19 prod-
cial challenges faced by manufacturers, in- ucts is required in a particular region consid-
cluding COVID-19 product developers, dur- ering the increase of the COVID-19 cases?
ing the COVID-19-crisis epoch are listed as This article proposes a Random Forest (RF)
follows: algorithmic approach to predict the required
number of productions of COVID-19 preven-
• an abrupt stalemate in material move- tive kits based on i) the demand of a partic-
ments, especially while transferring ma- ular location, ii) COVID-19 quarantine cases,
terials from international hubs or lock- and iii) the budget availability of a region. It
down locations, due to disruption in the introduces a RandomForest-assisted Collabo-
supply chains; rative COVID-19 Product Manufacturing (RFC-
• a poor quality in COVID-19 solutions CPM) framework. The proposed RFCCPM ap-
while adopting expeditious innovations proach includes a Threshold Acceptance (TA)
with limited machinery or experiments scheduling algorithm in the framework to pre-
in a short period from restricted lock- pare a manufacturing schedule that consid-
down locations; ers the geo-distributed nature of collabora-
tive manufacturing for quick voluminous pro-
• a differing working environment, es- ductions. The RFCCPM framework has the
pecially the concept of “Work from Home“,capability to initiate production and promotes
which urged a limited access to remote economy during COVID-19 or similar health-
manufacturing sites; and so forth. related crisis of the future. It maps produc-
tion tasks to available or functional manufac-
A growing volume of research and prod- turing hubs/units using a Threshold Accept-
uct development rapidly emerged globally to ing (TA) algorithm with the objective of im-
counterfeit the COVID-19 crisis and associ- proving the cost efficiency of manufacturing
ated challenges apart from the core medical units. It can consider the lockdown situation
solutions. For instance, solutions relating to of a region while producing products.
the digital working environment, adopting a The proposed research was experimented
long term planning for cash/resource utiliza- at the IoT Cloud Research laboratory and ob-
tion, delivery management, agile production served the prediction accuracy of around 98.4
of health-related products, and so forth, pro- percent while predicting the COVID-19 quar-
liferated to improve the near future economic antine cases of Kerala state using the Ran-
catastrophe. domForest algorithm; and, the manufactur-
Health departments and concerned officials ing cost efficiency of 66 percent while incor-
of smart cities are eager to proactively de- porating TA algorithm in the framework for
velop COVID-19 preventive kits and protect preparing the manufacturing task schedule.
the residents as healthcare resources, includ- In short, RFCCPM paves way for a cost- effi-
ing hospitals and doctors, are scarcely avail- cient production of products considering the
able to nurse the exponentially growing pa- availability of minimal employees during pan-
tients. However, manufacturing the preven- demic epochs such as COVID-19. The major
tive kits has challenges – i.e., the following contributions of the proposed work, as dis-
research questions need to be clarified / ad- cussed in this paper, include the following:
101
• RFCCPM, an RF-assisted manufactur- Tens of thousands of innovations and mech-
ing framework, was developed for pro- anisms have been initiated in the recent past
ducing COVID-19 products, including using AI [19], [32], [2], and the other innova-
preventive COVID-19 products such as tive machine learning technologies. For in-
face masks, face shields, hood caps, and stance, prediction models such as Deep Neu-
so forth, understanding the increase / ral Networks have been applied to study the
decrease of COVID-19 cases of a region; increase of COVID-19 patients and the cur-
ing status of different countries [8]. Authors
• the application of RFCCPM was exper- of [16] have predicted the number of proba-
imented considering fourteen MSMEs ble deaths that happen in the tenth day due
of the Kerala state of India; and, to COVID-19.
• the Threshold Accepting (TA) algorithm A few researchers have applied prediction
of the framework was analyzed – i.e., models to study the transmission pattern of
the experiments revealed the cost- ef- the COVID-19 virus. For instance, authors of
ficiency of 66 percent while producing [20] have proposed a compartmental model
COVID-19 preventive kits. to classify the transmission patterns of the
virus; authors of [5] have applied long short
The rest of the paper is described as fol- term memory (LSTM) models to analyze the
lows: i) Section 2 explores the state-of-the- risk involved in the spread of the virus; simi-
art research in the application of RandomFor- larly, authors of [3] applied LSTM models to
est considering COVID-19 situations; ii) Sec- differentiate corono-virus from the other res-
tion 3 discusses the inner details and func- piratory diseases such as pneumonia. In [14],
tionalities of the deep-learning assisted col- authors developed a mobile-enabled contact
laborative manufacturing platform; Section 4 tracing mechanism to avoid COVID-19 con-
investigates into the theoretical aspects of im- tacts. Authors of [1] have proposed a method
proving the cost involved in the production to improve the CNN training model as suffi-
of COVID-19 products while incorporating cient COVID-19 images were not available in
RFCCPM framework; Section 5 discloses the the early period of the virus outbreak.
experimental results that were carried out at It is a known fact that the impact of COVID-
the “work at home“ working environment by 19 highly affected the mobility of humans.
accessing the machines of the IoT cloud re- The speed in spreading the virus and the sever-
search lab; and, Section 6 consolidates the find- ity of occurrence differed from region to re-
ings and insights of the proposed work along gion. This puzzles almost all solution archi-
with a few future developments. tects [21]. The travel patterns changed as peo-
ple were led to commotions [23]. In [29], au-
thors studied the impact of lockdowns in three
2. Related Work University campuses of their vicinity; the au-
thors revealed the successive progress in the
Corona Virus 2019 (COVID-19) has marked
network traffic patterns due to the discharge
its footprint in over 200 countries with se-
of lockdown policies. In addition, approaches
vere acute respiratory syndrome which af-
were devised to engage human resources ef-
fected tens of millions of people. WHO has
ficiently using online systems, preferably by
reported that 10021401 COVID-19 cases were
social media [24] considering the lockdown
reported across the globe with a total 499913
and other idleness factors of cities. Notably,
number of deaths as on 29 June 2020 [11].
authors of [4] devised a value chain consid-
102
ering the lockdown locations. assisted Collaborative COVID-19 product man-
Many researchers and practitioners agree ufacturing (RFCCPM) approach. In a nutshell,
that online tools, including collaborative tools, the RFCCPM framework allows manufactur-
would become a mandatory point of sale for ers or smart city officials to quickly produce
overriding the emerging lockdown situations the demanding COVID-19 essentials, for ex-
in cities [31]. Authors of [7] expressed the ample, preventive kits, depending on the sta-
importance of a telemedicine approach in or- tus of locations in a cost-efficient manner. Fig-
der to protect the medical practitioners and ure 1 illustrates the entities of the framework.
non-COVID-19 patients while pursuing con-
sultations. A few researchers adopted mea- 3.1. RFCCPM Entities
sures to counteract the security challenges
of online platforms such as Facebook, Twit- The major entities involved in the collabo-
ter, Whatsapp, Zoom, Chatbots, VPN, and so rative production of COVID-19 products in
forth [30]. Similarly, authors of [25], quanti- a cost-efficient manner and their important
fied the online COVID-19 information. functionalities are described below:
Succinctly, a cloud-based online production-
enabling tool would increase the productiv- 3.1.1. Information Collector
ity of COVID-19 products. A few researchers
Information Collector is a cloud-based micro-
[13] [28] [10] have suggested a cloud-enabled
service solution that collects the required sta-
service model for production units. For in-
tus details of COVID-19 from smart city data
stance, authors of [15] have developed a cloud-
repositories after the reception of appropri-
based integration of manufacturing units in
ate permissions from them. A micro-service
order to enable a remote-access of the units;
is a loosely-coupled tiny service that are in-
Martino et al. [6] have proposed a semantic
dependently deployable in clouds. In gen-
representation for establishing Industry 4.0
eral, the number of COVID-19 patients, the
based cloud services; Saivash et al. [26] have
number of deaths due to the virus, and the
established a collaborative digital dentistry prac-
number of quarantined candidates are quite
ticing platform using cloud manufacturing con-
openly available in major cities of various coun-
cepts, and so forth; Gajamohan et al. [17] have
tries as they are involved in reporting the in-
proposed a cloud-based robotics platform.
fectious status to WHO. The Information Col-
Besides, Prateek et al. [22] has proposed a
lector entity, a golang based cloud service, pro-
computer vision-enabled approach to increase
vides the information in a csv format to the
the social distancing pattern in the manufac-
RandomForest prediction engine after convert-
turing location. However, not much research
ing the formats of the source repository.
work applies the cloud-based services to de-
velop COVID-19 products – i.e., very few re-
search works have been discussed to improve 3.1.2. RF Prediction Engine
the manufacturing aspects of COVID-19 pre- Predicting the number of quarantine cases in
ventive products. a particular location is mandatory to decide
on manufacturing the number of COVID-19
preventive products. Smart city officials could
3. RFCCPM Framework utilize the data to fix policies and sketch lay-
This section explains the entities involved in outs for a complete/partial lockdown in a lo-
the proposed RFCCPM framework, the RF- cation. Manufacturing preventive kits based
3.1 RFCCPM Entities 103
Figure 1: RFCCPM Framework
Figure 2: Processes Involved in the RFCCPM Framework
on the past information of the COVID-19 cases duction depending on the current informa-
might not satisfy the requirements. Two is- tion would be a futile decision. Instead, RFC-
sues are possible: CPM utilizes a RandomForest-based predic-
• A large volume of COVID-19 preven- tion of COVID-19 cases for the future months
tive kits may be required in a short pe- so that the framework could proactively rec-
riod – i.e., the productions might get ommend the concerned for deciding on the
hampered due to the limited availabil- production of the preventive kits. Besides,
ity of employees during the period; or, the RFCCPM enables a collaborated manu-
facturing platform that attains the manufac-
• Manufacturing a surplus amount of pre- turing schedule with the limited number of
ventive kits leads to under-utilization. employees during the raging pandemic pe-
The information is dynamic that the pro- riod in a cost-efficient manner.
3.1 RFCCPM Entities 104
In general, the RF algorithm is a subset of masks using N95 masks which provide
AI and machine learning [9]; it is an ensemble- higher protection from COVID-19 im-
based learning algorithm. The RF Prediction pacts or using cloths; ii) face shields
Engine entity of RFCCPM incorporates the may be manufactured using low-grade
algorithm in order to train and predict the disposable plastics or poly-carbonates;
COVID-19 quarantine cases of a region. The iii) shoe covers could be fabricated us-
training model is prepared from the informa- ing of non-woven materials or polypro
tion such as latitude, longitude, number of pylene fabrics; and so forth. In addi-
deaths, number of quarantined candidates, and tion to the COVID-19-related products,
so forth, for the locations where the preven- manufacturing the packing device that
tive kits are required. The engine attempts to holds the preventive products is also
improve the prediction accuracy depending considered as a task. The Task Designer
on the tuning parameters of the algorithm entity of the RFCCPM framework pre-
such as 𝑚𝑡𝑟 𝑦 and the other variable selection pares a set of optional combinations for
methods. More detailed information about manufacturing different COVID-19 prod-
the algorithm is found in the previous litera- ucts of preventive kits.
ture [27]. 2. Packaging Products: Packaging is a task
that needs to be accomplished by as-
3.1.3. Task Designer sembling all manufactured products into
the packaging device.
Tasks such as manufacturing face masks, face
shields, gloves, shoe covers, infrared thermome-
ters, sanitizer, kit cover, and packaging them 3.1.4. Scheduler
need to be formulated depending on the avail- Scheduler organizes tasks depending on the
able manufacturing units within the reach- cost factors of allotting tasks to specific man-
able vicinity. The Task Designer entity of the ufacturing units – i.e., it identifies the man-
framework attempts to pool tasks in a cloud ufacturing tasks given the information about
database, namely, mongodb database. It con- the available manufacturing units of a region/
siders several combinations of manufactur- location. Although the main objective of the
ing options while designing tasks. The tasks Scheduler entity is to reduce the manufactur-
are described in a machine-readable format ing cost, it has to consider several other pa-
such as XML, jade, or JSON. rameters such as the availability of appropri-
The tasks included in the RFCCPM frame- ate manufacturing units. For instance, if a
work are classified into two broad categories: face mask needs to be manufactured using
i) Manufacturing tasks and ii) Packaging tasks: N95, an appropriate manufacturing unit that
1. Manufacturing COVID-19 Products: Man- is nearest to the vicinity should be available.
ufacturing COVID-19 preventive kits in- Besides, it has to consider the prediction re-
clude manufacturing products such as sults available through the RF prediction en-
face masks, face shields, gloves, hood gine. This is crucial as lockdown situations
caps, shoe covers, thermometers, sani- could hamper the transfer of products for the
tizers, and so forth. These products are final preparation of predictive kits. Thresh-
often manufactured in different cate- old Accepting (TA) [18] [27] scheduling algo-
gories. For instances, i) face masks could rithm is applied in the framework although
be manufactured at different levels of several other scheduling algorithms could be
protections – i.e., manufacturing face implemented in the framework. The inner
105
details of the cost-efficiency of the framework, enable smart city officials or enterprises
while utilizing the Scheduler entity, is discussed to quickly transport the required amount
in Section 4. of preventive kits before the commence-
ment of the “lockdown“.
3.1.5. Processes 5. At last, an appropriate schedule of tasks
depending on the number of available
The following points highlight the crucial pro-
manufacturing units is prepared based
cesses (see Figure 2) involved in the collab-
on the TA algorithm. The scheduling
orative manufacturing of COVID-19 preven-
aims at reducing the costs involved in
tive kits from geo-distributed manufacturing
the production of the entire COVID-19
locations considering the forecasting infor-
preventive kits of a region.
mation of the RF Prediction Engine:
1. At first, a smart city authority or an
enterprise initiates the interest to pur- 4. Cost-Efficiency
chase preventive kits of COVID-19 in Mechanism
order to protect their residents or to
sell them in shops of their jurisdiction. The cost-efficiency, while manufacturing the
2. Second, the Information Collector cloud COVID-19 preventive kits, is tasked by the
service is invoked. The service gets ac- application of collaborative product manufac-
cess to the nearest COVID-19 data repos- turing processes. Besides, the collaborative
itory. The data is parsed, tidied, and efforts are guided by the Scheduler entity of
formatted as per the requirement of the the RFCCPM framework. It applies TA al-
RF Prediction Engine of the framework. gorithm for identifying a cost-efficient task
3. Third, the tasks are finalized depending schedule given the number of tasks and the
on the available manufacturing units product manufacturing units. This section
that are accessible within the vicinity explains the TA-based scheduling approach
and the manufacturing options of dif- while preparing the schedule.
ferent product categories.
4. Fourth, the quarantine information of
different locations, where the accessi-
4.1. TA Scheduling Approach
ble manufacturing units are located, is TA of the RFCCPM framework initially col-
predicted using the RF Prediction engine lects the list of products 𝑃𝑖 to be manufac-
of the framework. Thus, the schedul- tured and the corresponding manufacturing
ing of tasks to a particular location could sites 𝑀𝑠 . The products are often expressed in
be decided on considering the future different categories which are represented as
issues of the virus – i.e., if the number 𝐶𝑗 ∈ 𝑃𝑖 . The production of an entire COVID-
of COVID-19 cases would be higher in 19 preventive kit is represented as a set of
a location/region, there is a high possi- tasks. For example:
bility of a requirement of more number
of preventive COVID-19 kits. Thus, the C2{P1}, C1{P2}, C3{P3}, C1{P4},
production of products could be accord- C2{P5}, C1{P6}, C2{P7}, C2{P8}
ingly increased by rerouting manufac- ---> Task Sets
turing tasks to multiple manufacturing MS3, MS2, MS7, MS10, MS11,
locations in a collaborative manner. Sim- MS9, MS4, MS1
ilarly, predicting the information could
4.1 TA Scheduling Approach 106
old, final threshold, threshold value, number
of iterations, and the reduction step size of
TA are assigned to the algorithm.
In the meantime, the costs involved (𝐶𝑜𝑠𝑡)
for manufacturing the individual category of
products from various manufacturing sites 𝑀𝑠
and their corresponding manufacturing time
𝑇𝑝 are collected in the cloud-based mongodb
database. In addition, the distance 𝑑𝑚 between
the manufacturing units of various locations
is stored in the database.
Next, depending on the task set produced
by the TA algorithm, the solutions are ob-
tained for each manufacturing task set 𝑆𝑡 –
i.e., i) the cost 𝑇 𝐶𝑡𝑠 involved in the produc-
tion of an entire COVID-19 preventive kit of a
task set is calculated depending on the avail-
able manufacturing costs. Note that the prod-
ucts could be of different categories in the
COVID-19 preventive kit; ii) similarly, the to-
tal manufacturing time 𝑇 𝑇𝑝 of the task set
Figure 3: Threshold Accepting Scheduling Algo- is evaluated; and, the total distance 𝑇𝑑𝑚 in-
rithm – Flowchart volved for transporting the manufactured prod-
ucts between the manufacturing units in a
task set is calculated. Depending on these
where, C represents categories of P prod- values, the combined objective 𝐶𝑂 of a task
ucts. For instance, a face mask product man- set is calculated as shown in the equation 4.1.
ufactured using N95 in the manufacturing site
𝑀𝑆3 is represented as C2{P1}–𝑀𝑆3; a face 𝐶𝑂 = ∑(𝑇 𝐶𝑡𝑠 , 𝑇 𝑇𝑝 , 𝑇𝑑𝑚 ) (1)
shield manufactured using polycarbonate in
𝑀𝑆2 is represented as C2{P1}–𝑀𝑆2. A com- The 𝐶𝑂 of each task set 𝑆𝑡 in a population
plete list of products and their categories is set 𝑃𝑂 is evaluated in an iterative manner.
shown in Table 1. In the above task set, eight However, only the best task set is stored in
products are included. the database considering the minimal manu-
Next, multiple sets of manufacturing tasks facturing costs of the production tasks. Once
is formulated in one population 𝑃𝑂 of TA – when the entire population set 𝑃𝑂 is evalu-
i.e., each population set has collections of task ated, the local best task set 𝐿𝑜𝑐𝑎𝑙𝑆𝑡 is recorded
sets 𝑆𝑡 . Each manufacturing task set 𝑆𝑡 , the along with the obtained manufacturing time
preventive kit, consists of one category of the or costs. Next, the populations are perturbed
variety of products such as face mask, face with newer combinations and their correspond-
shield, gloves, head cap, shoe cover, thermome- ing 𝐶𝑂 of task sets 𝑆𝑡 is recorded. The thresh-
ter, sanitizer, and package. Besides, the ini- old value 𝑡ℎ of TA is reduced while increasing
tialization parameters that are required for the populations. Whenever the obtained 𝐶𝑂
the scheduling algorithm such as initial thresh- reaches the pre-assigned minimal 𝐶𝑂 value
107
or whenever the threshold value 𝑡ℎ attains Table 1
the final threshold value, the algorithm stops Categories of Products for COVID-19
the further creation of iterations or evalua- Sl.No Product Type
tions. Finally, the global best task set 𝐺𝑙𝑜𝑏𝑎𝑙𝑆𝑡 1 MaskC1 Cloth-based
is recorded based on the minimal 𝐶𝑂 values 2 MaskC2 N95
among the local task sets 𝐿𝑜𝑐𝑎𝑙𝑆𝑡 . The pic- 3 MaskC3 Face printed
torial representation of the TA algorithm is 4 FshieldC1 Polycarbonate
shown in Figure 3. 5 FshieldC2 Disposable
6 FshieldC3 Kid Type
7 GloveC1 Latex Gloves
5. Experimental Results 8 GloveC2 PVC Gloves
9 GloveC3 Plastic Gloves
This section manifests the importance of the 10 HcapC1 Non-Woven
RFCCPM framework. At first, the experimen- 11 HcapC2 Cloth type
tal setup is explained; next, the accuracy ob- 12 ScoverC1 Disposable
tained due to the RF Prediction engine is re- 13 ScoverC2 Non Woven Type
vealed; and, at last, the identification of the 14 ThermoC1 Fancy
15 ThermoC2 Wall mounted
manufacturing task schedule considering the
16 SanitizerC1 Alcohol-based
availability of manufacturing units and the 17 SanitizerC2 herbal
costs involved is disclosed using TA algorithm. 18 PackageC1 Plastic
19 PackageC2 Leather
5.1. Experimental Setup
The experiments were carried out at a DELL information of a particular location in the Ker-
precision tower 7810 machine of the IoT cloud ala state of India. The datasets were collected
research laboratory. The machine utilizes the from the COVID-19 repository of the Kerala
4.15.0-106-generic kernel Ubuntu version. The government site [12]. It was modified with
predictions were carried out by prediction al- appropriate latitude and longitude informa-
gorithm written in R programming language tion for the manufacturing locations of con-
version R4.0.0 and the services were written sideration. The datasets had values recorded
using golang version v1.14. The entire ex- from 1.3.2020 to 26.6.2020.
periments were carried out considering the
list of products and their categories illustrated 5.2.1. Validation Results
in Table 1; Nineteen MSMEs of fourteen loca-
tions within one government agency of Ker- At first, the manifestation of utilizing the RF
ala were utilized for calculating the manufac- algorithm while predicting the possible num-
turing costs. ber of quarantine candidates at a location was
validated. To do so, fifty percent of the ob-
servations were utilized for creating a train-
5.2. RF Predictions
ing model and the other fifty percent of the
In order to find the requirement for produc- observations were tested using the RF algo-
ing the number of COVID-19 preventive kits rithm.
at the required prices or budget, predictions Figures 4 depicts on the training and the
were undertaken to forecast the quarantine testing values of the number of quarantine
candidates of four different manufacturing MSME
5.2 RF Predictions 108
Kasaragod
Wayanad
0 30 60 90 120 0 30 60 90 120
DateID DateID
(a) Quarantine Cases in Kasaragod (b) Quarantine Cases in Wayanad
Thiruvananthapuram
Kottayam
0 30 60 90 120 0 30 60 90 120
DateID DateID
(c) Quarantine Cases in Kottayam (d) Quarantine Cases in Thiruvananthapuram
Figure 4: Validation Results of RF for Four Manufacturing Locations
locations, namely, Kasaragod, Wayanad, Kot- 5.2.2. Accuracy Improvements
tayam, and Thiruvanthapuram. The y-axis of
The prediction accuracy, which is measured
the Figures describes the number of quaran-
as 𝑅 2 values, was evaluated to understand if
tine cases for these locations. The x-axis dis-
further improvements are possible while tun-
cusses the increasing date of COVID-19 which
ing the algorithm-specific parameters. It was
is represented as unique identifiers.
observed that there were a few possibilities
The training values are represented in points
such as including more number of variables
and the tested values are shown in blue lines.
while dividing at each tree node (𝑚𝑡𝑟 𝑦) to
It could be observed from the figures that the
improve the prediction accuracy. Accordingly,
training and testing values are almost inclined
experiments were held by increasing the num-
to one another – i.e., the prediction accuracy
ber of variables for establishing the training
is better for the experiment.
models. Table 2 discloses the prediction ac-
Hence, it is proven to utilize an RF algo-
curacy values that were observed while con-
rithm for predicting the COVID-19 quaran-
ducting experiments with 𝑚𝑡𝑟 𝑦=2, 𝑚𝑡𝑟 𝑦=5,
tine cases which relate to the number of re-
𝑚𝑡𝑟 𝑦=10, 𝑚𝑡𝑟 𝑦=15, and 𝑚𝑡𝑟 𝑦=20, for four-
quired COVID-19 preventive kits in a city or
teen manufacturing locations of Kerala.
a location of different geo-spatial monitoring
As seen in Table 2, it was inferred that the
points.
prediction improvement of over 5 percent is
5.2 RF Predictions 109
Table 2
Prediction Accuracy Improvements in RF Algorithm
Sl.No Location mtry=2 mtry=5 mtry=10 mtry=15 mtry=20
1 Kasaragod 0.9459097 0.9733548 0.9774419 0.9796228 0.9796228
2 Kannur 0.9711467 0.9766201 0.9708988 0.9773691 0.9773691
3 Wayanad 0.9178817 0.9319097 0.9448657 0.9482104 0.9482104
4 Kozhikode 0.949926 0.9516512 0.9603881 0.9583597 0.9583597
5 Malappuram 0.9659479 0.9661579 0.96704 0.9681757 0.9681757
6 Palakkad 0.9541766 0.9570016 0.9545269 0.9490768 0.9490768
7 Thissur 0.9732814 0.9693427 0.9663719 0.9622127 0.9622127
8 Kochi 0.9260813 0.9397723 0.9371033 0.9331185 0.9331185
9 Idukki 0.9305873 0.9288297 0.935149 0.9213636 0.9213636
10 Kottayam 0.9479298 0.9635435 0.9720244 0.9841577 0.9841577
11 Alappuzha 0.9690321 0.9720363 0.977422 0.9779831 0.9779831
12 Pathanamthitta 0.9578222 0.9545819 0.9585167 0.957121 0.957121
13 Kollam 0.8263694 0.810251 0.7684122 0.7723276 0.7723276
14 Thiruvananthapuram 0.9043191 0.9008584 0.8888427 0.8484054 0.8484054
noticed in several locations – for instance, note Table 3
the prediction accuracy observed in Thiru- Number of Quarantine Cases in Kerala Districts
vananthapuram manufacturing station while SlNo Location 35𝑡ℎ Day
improving 𝑚𝑡𝑟 𝑦 values from 2 to 20; sim- 1 Kasaragod 433.52
ilar is the case with the other manufactur- 2 Kannur 1196.2
ing locations such as Kollam, Kottayam, and 3 Wayanad 394.02
Kasaragod. 4 Kozhikode 252.35
5 Malappuram 415.76
5.2.3. Prediction Results 6 Palakkad 605.71
7 Thissur 291.02
Finalizing the algorithm-specific parameters 8 Kochi 2470.02
to obtain higher accuracy – i.e., 𝑚𝑡𝑟 𝑦=20 and 9 Idukki 612.55
number of trees 𝑛𝑡𝑟𝑒𝑒=100, predictions were 10 Kottayam 1540.22
achieved for the future. The obtained predic- 11 Alappuzha 199.48
tion results were illustrated in Table 3. 12 Pathanamthitta 147.6
From Table 3, smart city officials or health 13 Kollam 629.7
department or the concerned officials could 14 Thiruvananthapuram 1084.24
decide to instruct Micro-Small-Medium-Enterprises
(MSMEs) for manufacturing COVID-19 pre-
ventive kits in a cost-efficient manner. However, the recommendations about which
For example, the smart city officials of Thiru- manufacturing MSME needs to manufacture
vanthapuram shall decide to procure 20 COVID-the products in a cost-efficient manner would
19 preventive kits for the thirty-fifth day based be dependent on the TA algorithm of the frame-
on the recommendations of the RFCCPM frame-work. Besides, the TA parameters define the
work – i.e., a production of 20485 preventive tuning of these recommendations in an ele-
kits has to be manufactured at a lower cost. gant manner.
5.3 Manufacturing Jobs – Cost Efficiency 110
Figure 5: Local Best CO Obtained over 50 Populations
5.3. Manufacturing Jobs – Cost value and the manufacturing cost identified
Efficiency for different sites vary for different popula-
tions – i.e., the algorithm searches for the MSMEs
It is mandatory to tender works to different or manufacturing units that offer a minimal
manufacturers depending on the quality and manufacturing cost. Even if a lower cost is
the cost involved in the production of the prod- identified in any one of the task set of a pop-
ucts. Questions such as i) Who needs to be ulation, the task sets were perturbed in or-
offered the tender? Whether one manufac- der to search for the best possible solution
turer could handle manufacturing all prod- in the consecutive iterations or populations.
ucts? This subsection identifies the job sched- In average, the manufacturing cost obtained
ule that considers the cost-efficient manufac- among local best task sets 𝐿𝑜𝑐𝑎𝑙𝑆 is Rs. 11013.
𝑡
turing of products using the TA algorithm. From the available local best values of task
The following TA parameter settings were sets, the global best value 𝐺𝑙𝑜𝑏𝑎𝑙𝑆 is calcu-
𝑡
utilized while searching for the cost-efficient lated by the TA algorithm. It is noticed that
collaborative COVID-19 preventive kit man- the minimal 𝐶𝑂 value is found as 5454; the
ufacturing units or MSMEs: minimal manufacturing cost is Rs.3734 for man-
Threshold = 0.0 ufacturing one preventive COVID-19 kit; and,
Initial_Threshold = 0.099 the manufacturing time is 1620 seconds. The
Final_Threshold = 0 task set that levied the global best result as
Threshold_Reduction = 0.001 per the TA algorithm 𝐺𝑙𝑜𝑏𝑎𝑙𝑆𝑡 is given be-
Number of Iterations = 1000 low:
Population Size = 50 C3{P1}, C1{P2}, C2{P3}, C1{P4},
C2{P5}, C1{P6}, C1{P7}, C1{P8}
The local best value 𝐿𝑜𝑐𝑎𝑙𝑆𝑡 obtained for
MS7, MS3, MS5, MS13, MS10,
the experiments consisting of 50 populations
MS4, MS4, MS3
with 1000 iterations of TA algorithm is shown
in Figure 5. As seen, the face mask of category 3 is sched-
The 𝐶𝑂 value is calculated using Eqn. 4.1. uled for the manufacturing site 𝑀𝑆7; face shield
It could be observed that the local best 𝐶𝑂 of category 2 is scheduled in 𝑀𝑆3; hand gloves
111
of category 3 is scheduled in 𝑀𝑆5; hood cap References
of category 1 is scheduled in 𝑀𝑆13; shoe cover
of category 2 is allotted to 𝑀𝑆10; thermome- [1] Abdul Waheed, M. Goyal, D. Gupta, A.
ter of category 1 is scheduled in 𝑀𝑆4; sani- Khanna, F. Al-Turjman and P. R. Pin-
tizer of category 1 is allotted to 𝑀𝑆4; and, the heiro, CovidGAN: Data Augmentation
packaging of category 1 is scheduled in 𝑀𝑆3. Using Auxiliary Classifier GAN for Im-
The detailed information of the categories of proved Covid-19 Detection, in IEEE Ac-
the products that are under consideration in cess, Vol. 8, pp. 91916–91923, 2020.
this article is listed in Table 1.
[2] Albahri A.S., Rula A. Hamid, Jwan k. Al-
The cost-efficiency of the identified global
wan, Z.T. Al-qays, A. A. Zaidan, B. B.
best task set 𝐺𝑙𝑜𝑏𝑎𝑙𝑆𝑡 when compared to the
Zaidan, A O. S. Albahri, A. H. AlAmoodi,
average manufacturing cost values of the lo-
Jamal Mawlood Khlaf, E. M. Almahdi,
cal best results, which were identified from
Eman Thabet, Suha M. Hadi, K I. Mo-
the population sets of the TA algorithm, is
hammed, M. A. Alsalem, Jameel R. Al-
recorded as 66 percent.
Obaidi, H.T. Madhloom, Role of biolog-
ical Data Mining and Machine Learning
6. Conclusion Techniques in Detecting and Diagnos-
ing the Novel Coronavirus (COVID-19):
The rapid dynamics of the COVID-19 pan- A Systematic Review, Journal of Medical
demic has manifested the requirement of in- Systems, Vol. 44, No. 122, 2020.
novations in various sectors, including the man-
[3] Ali M. Hasan, Mohammed M. AL-
ufacturing/fabrication sector. This article pro-
Jawad, Hamid A. Jalab, Hadil Shaiba,
posed an RFCCPM framework that combines
Rabha W. Ibrahim, and Ala R. AL-
the Random Forest and Threshold Accepting
Shamasneh, Classification of Covid-19
algorithm for enabling collaborative manu-
Coronavirus, Pneumonia and Healthy
facturing of COVID-19 preventive kits in a
Lungs in CT Scans Using Q-Deformed
cost-efficient manner. The framework was
Entropy and Deep Learning Features,
evaluated considering the COVID-19 quaran-
in Entropy Journal, Vol. 22, No. 517,
tine information and MSME enterprises of four-
doi:10.3390/e22050517, pp. 1 – 15, 2020.
teen manufacturing locations in Kerala, In-
dia. The necessity of the RFCCPM framework [4] Alok Baveja, Ajai Kapoor, Benjamin
was manifested through experiments that re- Melamed, Stopping Covid-19: A
vealed a cost efficiency of 66 percent for the pandemic-management service value
identified job schedule. chain approach, Annals of Operations
Research, Vol. 289, pp. 173–184, 2020.
Acknowledgment [5] Ayan Chatterjee, Martin W. Gerdes, and
Santiago G. Martinez, Statistical Explo-
The authors would like to thank IIIT-Kottayam rations and Univariate Timeseries Anal-
officials, AIC-IIITKottayam, and AIM officials, ysis on COVID-19 Datasets to Under-
for providing constant support through out stand the Trend of Disease Spreading
the research / entrepreneurial career. and Death, in Sensors, Vol. 20, No.3089,
doi:10.3390/s20113089, pp. 1–28, 2020.
References 112
[6] Beniamino Di Martino, Valeria Di [14] Enrique Hernández-Orallo, P. Manzoni,
Traglia, and Ivan Orefice, Semantic C. T. Calafate and J. Cano, Evaluating
Representation of Cloud Manufacturing How Smartphone Contact Tracing
Services and Processes for Industry 4.0, Technology Can Reduce the Spread
in procs. of CISIS 2019, AISC 993, pp. of Infectious Diseases: The Case of
817–826, 2020. COVID-19, in IEEE Access, Vol. 8,
pp. 99083–99097, doi: 10.1109/AC-
[7] Bokolo Anthony Jnr, Use of Telemedicine CESS.2020.2998042, 2020.
and Virtual Care for Remote Treatment
in Response to COVID-19 Pandemic, in [15] Florin Anton, Theodor Borangiu, Sil-
Journal of Medical Systems, Vol. 44, No. viu Răileanu, Silvia Anton, Nick Ivă-
132, 2020. nescu, and Iulia Iacob, Secure Sharing
of Robot and Manufacturing Resources
[8] Bouhamed Heni, COVID-19, Bacille in the Cloud for Research and Develop-
Calmette-Guérin (BCG) and Tuberculo- ment, in proc. of RAAD 2019, AISC 980,
sis: Cases and Recovery Previsions with pp. 535–543, 2019.
Deep Learning Sequence Prediction, in
Ingénierie des Systèmes d’Information, [16] Furqan Rustam et al., COVID-19 Future
Vol. 25, No. 2, pp. 165–172, 2020. Forecasting Using Supervised Machine
Learning Models, in IEEE Access, Vol.
[9] Breiman L., Random Forests in Machine 8, pp. 101489–101499, doi: 10.1109/AC-
Learning, Vol. 45, pp. 5–32, 2001. CESS.2020.2997311, 2020.
[10] Brintha N.C, Shajulin Benedict, and [17] Gajamohan Mohanarajah, Dominique
Winolyn J., Resource Allocation in Cloud Hunziker, Raffaello D Andrea, Markus
Manufacturing using Bat Algorithm, in Waibel, Rapyuta: A Cloud Robotics
International Journal of Manufactur- Platform, in IEEE Transactions on
ing Technology and Management, In- Automation Science and Engineer-
derscience publishers, https://doi.org/10. ing, Vol. 12, No. 2, pp. 481–493, doi:
1504/IJMTM.2020.107309, Vol. 34, No. 3, 10.1109/TASE.2014.2329556, April 2015.
pp. 296–310, 2020.
[18] Gunter Dueck and Tobias Scheuer,
[11] COVID-19 Status Report from WHO, Threshold Accepting: A General purpose
in https://www.who.int/emergencies/ optimization algorithm appearing supe-
diseases/novel-coronavirus-2019/ rior to simulated annealing, J. of compu-
situation-reports, accessed in June 2020. tational physics, Vol.90, No. 1, pp. 161-
[12] COVID-19 Status in Kerala, https: 175, 1990.
//dashboard.kerala.gov.in/, accessed in [19] Mohammed Abdel-Basset, R. Mo-
June 2020. hamed, M. Elhoseny, R. K. Chakrabortty
[13] Dae Jun Ahn and Jongpil Jeong, Design and M. Ryan, A Hybrid COVID-19
and Analysis of OpenStack Cloud Smart Detection Model Using an Improved
Factory Platform for Manufacturing Big Marine Predators Algorithm and a
Data Applications, in proc. of ICCSA Ranking-Based Diversity Reduction
2019, LNCS 11620, pp. 53–61, 2019. Strategy, in IEEE Access, Vol. 8, pp.
79521–79540, 2020.
References 113
[20] Nita H. Shah, Ankush H. Suthar, and [26] Siavash Valizadeh, Omid Fatahi Valilai,
Ekta N. Jayswal, Control Strategies to Mahmoud Houshmand, and Zahra
Curtail Transmission of COVID-19, in Vasegh, A novel digital dentistry
International Journal of Mathematics platform based on cloud manufactur-
and Mathematical Sciences, Vol. 2020, ing paradigm, in Int. J. of Computer
No. 2649514, pp. 1–12, 2020. Integrated Manufacturing, 2019.
[21] Piotr Staszkiewicz, I. Chomiak-Orsa [27] Shajulin Benedict, V. Vasudevan and
and I. Staszkiewicz, Dynamics of the R. S. Rejitha, Threshold Accepting
COVID-19 Contagion and Mortality: Scheduling Algorithm for Scientific
Country Factors, Social Media, and Mar- Workflows in Wireless Grids, Fourth
ket Response Evidence From a Global International Conference on Networked
Panel Analysis, in IEEE Access, Vol. Computing and Advanced Information
8, pp. 106009–106022, doi: 10.1109/AC- Management, Gyeongju, pp. 686–691,
CESS.2020.2999614, 2020. doi: 10.1109/NCM.2008.38, 2008.
[22] Prateek Khandelwal, Anuj Khandel- [28] Silviu Raileau, Florin Anton, Theodor
wal, Snigdha Agarwal, Deep Thomas, Borangiu, Silvia Anton, Maximilian
Naveen Xavier, Arun Raghuraman, in Nicolae, A cloud-based manufacturing
urlhttps://arxiv.org/pdf/2005.05287.pdf, control system with data integration
accessed in June 2020. from multiple autonomous agents, in
Computers in Industry, Vol. 102, pp.
[23] Qian Liu, Dexuan Sha, Wei Liu, Paul 50–61, 2018.
Houser, Luyao Zhang, Ruizhi Hou, Hai
Lan, Colin Flynn, Mingyue Lu, Tao [29] Thomas Favale, Francesca Soro, Mar-
Hu, and Chaowei Yang, Spatiotem- tino Trevisan, Idilio Drago, Marco Mel-
poral Patterns of COVID-19 Impact lia, Campus traffic and e-Learning dur-
on Human Activities and Environment ing COVID-19 pandemic, in Computer
in Mainland China Using Nighttime Networks, Vol. 176, No. 107290, pp. 1–9,
Light and Air Quality Data, in Remote 2020.
Sensing, Vol. 12, No. 1576, pp. 1–14,
[30] Tim Weil and San Murugesan, IT Risk
doi:10.3390/rs12101576, 2020.
and Resilience Cybersecurity Response
[24] Qiang Chen, Chen Min, Wei Zhang, Ge to COVID-19, in IT Professional, pp. 4–
Wang, Xiaoyue Ma, Richard Evans, Un- 10, 2020.
packing the black box: How to promote
[31] Vinton G. Cerf, Implications of the
citizen engagement through government
COVID-19 Pandemic, in Communica-
social media during the COVID-19 cri-
tions of the ACM, Vol. 63, No. 6, pp. 7,
sis, (to appear), in Computers in Human
2020.
Behavior, Elsevier, in https://doi.org/10.
1016/j.chb.2020.106380, 2020. [32] Wang Y., Haiyan Hao, Lisa S.P., Ex-
amining risk and crisis communications
[25] Richard. F. Sear et al., Quantifying of government agencies and stakehold-
COVID-19 Content in the Online Health ers during early-stages of COVID-19 on
Opinion War Using Machine Learning, Twitter, Computers in Human Behavior,
in IEEE Access, Vol. 8, pp. 91886–91893, Vol. 114, No. 106568, 2021.
2020.