International Journal of Production Research

ISSN: (Print) (Online) Journal homepage: www.tandfonline.com/journals/tprs20

Design and development of automobile assembly

model using federated artificial intelligence with
smart contract

Arunmozhi Manimuthu, V. G. Venkatesh, Yangyan Shi, V. Raja Sreedharan &

S. C. Lenny Koh

To cite this article: Arunmozhi Manimuthu, V. G. Venkatesh, Yangyan Shi, V. Raja Sreedharan
& S. C. Lenny Koh (2022) Design and development of automobile assembly model using
federated artificial intelligence with smart contract, International Journal of Production
Research, 60:1, 111-135, DOI: 10.1080/00207543.2021.1988750

To link to this article: https://doi.org/10.1080/00207543.2021.1988750

© 2021 The Author(s). Published by Informa

UK Limited, trading as Taylor & Francis
Group

Published online: 26 Oct 2021.

Submit your article to this journal

Article views: 8646

View related articles

View Crossmark data

Citing articles: 35 View citing articles

Full Terms & Conditions of access and use can be found at

https://www.tandfonline.com/action/journalInformation?journalCode=tprs20
INTERNATIONAL JOURNAL OF PRODUCTION RESEARCH
2022, VOL. 60, NO. 1, 111–135
https://doi.org/10.1080/00207543.2021.1988750

Design and development of automobile assembly model using federated

artificial intelligence with smart contract
Arunmozhi Manimuthu a , V. G. Venkateshb , Yangyan Shic , V. Raja Sreedharan d and S. C. Lenny Kohe
a Nanyang Technological University, Singapore; b EM Normandie Business School, Metis Lab, Le Havre, France; c Department of Management,

Macquarie Business School, Macquarie University, Sydney, Australia; d Rabat Business School, Université Internationale de Rabat, Rabat,
Morocco; e Advanced Resource Efficiency Centre and Management School, The University of Sheffield, Sheffield, UK

ABSTRACT ARTICLE HISTORY

With smart sensors and embedded drivers, today’s automotive industry has taken a giant leap in Received 19 April 2021
emerging technologies like Machine learning, Artificial intelligence, and the Internet of things and Accepted 24 September 2021
started to build data-driven decision-making strategies to compete in global smart manufacturing. KEYWORDS
This paper proposes a novel design framework that uses Federated learning-Artificial intelligence (FAI) Artificial intelligence;
for decision-making and Smart Contract (SC) policies for process execution and control in a com- blockchain; federated
pletely automated smart automobile manufacturing industry. The proposed design introduces a machine learning; original
novel element called Trust Threshold Limit (TTL) that helps moderate the excess usage of embedded equipment manufacturer;
equipment, tools, energy, and cost functions, limiting wastages in the manufacturing processes. This smart contract
research highlights the use cases of AI in decentralised Blockchain with smart contracts, the com-
pany’s trading policies, and its advantages for effectively handling market risk assessments during
socio-economic crisis. The developed model supported by real-time cases incorporated cost func-
tions, delivery time and energy evaluations. Results spotlight the use of FAI in decision accuracy for
the developed smart contract-based Automobile Assembly Model (AAM), thereby qualitatively limiting
the threshold level of cost, energy and other control functions in procurement assembly and manu-
facturing. Customisation and graphical user interface with cloud integration are some challenges of
this model.

1. Introduction
electronic controller units, actuators, and embedded soft-
Digital transformation and technology adoption tend to ware are critical in handling the generated data resources
enhance the quality and quantity of assembly, distribu- from individual equipment and processes. Industries
tion, and manufacturing in a fully automated smart man- widely use artificial intelligence and machine learning
ufacturing enterprise (Manimuthu et al. 2021). All these mainly for data processing and analysis, and decision
technology-driven approaches require smart infrastruc- making (O’Leary 2013), whereas big data and the Internet
ture and customised business plans with market strate- of Things (IoT) for decision analytics and data collec-
gies to boost the production line-ups, thereby enhancing tion, respectively. Data-driven decision-making provides
their manufacturing capabilities (Jain, Shao, and Shin solid evidence in improving productivity and enhancing
2017). Smart manufacturing involves domain-related the manufacturing and assembly processes, thereby helps
technology adoptions that target an achievable decision in monetary gain and accountability in real-time (The-
in the manufacturing ecosystem (Elverum and Welo orin et al. 2017). Many convolutional methodologies of
2016). Some smart technologies that are actively used in today’s industrial practices are getting a smarter trans-
today’s modern and fully automated industries include formation due to these technological advancements and
Artificial Intelligence, Bigdata, Blockchain, Robotics, and digital adoption, in particular the increasingly prominent
Machine Learning (Koh, Dolgui, and Sarkis 2020; Singh, role of blockchain, artificial intelligence and machine
Rathore, and Park 2020). learning (Liao et al. 2017; Xu, Xu, and Li 2018; Dolgui
These technologies assist in collecting, processing, and et al. 2020; Koh, Orzes, and Jia 2019; Pournader et al.
assembly to polishing, fitting, and distributing data to 2020; Koh, Dolgui, and Sarkis 2020). All the processes
the commercial and industrial markets. Smart sensors, are fully automated, sophisticated software tools and cus-

CONTACT S. C. Lenny Koh S.C.L.Koh@sheffield.ac.uk

© 2021 The Author(s). Published by Informa UK Limited, trading as Taylor & Francis Group
This is an Open Access article distributed under the terms of the Creative Commons Attribution-NonCommercial-NoDerivatives License (http://creativecommons.org/licenses/by-nc-
nd/4.0/), which permits non-commercial re-use, distribution, and reproduction in any medium, provided the original work is properly cited, and is not altered, transformed, or built upon
in any way.
112 A. MANIMUTHU ET AL.

tomised control methodologies help industries achieve industries in predicting and projecting their target pro-
efficiency (Gupta et al. 2020). duction and manufacturing procedures not to be any
Based on the infrastructure and manufacturing capac- loss to initial investments of sectors. Second, to our
ity, technologies can be incubated and handled effectively best knowledge, the literature lacks a deliberation on
during every transformation stage available in the assem- integrating smart contracts for controlling parameters
bly and testing process(Manimuthu and Ramadoss 2019). such as energy and cost values of different components,
Rather than investing in machinery and goods, industries especially for complex environments such as automobile
focus more on technology adoption. It helps them miti- manufacturing, which warrants the assistance of smart-
gate the market risks that directly affect their profits in decision framework. Third, the literature remains far
a competitive industrial environment (Khan and Byun from reporting the application of federated learning sys-
2020). Besides, skilled labour and smart infrastructure tems in a real-time manufacturing systems perspective,
design prove that digital adoption can make technology- though the domain receives some focus only in recent
oriented process enhancements profitable, forecasting times (Lu et al. 2020; Pokhrel and Choi 2020). To address
the market trends and investment capabilities. Such these gaps, the study proposes the research question: How
transformation plays a critical role in enhancing sustain- to design and deploy Federated Learning-Artificial intel-
ability and quality assurance besides trust management in ligence (FAI) assisted smart decision-making system for
the production and manufacturing processes(Yu Zhang automobile manufacturing environments?
and Wen 2017). The main objective of this federated learning model is
Besides, contractual formalities and guaranteed return to introduce a nominal range called Trust Threshold Limit
of investment using smart control and operational strate- (TTL) that helps the system sustain any business pro-
gies help industries sustain global markets. Apart from cess/method with minimum freedom from excess usage
market risk, the small and medium enterprises (SMEs) in terms of energy and cost without facing losses. Our
concentrate more on curtailing the production and man- work defines TTL as a maximum limit that industrial
ufacturing cost and energy consumption. Such envi- processes use to minimise process losses. It provides the
ronments involve consistent investment plans and flex- functional entity value, including the permitted level of
ible return policies with suppliers and developers. Now purchase and energy usage compared with their max-
the focus is on leveraging smart contracts (SCs), a self- imum risk through the smart-decision framework. All
executing decentralised blockchain-based procurement these available attributes are modelled and deployed in
mechanism towards data transparency. It helps monitor the developed design, unique to production industries.
and control third-party interventions, hidden broker- The study is a pioneering one for the automotive man-
ages, real-time consumption, and unauthorised activities ufacturing industry in multiple aspects. First, it deliber-
(Wang et al. 2019; De Giovanni 2020). Thus, from pro- ates how the smart contract is involved in the control,
curement to design and from processing to control, all execution, and legalisation of manufacturing and distri-
the critical elements involved in supply chains need to bution of spare parts and components required for the
be closely monitored before deploying and testing the automobile manufacturing process (Magazzeni, McBur-
latest digital technologies relying on a data-driven and ney, and Nash 2017). Second, the study deliberates the
collaborative model (Xu and Dang 2020). This cooper- effectiveness of using machine learning, especially feder-
ative mode allows sharing the datasets from a centralised ated learning, for computing suitable TTL values for each
data repository, often referred to as a federated learning tool, method, and component in manufacturing environ-
system, which handles product movement, energy con- ments. Third, the study developed an AI-enabled Auto-
sumption, and other real-time data through embedded mobile Assembly Model (AAM) that stresses the need and
systems (Treleaven, Brown, and Yang 2017; Zheng et al. importance of IoT and ML-based data-driven decision-
2020). However, the deliberation on its (federated system) making. Thus, it offers a perspective on the role of nego-
relevance to practice is still at the nascent stage though it tiable entities such as smart contracts in processing real-
leaves scope for diversified objectives. time purchase and demand information (Yuanyu Zhang
Against this background, this study recognises a few et al. 2019). Use cases of AAM include analyzing the pro-
research gaps to explore the integration of manufactur- ductivity and distribution when SC and TTL are in place.
ing processes with smart systems at process and module Critical elements like energy, cost function, time, and
levels. First, the current models do not consider thresh- productivity are remarkably improved using AAM as a
old levels for key parameters such as energy consumption reference framework in the industry.
and individual component/module manufacturing costs. The remainder of this paper is organised as follows.
Under market eventualities, these values will assist the Section 2 reviews the recent literature on blockchain
 INTERNATIONAL JOURNAL OF PRODUCTION RESEARCH 113

and smart manufacturing processes. Section 3 dis- can be mapped and modelled using suitable policy-based
cusses model components and control parameters. blockchain applications. The application allows manu-
Section 4 explains the modelling and design method- facturers to track their product movements and traceabil-
ologies. Experimental study and use case analysis are ity on the supply of goods and equipment among com-
shown in section 5. Next, section 6 elaborates the sim- panies, vendors, and suppliers (Gonçalves et al. 2021).
ulation and experimental analysis. Section 7 discusses Risk accessibility, especially on large-scale production
the findings, and Section 8 deliberates both theoretical and manufacturing with suitable network and blockchain
and managerial implications. Finally, Section 9 concludes aided supply help ease the system-centric smart automa-
with limitations and future scope. tion environment (Manimuthu et al. 2019; Manimuthu
and Dharshini 2021; Mohanta et al. 2020)). SC and
market policy aims to showcase component and fin-
2. Literature background
ished product-wide deliverables. Industries also encour-
The review process adopted a systematic exercise age legitimate and legally available SC, especially for risk-
by retrieving the relevant publications from differ- prone industries (Baryannis et al. 2019). Thus, companies
ent repositories through the following search strings: need to functionally incubate, implement and plan suit-
(TITLE-ABS-KEY (blockchain) OR TITLE-ABS-KEY able infrastructure design in manufacturing and supply
(smart AND contracts) OR TITLE-ABS-KEY (federated chains to effectively utilise the blockchain in investment,
AND artificial AND intelligence) OR TITLE-ABS-KEY retail, export-import, and pre-and post-processing of raw
(machine AND learning) OR TITLE-ABS-KEY (artifi- materials in real-time (Min 2010).
cial AND intelligence) AND TITLE-ABS-KEY (automo- Enterprise resource management and control system
tive AND assembly) OR TITLE-ABS-KEY (automobile strategies need to have long upgrades. In few cases, these
AND assembly)) AND (LIMIT-TO (DOCTYPE, ‘ar’) OR strategies require infrastructure and automation invest-
LIMIT-TO (DOCTYPE, ‘re’)) AND (LIMIT-TO (LAN- ments to utilise the blockchain primitives in their work-
GUAGE, ‘English’)) AND (LIMIT-TO (SRCTYPE, ‘j’)). place efficiently. As a nascent technology, blockchain with
The below sub-sections summarise the recent literature other smart technologies needs to be effectively han-
around blockchain, artificial intelligence in manufactur- dled and efficiently used in today’s modern manufac-
ing, and federated learning systems. turing industries. The reliable and smart manufacturing
process requires a focus on product reception to sup-
ply (Manimuthu et al. 2021). Customer markets need to
2.1. Blockchain in manufacturing
be studied to ensure a sustainable market index in the
Manufacturers are actively involved in developing and growing global trade environment. According to agile
deploying blockchain in their industrial practices due to and smart manufacturing companies’ market valuation
many functional aspects: operations control, risk man- and customer index, reliable and quality assured prod-
agement, active process control, and additive manu- uct delivery between manufacturers and consumers is
facturing supply chain(Macrinici, Cartofeanu, and Gao very narrow. This gap requires substantial steps that war-
2018). These factors help them gain more visibility rant long-term business trend forecasting and market
in market risk and obtain asset tracking availabil- investments (Manimuthu et al. 2021).
ity throughout their market venture capitalisation of Blockchain has proven potential in influencing man-
investments in real-time. It can influence the design, ufacturing and supply chain practices. Critical elements
control, process, assessment, and delivery of products such as energy consumption, cost, processing, and con-
at both ends of the industry. With global supply- trol strategies are integrated towards sustaining the prod-
demand forecasting, auditing the control strategies and uct supply under the required specifications. Policy for-
cost function are critically important, thereby foster- mulation, government regulations, and legal advisories
ing customers’ trust and market sustainability(Allam become part of industrial practices; blockchain with
and Dhunny 2019). All these key scaling factors will operational and trade policies help in guiding and pro-
adversely affect the system performance (Yuanyu Zhang liferating these regulatory gaps (Andoni et al. 2019).
et al. 2019). Some of the industrial impacts of using Recently, Venkatesh et al. (2020) propose a blockchain
blockchain include supply-chain monitoring, data-driven environment to track the social sustainability dimen-
decision making, asset tracking, control, process manage- sion of manufacturing industries. Thus, the objectives of
ment, trust validation, quality and quantity assurance, supply chain transparency, effective management of tech-
market risk forecasting, energy and cost management, pol- nologies, and deploying smart innovation tools together
icy formation, and risk management. Thus, from procure- offer a win-win situation to all participating industries
ment to delivery, all the raw materials to finished goods of a blockchain-enabled automation environment. In
114 A. MANIMUTHU ET AL.

addition to the supply-chain practices and operations, and processing techniques performed by labours. Thus,
fostering the state-of-the-art design that helps to get transforming the workplace with manual process into
maximum potential with available resources becomes a robust and customisable AI-driven operation tends
instrumental (Ivanov, Sokolov, and Dolgui 2014; Saberi to prove their betterment in areas like pre-and post-
et al. 2019). Summing up all the critical elements for the processing, control and application delivery, and other
design, a novel smart and comprehensive operations and supply chain practices in real-time(Allam and Dhunny
business model is developed. 2019; Kolvart, Poola, and Rull 2016; Singh, Rathore, and
Park 2020).
As a core component for Industry 4.0 and smart IoT
2.2. AI in automotive industries systems, AI never stops enhancing the business model
Processing capabilities and error detection/identification where it helps to transform the industry to compete
are a few critical elements in the smart manufacturing in the global market (Omohundro 2014; Parunak 1996;
industries. A highly reliable smart system needs to be Wang et al. 2018). Thus, in today’s business model, indus-
deployed for automated and completely robotic control tries aim to incorporate AI in their designs and offer
units in assembly, control, and movements (Min 2010; smart solutions to data-driven decision-making proce-
Yuanyu Zhang et al. 2019; Fenwick and Vermeulen 2019; dures. Due to the advent of developing smart technolo-
Cioffi et al. 2020). Due to the enormous range of data gies, the transformation of sectors to adopt industrial 4.0
computing and fast processing capabilities, AI is being and industrial IoT standards is getting linear growth. Pro-
implemented in industries and other smart technologies duction, control, processing, and manufacturing are vital
(Refer to Table 1). A wide range of customised AI algo- areas that get boosted with these smart techniques of
rithms is readily available in the market to do various operations (Rane and Narvel 2021). AI provides an inde-
operations such that algorithms can mimic the actions pendent and stand-alone solution to numerous inventory

Table 1. Studies related to emerging technology from the automotive industry.

S.no Authors & Year Domain Focus Methods Applications
1 Gonçalves et al. (2021) Automotive Industry Decision-Making Multivariate approach Forecasting Assembly
process
2 Guo and Ryan (2021) Auto Assembly Line Risk-Averse Optimisation Mixed-Integer Programming Large Vehicles Assembly
3 Kong et al. 2021 Mobility Services Decision-Making Bloom Filter Driver Performance
Evaluation
4 Mishra, Mahanty, and Automotive Industry Servitisation Graph-Theoretic Approach Quality Concerns
Thakkar (2021)
5 Loading et al. (2021) Automotive Industry Judgment Analysis Discrete Event Simulation Manufacturing systems
6 Raj Kumar Reddy et al. Automotive Industry State Of the Art Clustering Analysis Vuca World
2021
7 Shahbazi and Byun 2021 Automotive Industry Data Analytics Hybrid Prediction Models Smart Manufacturing
8 Alavian et al. (2020) Automotive Industry Continuous Improvement Industry 4.0 Production Systems
9 Dutta et al. 2020 Blockchain Technology State Of the Art Literature Review Business Visibility
10 Gupta et al. 2020 Autonomous Vehicle State Of the Art Literature Review Cybersecurity
11 Hadian et al. (2020) Automotive Industry Decision-Making Vikor – MCDM Outsourcing
12 Jabbar et al. 2020 Automotive Industry Decentralized Platform Internet Of Vehicles Vehicle Communication
13 Kim, Jung, and Hu 2020 Automotive Industry Smart Contracts Deep Learning Dashcam Application
14 Xia et al. 2020 Vehicle Technology Blockchain Transactions Bayesian Game Electronic Trading
15 Xu and Dang (2020) Automotive Industry Causal Analysis Digital Cause & effect Diagram Knowledge Management
16 Copeland et al. 2019 Vehicle Sourcing Edge Communication Network Function Virtualization Essential Services
17 Samarakoon et al. (2020) Automotive Industry Decision-Making Federated Learning Vehicular communications
18 Sharma, Kumar, and Park Automotive Industry Distributed Framework Node Selection Algorithm Smart City
2019
19 Erfurth and Bendul (2018) Automotive Industry Manufacturing networks Cross Case Study Global Manufacturing
20 Kumar et al. (2018) Integrated Planning Production control Modeling And Simulation Production Scheduling
21 Sharma, Kumar, and Park Vehicle Technology Distributed framework Blockchain Smart Environment
(2019)
22 Jain, Shao, and Shin (2017) Automotive Industry Data Analytics Performance Analysis Process Modelling
23 Wei et al. (2017) Auto Part Manufacturer Optimization Algorithm Support Vector Machines Manufacturing Process
Quality
24 Keivanpour, Ait-Kadi, and Automotive Industry Decision-Making Fuzzy Logic End-Of-Life Vehicle
Mascle (2017)
25 Theorin et al. (2017) Agile Manufacturing Information Systems Event-Driven Architecture Manufacturing Systems
26 Elverum and Welo (2016) Automotive Industry Innovation Management New Product Development Rapid Prototyping
27 Gupta and Vardhan (2016) Automotive Industry Equipment Effectiveness Production Cost Equipment effectiveness
28 Lacerda, Xambre, and Automotive Industry Continuous Improvement Value-Stream Mapping Component Manufacturing
Alvelos (2016)
Note: MCDM – Multi-criteria decision making.
 INTERNATIONAL JOURNAL OF PRODUCTION RESEARCH 115

and supply-chain problems existing within the indus- with sensory data in a single window for Grouping-
trial environment. All these fully automated frameworks > Processing- > Analysis- > Interpretation- > Mapping- >
run with the help of customised AI-IoT algorithms with Modelling- > Training- > Feature Extraction- > Decision
potential risk assessment features. Thus, manufacturing Making. Data are stored after all these processes in a
and supply chain industries are constantly looking for local data repository or data centre (based on industry
these robust and customisable smart innovations to be infrastructure and data generated).
adopted in their workplace (Bhamra, Dani, and Burnard In the developed FAI model, regression methods
2011; Gupta et al. 2020; Macrinici, Cartofeanu, and Gao are used a few data processing sections for data nor-
2018; Singh, Rathore, and Park 2020). malisation and probability on error identification/rec-
tification during sampling and training. This training
model depends on few critical parameters as follows:
2.3. Federated learning
The collaborative learning mode of machine learning (1) Batch size or Repository sampling values (B)
helps iteratively train vast amounts of data from each (2) Number of data entries/iteration (D)
of the embedded sensors and devices installed inside (3) Total number of nodes or components (N)
every manufacturing equipment. Data from the vendors, (4) Deployed FAI models (M)
suppliers, and stockholders are collected and processed (5) Training range (R)
before transferring to the following sections for further (6) Data sampling rate (S)
operations. IoT plays a significant role in the instant
delivery of accumulated data from the devices. Such data Depending on the data obtained from vendors, sup-
have critically been used in training to obtain functional pliers, and stocks, values are modelled and trained. If
insights on different goods and commodities in real- datasets are generated from a single period, sampling
time. One of the key factors that assist in designing and rate needs are assigned for training and processing to a
modelling accumulated sensory data is standard console separate FAI model. In such cases, training values must
built-up, where data can be stored as packets with times- be correlated and normalised to avoid sampling errors
tamps (O’Leary 2013). Thus, centralised federated learn- affecting the overall outcome for a particular component
ing is performed whenever the data is obtained using in the system model(Gupta et al. 2020; Treleaven, Brown,
IoT-enabled smart sensors that tend to get accumulated and Yang 2017; Zheng et al. 2020). Thus, to avoid such
on an arbitrary basis and logged based on a specific cus- discrepancies at the output, further modelling is done
tomised AI algorithm. The majority of reported studies using batch processing. Each batch consists of compo-
were conceptualised and developed using conventional nents of the same type obtained from multiple vendors
techniques and approaches. The review confirms that the at the same frequency for processing. Each of the tools
literature remains far from using a federated-learning sys- and features is identified and grouped with specific iden-
tem in manufacturing to improve efficiency, even though tification (ID) numbers. Consider an item ‘A’ obtained
there is a growing interest in recent times. This provides from vendor ‘X’ at a time ‘t’. Then following notations
an opportunity to study the application of FAI as a smart are provided by the user to retrieve a particular A at any
decision-making framework in an automotive assembly time instance ‘ti ’.
process. Apart from the consummate design and plan-
ning features, using federated models provides lot of (1) Total
number of components from particular vendor
managerial insights for effective operations and business = X = X1+X2+ . . . ..+Xn
practices in real-time. Some of the obsoleted models exit- (2) Total number of samples recorded at the time ‘t’ = Y
ing in current practices can be easily tuned, restructure (3) Number of clients involved in the same process = Z
or replaced using the learning models. A detailed lit- (4) Training dataset value for X at any time ‘t’ = W (R,
erature studies about design elements, decision models B) where W is the weight of a particular dataset.
and operation support entities are tabulated (Table 1) to (5) Error accumulated during training = E
give broader insights AI, and Blockchain in industrial use
cases. Thus, during the training and while applying FAI,

3. Model design background and key 

n
Y(Z)
Tested sampling rate for component X =
parameters W(Rn, Bn)
t=1
Federated Learning with AI (FAI) algorithms helps in E
+ + S (1)
bringing different steps of algorithms and codes along W(Rn, Bn)
116 A. MANIMUTHU ET AL.

n 
  Both regulatory and technical policies can be bundled
 S D
Where S = + (2) together using SC as FAI helps in foreseeing the risk
t=1
B+N S
factors in prediction, planning, procurement, purchase,
Samples tested correction and modelling using FAI at manufacturing, assembly, and delivery(Cioffi et al. 2020;
time ‘t’ is obtained using the equation (1). Similarly, the Khan and Byun 2020). Thus, to validate this blockchain
dataset for all the components is modelled and stored during the loading, unloading, and distribution of com-
in a shared data repository. This data repository can ponents and tools in and out of the industry and to study
be revoked using their unique packet identifier (PID). SC’s limitations, functionalities, and features, FAI helps
Thus, using suitable PID and sampling test results, the manage the potential risk in the manufacturing process.
correct proportion of tools and components from pro- Data with the least possible error values help fetch the
cessing to production can be revoked from the storage. desired benefits of utilising AI and SC in the workplace.
These capabilities are provided with every section until A new normalisation phenomenon is introduced in the
the final product is delivered to the clients(Parunak 1996; manufacturing process to cut off accumulated errors,
Allam and Dhunny 2019;). Apart from the information and data mismatch from obtained datasheets to pro-
handling and data processing, FAI helps to handle the cessed data (Nofer et al. 2017; Sayeed, Marco-Gisbert,
stock comparison and pricing values. This helps monitor and Caira 2020). This method is data-centric, and nor-
volatile market share and commodity pricing options of malisation helps mitigate the system’s accumulated data
components or raw materials subjected to market invest- errors without affecting the policy to a greater extent.
ment risk and economic crisis. Thus, their cost of pro- This novel error limiting factor is called Trust Threshold
duction, usage, wastages, and energy involved in utilising Limit (TTL).
them are critically modelled and planned at each section
(O’Leary 2013; Wang et al. 2018). 3.2. Trust Threshold Limit (TTL)
TTL refers to the maximum limit that any goods and
3.1. AI and Smart contract components involved in assembly, manufacturing, and
Algorithms and learning methodologies involved in delivery can be experimentally utilised with minimum
smart manufacturing entities vary with time. Thus, this wastage or losses. Thus, as the name indicates, TTL sets
basic featurization endorses the smart contract (SC) the threshold limit for all the tools and devices actively
usage that extends the production process visibility to participating in product delivery. SC depends on the
engineer and train the functions and standard oper- permitted TTL limit for each entry in the data socket,
ating procedures. SC is framed between the suppliers exceeding which the chance of loss in the product market
and the manufacturers (mutual agreement) on the risk is high. With TTL value, the responsibility of the policy-
of raw materials, processing, testing, and validation of maker and FAI function is to optimise the scaling factor
goods(Baryannis et al. 2019; Magazzeni, McBurney, and for the component within that particular limit(Mohanta
Nash 2017; Min 2010). SC has provisions to include the et al. 2020; Sayeed, Marco-Gisbert, and Caira 2020; Yu
insurance schemes available for the assembly, inventory, Zhang and Wen 2017; Yuanyu Zhang et al. 2019). The
and product delivery. Since the entire operation is han- novel decision aid model and the threshold limit values
dled using FAI and SC, the arbitration of data obtained helps the operations, production and logistics in their
from all the market shareholders is modelled for effec- purchase, procurement, distribution, and delivery. Apart
tive management in real-time. Some of the segments that from the normal supply-chain and logistics operations
involve SC in the production and manufacturing in the (Ivanov et al. 2016, 2019), the developed state-of-the-art
industry include: TTL values helps the existing business model to incubate
them for better component and product movements in
(1) Raw Materials their real-time industrial environment.
(2) Financing and Stock valuation Example: Consider equation (1) where testing sam-
(3) Insurance and market risk ples for product X are obtained by FAI and weights of X.
(4) Delivery and Transportation When it comes to TTL, the maximum limit value needs to
be obtained using the same weights and data distribution.
All these four SCs are actively enabled during the man- Thus TTL (X) can be modelled as follows:
ufacturing and assembly process. AI models help find 
n 
n
W(R, B) E
suitable negotiation factors for cost, energy, market risk TTL(X) = +
by predicting and forecasting the feasibility of SC poli- t=1
Y t=1
1
(B + D) ∗ R+Z ∗ X
cies before the commencement of works in the industries. (3)
 INTERNATIONAL JOURNAL OF PRODUCTION RESEARCH 117

1  X for decision-making and validation in real-time. SC is

Where X  = B
+
D+R W(R, B) provided with the market risk knowledge and resource
 utilisation metrics during all the stages of data accu-
+ Z ∗ W(R, B)/ X+E (4) mulation, processing, and control, thereby ensuring the
market standards for better product delivery. Policymak-
For a manufacturing and production unit to operate
ers and standard decision-making units critically evalu-
with full potential both in terms of benefits from sup-
ate the outcomes at every stage. Finally, the end prod-
ply and process/procurement/purchase, equations 3 and
uct is delivered with the same features to the consumer
4 will significantly help. Design testing and analysis fall
market(Gupta et al. 2020; Min 2010).
under the same umbrella of TTL. Timestamp and compo-
nent ID are used as a reference entity to model the FAI at
different sections with other industrial procedures (Min
4.1. Stage 1: Data collection and classification
2010; Nofer et al. 2017).
The entire process of product design and delivery relies
solely on the accuracy of data gathered from each of
4. Methods
the available components in the system. Discrete datasets
The model framework is structured under four stages from each supplier, vendor, and stockholder are actively
of sequential processes, as shown in Figure 1. The pro- collected along with their market risk policy to frame
cess gets started with data collection, followed by data their suitable SC. As these data are entirely obtained from
normalisation. The processed data is tested and analyzed the IoT-enabled smart sensors, these data require a lot
using suitable testing and validation methods (support of pre-and post-processing functions to be performed.
vector machine learning). All these segmented and nor- Information about manufacturing, warranty, composi-
malised datasets are then modelled and trained using FAI tion, maker’s policy, structural details are logged and

Figure 1. Research Flow.

118 A. MANIMUTHU ET AL.

modleled during the collection process. Sensors, actua- 4.2. Stage 2: Grouping and normalisation
tors, and other embedded electronic equipment collect
All the logged data and support vectors are moved to the
the data and efficiently hand them to the central data
next section of data normalisation. The data with errors
repository for further processing and analysis. After the
are identified and removed/correlated with the next least
data collection, the accumulated data are modelled and
possible error values required for processing the tools.
mapped with their unique ID. The equipment data are
Thus, error rectification is completely done during this
stored in their respective local data repositories for long-
process of data normalisation. In some worst-case scenar-
term referencing and policy formulation. These data are
ios, if the schematic of a particular component consec-
then retrieved for the 1st stage of processing, called Trust
utively fails to meet the expectations, then the financial
validation. The data obtained and stored from the sup-
blockchain is invoked using the SC. One of the most
pliers and stockholders are cross-referenced and verified
prominent examples of today’s automotive world for such
to prove confidence and originality regarding measure-
financial SC and their smart policy is recalling Honda
ments, validity, defects, or other deformities. If any flaws
Model cars due to their faulty airbags. Under such cir-
are identified during this stage of trust validation, SC will
cumstances, group policies are shared between all the
immediately help manufacturers raise a replacement or
involved parties, from suppliers to manufacturers. Due
refund the stocks. This blockchain feature needs to be
to these unexpected circumstances, the loss incurred is
mentioned at the time of policy formation. This can be
equally shared by all the commodity vendors and compa-
materialised and followed as a standard defect identifi-
nies (Sayeed, Marco-Gisbert, and Caira 2020; Zheng et al.
cation and originality check of goods and commodities
2020). If the defect is identified in the product (Airbag
covered under the same risk assessment policy(Khan and
as in Honda cars), then the faulty product id is revoked
Byun 2020). In addition to the existing literature (Ivanov,
from the database for vendor identification. If the ven-
Sokolov, and Dolgui 2014, 2019), the impact of engag-
dor SC is not assigned under such financial blockchain,
ing smart technologies are explained in detail with the
then the company is entirely liable for the incidents. If
help of AI and blockchain. This sheds lights for new
SC is derived in favour of the manufacturer, then the
research and design perspectives of altering the indus-
vendors will take the whole responsibility and address
trial design for performance enhancement and monetary
the incidents with compensation or replacement of the
benefits.
entire automobile itself. Thus, data error normalisation
The datasets are modelled using the support vec-
and schematic verification play a major role in training
tor machine learning model (SVML) in the post-trust-
and SC formation.
validation phase. During this unsupervised ML train-
ing, the dataset is provided with classifiers where each
classifier offers the information on the source of goods, 4.3. Stage 3: Control, verification, and analysis
timestamp, and purchase details. All these data are mod-
elled and mapped under SC group policy intended to Two of the critical stages in the manufacturing and
avoid the market risk practice. Depending on the sup- assembly in the fully automated smart manufacturing
plier’s datasheets and procedures, the manufacturer has industry are verification and analysis. In this stage, the
the freedom to plan additional policies if the system’s per- component datasets are evaluated for fixing their TTL
formance predicts a better ratio than expected during the range. This range plays a significant role in energy usage,
policy formation(Mohanta et al. 2020; Yu Zhang and Wen supply-demand management, cost, and flawless product
2017). delivery. Normalised data obtained after verifying the
Each support vector is identified and mapped with schematics is shared with the assembly and verifica-
their classifiers and logged in the same database for each tion section. TTL of the component is set before start-
reference and identification. All the mapped dataset is ing the process (Magazzeni, McBurney, and Nash 2017;
logged in a customised database for easy and smooth Yu Zhang and Wen 2017). Once the process is initi-
identification and utilisation. During the end-product ated, TTL for that tool is picked from the data model
delivery, these classifiers and the support vectors are for fitting and assembly. During the assembly and fit-
invoked from the system database for final checking and ting schemes, the dataset is verified for TTL; thereby,
clearance. In case of any contradiction to the proposed the reference limit for the whole processing mecha-
details, S and FAI values, will be cross-checked and mod- nism solely depends on the limit set by TTL. In this
ified as per the risk assessment SC in the system (Mag- system driven control and processing, all the accumu-
azzeni, McBurney, and Nash 2017; Yuanyu Zhang et al. lated datasets are going through series of processes as
2019). follows:
 INTERNATIONAL JOURNAL OF PRODUCTION RESEARCH 119

(1) Fitting credibility in real-time(O’Leary 2013; Wang et al. 2018).

(2) Assembly Working performance and service satisfaction from the
(3) Verification end-users play a vital role in framing and fixing the SC
(4) Analysis once the product is available for usage. A Timeline for the
(5) Testing next bulk production relies entirely on the target audi-
(6) Polishing under the same TTL value from the SVML ence’s satisfactory report analysis, which takes time to
training used for deriving SC. obtain in real-time.

All the SC under this stage is assigned under the insur-

ance and market risk blockchain category. SC associated 4.4. Stage 4: Decision making, FAI, and product
with insurance blockchain helps the retail business ven- dispatch
tures to take part in the investment procedures, thereby The next critical and most crucial stage of AAM design
helps in boosting their market share and business devel- is the decision-making by which the real-time testing is
opment processes (Yuanyu Zhang et al. 2019). Market lia- evaluated. Data obtained from this stage is used as a ref-
bility is insured, and the IoT devices closely monitor the erence for all SC forms considered a baseline scheme
actions performed at each section. As the data generated for identifying the TTL values. The convolutional data
from embedded devices are a continuous process, classi- processing method offers more error functions than the
fiers and the support vectors are assigned instantly irre- FAI scheme (Giancaspro 2017; Macrinici, Cartofeanu,
spective of the functionality and operations performed at and Gao 2018). Apart from the analysis and training, the
different stages of the assembly and delivery process. SC and training datasets are arbitrarily cross-referenced
If any mismatch occurs during the production and for smooth data interpretation and model evaluation.
delivery stages, the dataset is retrieved from the local From the study, key findings include market valuation for
database, and their TTL is critically examined with their each product, cost and distribution quantity to individ-
support vectors. Thus, errors are eliminated. Compar- ual shareholders, profit-loss margin and transportation
ing with the existing industrial setup, engaging smart and insurance coverage, quantity, and quality of goods
technologies will give maximum insights about the cru- delivery, etc. By doing this model evaluation using TTL
cial designs and process automation. Joining up with the and FAI, the decision can be obtained for the AAM
available resources, the operations and control schemes design that includes energy usage, cost, and components
can be modelled for maximum benefits. For every stage quantity(Allam and Dhunny 2019).
of product design and fitting, the SC can be provided
with suitable regulatory primitives, thereby bringing all
the datasets under a single database (Zheng et al. 2020). 5. Case study
Insurance companies offer higher flexibility in share
exchanges for policymaking and production strategy Implementation of the developed model in a smart indus-
analysis subjected to market risk and investment returns. trial environment helps to understand the benefits of
For investment returns, the predicted performance by the using TTL and FAI in real-time. As the company can
FAI schemes holds a high hand in the market demand incubate the necessary blockchain techniques, the infras-
with a better supply chain ratio. At the same time, all tructure requirement has diversified requirements. The
the involved third-party vendors and suppliers try to operations and control scheme of the developed AAM
seek more capital investors. To ensure assured returns are thoroughly evaluated in real-time. Implementing the
from the entitled policies, commonly available risk fac- SC in the workplace without halting the operations in
tors include: the industry is quite challenging, but the time taken to
implement is significantly less than any other existing
(1) Raw material cost conventional models.
(2) Supply-demand ratio
(3) Goods quality
5.1. Company background
(4) Delivery time
(5) Economic crisis Small and Medium Enterprises (SMEs) are actively
(6) Market credibility involved in emerging the latest technologies in their
working strategies. One such SME is situated in Europe,
Thus, this stage involves a high risk-high return if the where it assembles various automotive parts manufac-
expected product reaches the target audience within the tured across the globe. They have an automated assem-
stipulated time. Thereby effectively balances operational bly, distribution, and small-scale manufacturing unit
120 A. MANIMUTHU ET AL.

(locally develop few components). As a new commer- component classification and analysis. Data are modelled
cialisation strategy in 2020, they started implementing and associated with the Raw Material Blockchain. Stage
the SC Blockchain technique in their vehicle assem- 2 comprises embedded equipment and machinery data
bly and distribution process. Initially, they tested the that helps in the investigation, grouping, and classifi-
scheme for their domestic warehouse operations, which cation of data. This section is closely associated with
assembles embedded components essential for product financial and insurance blockchain. The final stage in the
delivery, as their concentration was only on transporta- design of AAM for XZY company is the final assembly of
tion blockchain. Step-by-step, they have expanded their the product, where all the sections and stages are involved
market to neighbouring countries like Malaysia, India, in achieving the desired outcome. In this stage, 3, TTL,
and few others (Baryannis et al. 2019; Yuanyu Zhang and SC help finalise the FAI values by which the company
et al. 2019). Since the company is about eight years stakeholders will plan and execute their business plans in
old, the global response index for testing and emerg- real-time.
ing new techniques in their design is quite challeng- The vehicle design has many smart embedded com-
ing in the beginning. However, their initiative is rela- ponents. Thus their working conditions, testing, and
tively new to the commercial vehicle market. Due to field assembly values need to be processed and modelled
confidentiality and workplace design ethics, the com- (Yuanyu Zhang et al. 2019). Depending on the indus-
pany name is kept anonymously as XYZ company. The trial standards, market needs, and commercial value, the
company’s goal is to deliver commercial trucks to var- quality and quantity of tools, software, and devices are
ious customers across the world. The company aims to mapped and installed. Figure 2 exhibits details of com-
implement AI models, IoT, and other ML techniques mercial vehicles assembled in the plant. The vehicle con-
are driven decision-making. There are three stages in sists of much electronic equipment that acts as embed-
the vehicle assembly and delivery process. Each stage ded agents and assists the IoT system in data collection,
involves new technologies and standards AI methodolo- distribution„ and storage.
gies that aim to bring a solid result-oriented profitable
design. The developed design helps the company com-
pete effectively in the local and global market for the 5.2. Business case
long term. Component procurement, vendor selection, equipment
The other objective of implementing SC and AI is to identification and market valuations, and the best com-
provide a hassle-free and risk-less automobile delivery mercial value for the product are critical zones focused
environment that facilitates desirable profit. This leads in the design. Data collection, processing, and analy-
to less impact due to socio-economic crises or any other sis based on various factors like energy usage, product
strategies from existing competitors. The trucks assem- loss/damages, and cost of production are critically consid-
bled from this factory vary from 10–18 wheels, and each ered in the modelling, evaluation, and implementation of
truck goes through the same assembling stages using AI them in the workplace. Implementing the AAM helps to
and Blockchain(Nofer et al. 2017). Stage 1 consists of improve productivity, competitive market pricing for the

Figure 2. Components involved in Industry Grade Commercial Truck.

 INTERNATIONAL JOURNAL OF PRODUCTION RESEARCH 121

vehicle delivered, and security for all the investors. Fur- Table 2. Project charter.
ther investigation includes 1. Component classification Business Case: Statistical modelling and efficient cost reduction and
based on loading and unloading time and market cost. 2. energy consumption in stages like loading, transportation, procurement,
production, and assembly. Sensor-based embedded application processes
Analyse the pricing values from all the vendors, suppliers, are fully automated. Modelling and design evaluation is done using R
and stockholders. 3. It assists in arranging the compo- programming. Industry-grade simulation software helps to derive the
required values and thresh limits for each component involved in the
nents with their respective logistic sections. 4. Processing design. Market procurement with minimum gain margin will provide
and evaluating the datasheets received from all the com- better scaling values when all the components are modelled using the
ponent delivery members. 5. Identify the component’s same procedures. This helps reduce the stock pricing and market valuation
by attracting many investors without the risk of the vehicle’s monetary
originality, insurance policy, and smart contract details loss and commercial value.
for practical usage at different sections in the XYZ com- Problem Statement: Implementing cost-effective procurement, processing,
and product delivery strategy using FAI and SC. Effectively utilise IoT and
pany. This helps effective utilisation of IoT devices and Machine Learning functions for data collection, analysis, estimation, and
smart components on the manufacturing floor. Data col- performance evaluation in a fully automated assembly and delivery unit
in the automobile industry.
lected are transferred to different processing and control Goal Statement: Minimising excess resource utilisation, cost, and energy
zones for interpretation, classification, and analysis. For in all the critical areas covered under the smart contract. Introducing a
novel performance measurement index helps the industry maintain a
this purpose, AAM uses support vector machine learning permitted level of profit-safe margin during purchase and holdings called
and classifiers. Trust Threshold Level (TTL).
Thus, data are modelled, mapped, and processed Team Requirement: Since the industry is fully automated, only 6–10 skilled
labourers are sufficient to assist the devices, data processing centres,
based on industrial requirements and product’s commer- and robotic platforms. Skillset includes troubleshooting in embedded
cial standards in real-time. Required data statistics are software tools, design testing, and high-level experience handling
machine learning algorithms and R models.
immediately processed and modelled using FAI to obtain Period: 13 months (Full Time) Jan 2020 to Feb 2021 for implementation,
the TTL value for any individual component involved in modelling, testing, and deployment of developed AAM in the workplace
with fully established FAI-TTL assistance with Smart Contract.
the system design. Stock availability, component usage Equipment used: Sensors, Actuators, Electronic Controller Units, Near Field
metrics, energy consumption at the workplace, and cost Communication Devices, RFID (Radio Frequency Identification) tags, and
of production are modelled individually in the AAM IoT enabled smart Transceiver units.
Software and Data Repository used: Local Data storage and processing
design. Thus, the final product assembly and delivery unit, Automotive Grade Design software like MATLAB and R modelling
involves a lot of complex data processing and training tool.
mechanism that assures better market value for the vehi-
cle without any losses in per capita investment for each
investor. As all the schemes modelling are done through from a coordinated and modelled central controlling unit
a smart contract, the AAM framework requires stage- where all the federated models are deployed for process-
by-stage data processing that uses IoT, machine learning ing and calculation in real-time.
models that significantly assist in curtailing losses at var- Once the details about the tools, components, and
ious stages of the assembly process. Table 2 shows the other miscellaneous elements involved in the design,
project charter carried out for AAM design and deploy- assembly, and production process are received, they were
ment in the XYZ company. labelled, and tagged for easy classification, identification,
and storage. The stocks are stored in the local reposi-
tory, and the same is modelled using Support vector ML.
5.3. Data collection and vector classification
In this process, each vendor ID is mapped and logged
In this design stage, goods, software, tools, and com- with their devices and tools. This information is cat-
ponents from all the parties are unloaded and verified egorised and provided as samples for the ML model
for their originality and standards in real-time. All stock for training and feature extraction purposes. Once their
values, component lists and specifications are labelled, detailed analysis is obtained, the stockpile is provided
tagged, and received at the unloading section. Stock- with classifiers. The stockpile and their unique compo-
list, storage requirements, processing features, marking nent ID can be invoked and identified at any stage of their
details like time and validity of the contract are mapped requirement in real-time. For this analysis and categori-
and stored with their respective suppliers (Guo and Ryan sation, AAM uses R packages, and the dataset is analyzed
2021). This stage uses a unique identifier for easy recov- for feature extraction using automotive software tools
ery in case of loss, damage, or faulty components. Data (Manimuthu et al. 2021). Key findings from this stage
about the suppliers are kept confidential throughout the include:
process as the SC policy will give additional safety and
security details from the manufacturer. Cross functional (1) Fault identification
embedded equipment can be combined together for data (2) Component Classification
collection and processing. Energy reading is obtained (3) Mapping of tools and Devices with their suppliers
122 A. MANIMUTHU ET AL.

(4) Product policy identification 5.6. Spare parts inspection and trust validation
(5) Stocks and Storage range and capacity index
The pivotal stage is to classify different tools and com-
(6) Market cost and Energy consumed during Unload-
ponents from manufacturers. During this stage, mar-
ing and storage in dedicated facilities
ket valuation and the component’s purchase cost are
critically studied. Forecasted market value is quoted as
From this stage, the raw materials are assigned and
the best market index during their training in real-
model with their specific SC type based on com-
time. All the datasets are trained and modelled using
pany requirement, market trend, and investors inter-
their unique ID and training vector classifiers. Database
est. Some primary SC types include Smart Legal con-
that holds the record about multiple components of the
tract, Decentralized Autonomous Organization (DAO)
exact specifications using SC procedures. Their label and
contract, Application logic Contract (ALC). In this AAM,
time tags are uniquely modelled, trained, and stored
components are categorised in the production process
under their unique classifiers and support vectors (Zheng
using only two SC: Smart Legal and ALC(Parunak
et al. 2020).
1996).
Once the automotive elements are classified based on
different datasheets, market values and usages are mod-
5.4. Smart legal contract (SLC) elled and trained using their support vectors and classi-
fiers. This helps identify and eliminate the data errors and
One of the commonly used blockchain elements where
accumulated processing errors during data reception and
all the elements, tools, and software investors and stake-
storage by different IoT devices installed at different crit-
holders are legally merged under one common agree-
ical industrial zones. Data play a critical role in TTL limit
ment says profit or loss needs to be shared legally under
identification for every industry-grade tool and embed-
accepted terms and conditions. This SC covers software,
ded component. Datasets with errors are processed sepa-
tools, and hardware, and the industry’s data centre for
rately rather than training with other error-free datasets.
future references. Trust among all involved parties is
Further data normalisation shows the datacentric abil-
ensured using SC. Market reliability, stock value pre-
ity of ML in processing the information among different
dictions, and product commercial value evaluations are
sections of the industry(Pokhrel and Choi 2020).
accountable and shared by all the investors using SLC.
After classification and modelling, the components
Easily accessible machine-level SC assists in tying the
are mapped with their respective section based on usage
consumer market with industries without any external
requirements in the assembly process. In this stage, errors
brokerages. Energy wastages calculation uses a digital
accumulated during the dataset training and evaluation
IoT environment. Most SMEs use this blockchain in
are updated, and new entries are stored. Sections involved
their product procurement, delivery, and market stock
in the XYZ company use this mapped dataset and their
analysis.
normalised values for smooth and easy identification of
components from multiple manufacturers. This classi-
5.5. Application logic contract (ALC) fication is based on their market valuation and stock
listings as well. Grouping and logging take place once
SC assists in using IoT devices for data collection, pro-
the trained values are available at respective data loggers.
cessing, analysis, and decision-making in automobile
Irrespective of time, cost, and energy usage, all the ele-
assembly and delivery. Most of the tools and methods
ments are logged and categorised using their classifiers
used in the AAM involve many application-specific IoT
and support vectors in this stage. Since the next stage
devices, support vectors, classifiers, and federated learn-
of vehicle design requires tools and components based
ing models. ALC helps bridge gaps in programming
on timing and assembly, these logged data are clustered
tools, system design software, and industrial standards
together (Manimuthu et al. 2021).
for assembly and production processes. ALC applies to
design and modelling. All the active components, irre-
spective of manufacturers and process, can be brought
5.7. Product design and schematic verification
under a common umbrella of SC without additional pol-
icy formation. Many investors agree to get involved in As all the datasets required for component assembly and
the logical contract without scrutinising their design and product structuring are readily available, the SC policy is
development details. Since XYZ company comes under checked before starting this process. During the process,
SME in Europe, the implementation and design pro- almost all the assembly operations will be completed,
cess is much more flexible while using ALC as one of and only the delivery of the final product will remain
the SC. in the industry. Before starting this process, product
 INTERNATIONAL JOURNAL OF PRODUCTION RESEARCH 123

Figure 3. Rendered Schematic Viewpoint for the Heavy-Duty Commercial Vehicle.

schematics, as shown in Figure 3, are critically analyzed available for usage and shared as a key reference to assem-
and evaluated deeply to come with the process initiation. ble every individual component (O’Leary 2013). In con-
The schematics include the components assembly with trast to the existing business models and industrial setup,
reference to the different viewpoints. The references are the developed design uses smart technologies, AI and
shared with all the divisions and the sections for their smart contract to assess and manage all the error prone
correct fitting and design verification in detail. Cross- zone in the business and industrial operations.
sectional views help the design process more compact
as the tools are categorised according to their require-
5.8. Product delivery
ments. Robots take care of the assembly process using
the embedded software, and R models help in data visu- After assembling the product, the usages are updated
alisation. Once verified and approved, this stage of pro- with their ID and TTL values in real-time. In this stage,
duction and assembly cannot be intervened by any of data obtained from the sensors and embedded devices are
the embedded devices in the middle(Singh, Rathore, and modelled and trained using FAI. Component’s classifiers
Park 2020). Only the emergency halt operation can be and support vectors are correlated for their usage and
performed. Steps and processes involved in this stage threshold limits in real-time. Once the process is initi-
include: ated, data obtained are relatively normalised and mapped
in parallel with their SC before getting transported
(1) Fitting to the final product delivery section(Gonçalves et al.
(2) Assembly 2021).
(3) Polishing Data about the components and tools, embedded soft-
(4) Testing ware, and smart sensors are cross verified for their market
(5) Analysis utility, licenses, and delivery cost with their intended con-
(6) Verification sumers or vendors for commercial stocking. Fault identi-
fication at this stage is crucial as the entire schematics of
In each section, ML values, trained classifiers ele- the system need to be reworked and revamped based on
ments, and the permitted storage level are recorded; So the TTL values. FAI training is again performed to restore
that there will not be any excess values in any sections. the system, just as every process involving the faulty com-
This helps to secure the system from any loss due to ponent. The component is replaced completely. If the
excess storage cost, and energy required to process and fault is identified during assembly or production, the
store the values. The product design schematics with entire items delivered during that time-stamped stage
all viewpoints, as shown in Figure 3, are kept readily and product classifiers are separated from the rest of the
124 A. MANIMUTHU ET AL.

sections. This helps secure the production without being 6. Design simulation and experimental analysis
affected much by the fault at different stages during mass
Model development, simulation, and software-based
production(Min 2010).
programming include three levels of data extraction and
Further, with TTL, the market pricing is carefully
modelling, as shown in Figure 4. In all the stages, fea-
identified and evaluated to get the best pricing value
ture extraction, error identification, rectification, model
during the final product delivery – SCs help bring qual-
processing, and data training are performed continuously
ity and quantity during the distribution of vehicles in
using SVML and FAI. The modelling scheme involves
the commercial market. Stakeholders and investors, par-
SC and TTL to ensure the proper commercial pricing to
ticipating venture capitals are provided with the vehi-
design and the required number of assembly and product
cle standardisation schemes and method of designs and
design components.
explained about the TTL and their performance in devel-
oping customer-centric smart vehicle design with an
assured profit margin to all the investors and partici- 6.1. Level 1: Data normalisation and feature
pating agencies in real-time in terms of market share, extraction
stocks and commodity and consumers commercial mar- R packages and industry-grade smart system software are
ket trust(Cioffi et al. 2020). used in bringing the full features available from every

Figure 4. Automobile Assembly Model (AAM) using TTL, FAI, and Smart Contract Blockchain.
 INTERNATIONAL JOURNAL OF PRODUCTION RESEARCH 125

component involved in the design. Training and eval- TTL and FAI while setting the FAI margin as 1.27% dur-
uation include stock lists, pricing values, and manufac- ing the purchase and installation of ECU alone. Then
turer support for smooth testing, loading, unloading, and the same process is done for every individual compo-
information extraction. The process involves fact sheets nent. The overall gain margin using FAI in the design
and design information obtained from original vendors includes a whopping 26.46% overall gain throughout the
and suppliers. This information is retrieved and modelled entire process. However, this whole FAI margin will not
using SVML, and classifiers are provided to each section be reflected the same at the end of product design and
for smooth identification(Nofer et al. 2017). Each com- delivery as the market share is prorated and has its stock
ponent is mapped and logged onto a local database for and investment fluctuations(Giancaspro 2017). The pro-
easy pick and place into different sections involved in the cedure is followed in the SME for almost ten months
process. Investors and stakeholders are advised about the unlike any other existing models, which may normally
market price. be implemented only for few months before deploying
on a large-scale industrial environment. Now it is fully
operational and exporting vehicles overseas with a solid
6.2. Level 2: Trust Threshold Limit identification and profitability range of in the body and commercial division
modelling split-up alone for each investor based on their investment
One of the critical levels of design and testing is function percentages apart from market gain and fluctuations in
evaluation and vector characterisation of components for real-time.
identifying their TTL values in the system. This TTL
references consumption, wastages, energy usage, and cost 6.3. Level 3: Federated learning AI and smart
function involved in the commercial vehicle assembly and contract
distribution process. With TTL, the loss margin is greatly
improved, and the stock pricing is kept on the desired The installation, procurement, purchase, and product
limit by all the involved component vendors. Support delivery process is subject to the cost and energy usages
vectors and the federated training values are modelled apart from market investments and stock prices. The
and smoothly channelised for best throughput as the training and evaluation of data and component’s speci-
limit is set during the initialisation process. Table 3 shows fications are based on their original suppliers (Fenwick
the impact of using TTL and SC in analyzing the risk and Vermeulen 2019; Yuanyu Zhang et al. 2019).
index in the commercial market. Before implementing The smart contract involved in the design includes
FAI, the use case of SC is not completely achieved in the the insurance policies that attempt to protect and secure
design as suitable infrastructure is not viably available. the initial investments made during the production and
Based on the contracts and mutual agreements, the pur- delivery process. In this SME, only two insurance policies
chase and string values go through their applicable SC. are incubated in the design process. 1. Pure Holding Type
Thereby the unwanted storing cost and the risk involved and 2. Intermediate Holding type.
in storing are significantly reduced. It is not mandatory
to pay additional costs, and energy wastages for mainte-
6.4. Pure holding type
nance are nullified completely. Thus, this implementation
directly reduces the risk involved in handling all forms Companies that participate in component investments
of products movements. If the SC is not implemented, and market share alone without product design and com-
the policymakers have the freedom to implement addi- ponent pricing come under this category. These invest-
tional maintenance charges along with storage and deliv- ment companies hold a significant share in the SC, and
ery cost. Almost 1.5% of the risk index is improved using the commercial market’s profit margin is also very high.

Table 3. Raw materials implemented with FAI training percentage and valuation index.
License License Smart Contract Risk Index (%) FAI Risk Index (%)
Elements (In Years) Grade Units (SC) Type (Before FAI) Margin (After FAI)
ECU 2 Industrial 25 ALC 14 1.27 9.8
Embedded Controller 1 Retail 50 ALC 25 2.14 17.5
Sensors 1.5 Vendor 120 Smart Legal 27 4.21 18.9
Actuators 1 Vendor 100 Smart Legal 30 5.31 21
Conveyors 3 Industrial 12 ALC 15 2.89 10.5
Data Stack 5 Retail 5 ALC 10 3.47 9.5
Logger 2 Retail 10 Smart Legal 5 4.01 3.75
Debugger 1 Vendor 10 Smart Legal 5 3.16 3.5
126 A. MANIMUTHU ET AL.

6.5. Intermediate holding type FAI for their best training attributes. Hence, with these
values, the best market price for each of the essential ele-
As the name suggests, instead of investing and acquiring
ments is obtained. Based on this, TTL is developed as iter-
a significant stake in the company, these companies will
ations for the next consecutive progression of matrices for
invest along with other retail investors in terms of cost,
the same companies.
energy, and component purchases. The main advantage
In this process of training and evaluation of goods and
of this type of holding is high return in the short term,
services, SC involved in the design helps to get better
but it also involves heightened risk due to market fluctu-
pricing values from the commercial market by identify-
ations. Companies in this category tend to be proactive
ing, analyzing, and estimating the pricing of goods and
in commercialising the automobile as soon as the prod-
commodities. It also the product movement from and
uct reaches the commercial market for consumer utility
to the company and commercial market respectively in
in real-time. Table 4 shows the insurance type and the
real-time.
SC values for each critical component with many ven-
The tariff rate, along with their market valuation,
dors and suppliers. In addition to these components and
are identified, modelled, and evaluated based on deliv-
tools, many other products are available in the vehicle
ery speed, charges/item, and market capitalisation(SC;
where the gross margin is shared between many local
and Enterprise value). Many vendors have their delivery
investors.
agencies for loading and unloading goods and commodi-
From table 4, it can be observed that the gross mar-
ties. Thus, TTL and FAI help identify the best pricing
gin gain for each component raised to an average of
agency for delivery and transportation in the commercial
1.5–2.3 percent irrespective of their SC and Insurance
market. Table 6 shows the delivery agency list and their
holding type. Thus, TTL and FAI help companies get bet-
enterprise valuations like market cap, enterprise value,
ter investment returns and market sustainability for the
and delivery charges before and after implementing FAI
long term.
and TTL in the workplace.
Similarly, the TTL tends to limit all the company’s
Similarly, the performance index is evaluated for the
share value based on their market trends and com-
dealers based on the revenue and gross sales. These data
mercial market fluctuations in real-time irrespective of
must be submitted as per the company policy and SC
the share contributions these companies have in the
blockchain for insurance, maintenance, and market share
vehicle design, development, and commercialisation. As
calculation in real-time. Thus, all the agencies and deal-
shown in Table 5, companies from A-G supplies different
ers produce their total retail and commercial revenue
embedded components, and the conveyors are purchased
data for FAI training and analysis. This helps to study
from companies X to Z in the local market. Due to the
and identify potential issues or threats or quality index
randomness of investments and stock prices, the valua-
and improve cost margin gain regarding distribution and
tion of each company is closely monitored and fed to the

Table 4. Insurance type and gross margin % of different components in AAM design.
Insurance Smart Contract SC Rate (%) Gross Margin SC Rate (%) Gross Margin
Elements Holding Type (SC) Type (Before FAI) (%) (Before FAI) (After FAI) (%) (After FAI)
Conveyor Pure ALC 2.23 4.20 4.68 6.59
Data Stack Intermediate ALC 3.01 5.32 7.21 8.32
ECU Intermediate Smart Legal 2.86 3.26 4.19 3.53
Sensors Immediate Smart Legal 4.33 5.32 10.35 9.81
Battery Pure Smart Legal 4.53 4.82 9.83 7.09
Software Pure ALC 2.87 2.94 3.80 3.61

Table 5. Companies market valuations for each critical component in AAM design.
Companies (Market Fluctuation %)
Elements A B C D E F G X Y Z
ECU 4.011 6.12 3.51 9.6 15.28 17.79 6.402
Embedded Controller 6.19 4.29 5.35 7.29 4.69 8.29 4.69
Sensors 3.67 4.67 3.14 8.21 4.36 7.26 3.03
Actuators 5.158 5.146 5.173 5.193 5.103 6.09 5.049
Conveyors - - - - - - - 11.93 10.29 9.143
Data Stack 5.21 6.21 5.179 4.29 9.21 7.29 6.73
Logger 4.019 9.27 7.95 7.63 6.72 6.1 4.03
Debugger 5.753 6.017 6.32 5.236 4.93 5.017 6.73
Note: Values highlighted in Red gives the best market valuation for a particular component using FAI and TTL.
 INTERNATIONAL JOURNAL OF PRODUCTION RESEARCH 127

Table 6. Transportation tracking with delivery rate of all dealers in AAM design.
Delivery Delivery Rate Market Enterprise Delivery Charges Delivery Rate
Dealer ID Charges (%) (Before FAI) Cap (%) Value (%) (%) (After FAI) (After FAI)
IN263 12.23 23.86 131.53 168.75 5.22 21.15
IN753 16.23 15.93 95.89 123.02 −16.68 13.22
IN632 17.30 14.40 89.28 114.55 −21.64 11.68
IN895 14.23 19.27 110.58 141.87 −6.67 16.56
IN412 9.21 20.31 110.76 142.11 7.44 17.60
IN724 11.04 21.42 118.15 151.59 4.54 18.71
IN244 19.23 20.15 119.96 153.91 −17.86 17.43
IN893 10.25 19.33 106.88 137.13 3.37 16.61
IN710 10.86 22.36 122.66 157.38 6.38 19.65
IN837 9.31 18.40 101.29 129.96 4.31 15.68
IN720 8.39 19.38 105.29 135.08 8.09 16.67
IN207 19.34 22.36 131.14 168.25 −14.81 19.65
IN663 14.30 23.64 132.50 169.99 −0.28 20.93
IN743 17.29 20.93 121.94 156.45 −11.83 18.22
IN552 13.70 19.30 110.20 141.39 −5.30 16.59
IN349 14.93 19.76 113.73 145.92 −7.68 17.05
IN735 15.73 24.36 137.53 176.45 −2.79 21.65
IN753 13.29 23.18 129.19 165.75 1.55 20.47
IN823 14.37 22.31 125.92 161.55 −2.46 19.60
IN634 14.20 21.34 120.90 155.11 −3.49 18.63
IN900 11.36 22.56 124.16 159.30 5.44 19.85
IN209 10.30 22.79 124.25 159.41 8.44 20.08
IN760 8.24 19.39 105.19 134.95 8.50 16.68
IN860 11.80 21.29 118.25 151.71 2.43 18.58
Note: Negative Value indicates gross margin exceeds the desired limit (Outperforms in Market Valuation).

Table 7. Dealers revenue and gross sale improvisation using FAI, Table 8. Energy evaluation during AAM design, testing, and
TTL, and SC during AAM design. evaluation.
Gross Sales Gross Sales Gross Energy Gross Energy
(Margin %) Revenue (%) (Margin Revenue Margin % Consumed/ Margin % Consumed/
Dealer ID (Before FAI) (Before FAI) %)(After FAI) (%)(After FAI) Category (Before FAI) Day (%) (After FAI) Day (%)
IN263 14.06 7.56 16.08 10.29 Loading 6.346 7.631 11.1055 6.1048
IN753 11.90 13.26 13.92 15.98 Polishing 3.21 12.78 5.6175 10.224
IN632 11.57 17.02 13.58 19.75 Assembly 6.27 15.364 10.9725 12.2912
IN895 12.72 5.77 14.74 8.49 Fitting 4.95 16.37 8.6625 13.096
IN412 11.73 8.56 13.75 11.29 Organizing 5.071 12.71 8.87425 10.168
IN724 12.71 5.80 14.73 8.53 Tunning 4.32 20.3 7.56 16.24
IN244 14.50 16.05 16.51 18.77 Testing 9.22 20.13 16.135 16.104
IN893 11.62 3.62 13.63 6.35 Control 13.432 15.019 23.506 12.0152
IN710 13.05 8.24 15.07 10.96 Unloading 11.432 8.31 20.006 6.648
IN837 10.97 4.29 12.98 7.02 Total Energy 128.614 96.2432
IN720 11.11 8.91 13.13 11.63
IN207 15.45 14.27 17.46 16.99
IN663 14.55 1.35 16.57 4.08
IN743 14.28 11.03 16.29 13.76 Apart from the cost and component distribution, it
IN552 12.58 4.56 14.60 7.29
IN349 13.12 6.87 15.14 9.59 is very much possible for the TTL to identify and ana-
IN735 15.26 4.44 17.27 7.16 lyze the energy consumed per product. Thus, Table 8
IN753 14.08 0.10 16.09 2.83 shows the reduction in energy consumption after imple-
IN823 14.02 3.29 16.04 6.02
IN634 13.57 3.81 15.59 6.53 menting TTL and FAI in the intelligent industrial unit.
IN900 13.28 7.27 15.29 10.00 It is estimated and recorded from unloading at the
IN209 13.07 10.71 15.08 13.44
IN760 11.08 9.36 13.09 12.08 warehouse throughout different production and assem-
IN860 12.87 3.40 14.89 6.13 bly stages. Thus, the loading, unloading, transportation,
assembly, fitting, and polishing costs, along with anal-
market limitations. Table 7 shows the similar gross mar- ysis and design schematic verification costs, are iden-
gin gain achieved by the dealers and commercial distri- tified, recorded, and processed. The processed value is
bution agencies after implementing FAI and TTL in their applied with FAI and SC policies for getting final mar-
operating procedures. Advancing further from the exist- ket evaluation and stock prices/unit of energy consumed
ing design (Ivanov et al. 2016), all the participating agents in the industrial process(Ivanov, Sokolov, and Dolgui
are provided with their stock movement details and pric- 2014, 2019). Figure 5 depicts the performance before and
ing values. This includes wastages, flaws and defects in after implementing TTL in the workplace. The tabulated
their goods and services. results are visually represented in figure 5 to compile and
128 A. MANIMUTHU ET AL.

Figure 5. Data visualisation and results comparison (before & after implementing TTL & FAI).

show the use case of TTL and the effect of using FAI and Blockchain into the commercial vehicle industry prac-
SC in the design process. Apart from the cost and energy, tices. Various studies on Blockchain, AI, and IoT have
the interest rate, delivery charges, and SC’s policies help detailed their usage from the industry perspective (Singh,
the industry achieve better performance and improved Rathore, and Park 2020; Yuanyu Zhang et al. 2019).
productivity in real-time. They also triggered the point that combining mul-
tiple technologies with data analytics tends to solve
many real-time problems in the industrial environ-
7. Discussions of findings
ment over many decades. Quality and performance val-
The study has designed and developed an Automobile uation, design analytics, and productivity enhancement
Assembly Model (AAM) concentrating on industrial- constitute several possible sustainability solutions to
grade commercial vehicle design. This study practically withstand in any global industrial market. In addi-
implemented and evaluated the developed design using tion to the additive manufacturing and process automa-
real-time data analytics and machine learning models to tions, the use of SC and IoT helps the industrial envi-
process, train, and test developed AAM. Before explain- ronment and active participants to obtain a detailed
ing the key findings, the study has opened doors for market response curated with the existing business
emerging technologies like IoT, AI, and SmartContract practices.
 INTERNATIONAL JOURNAL OF PRODUCTION RESEARCH 129

Data modelling, computational algorithm design, simu- as the dataset may not be sufficient to provide result-
lation, and schematic formations were again a big hurdle oriented decisions in real-time. Thus, flexible and robust
due to prevailing trust issues between stakeholders and infrastructure capable of emerging smart technologies in
investors. As the SC helps bridge this trust issue by pol- industrial practice is required in the automated indus-
icy formation and legal assistance, all the investors tend trial setup. The discussion made from the literature and
to have hassle-free investment plans in real-time. This practical industrial white papers shows the importance
design includes multiple procedures in terms of purchase, of decision-making models and flexible data process-
delivery, utility, and distribution. All the data are mod- ing metrics. This helps industries to improve resilience
elled using SVML, and Federated Learning ML techniques against fluctuating global market. Accordingly, the policy
as the data are effectively captured by the embedded IoT makers and regulatory management authorities will get a
in the industry. Developed AAM shows the use cases of clear understanding of the business process and manage
ML and IoT in fixing, evaluating, and processing all forms the operations with open information, transparency and
of data irrespective of sources without any deviation from visibility. This includes operations, logistics, distribution
their programmed functions. The programmed model and transportation.
uses a customised R programming scheme, and simu- The present AAM uses four form factors that act as
lations are done using company-owned industry-grade pillars in the design, development, and successful test-
licensing software. Thus, unbiased data is fed to the train- ing of the design scheme. It includes smart technologies
ing, testing, and evaluation model at all assembly stages and industrial standardizations. Form factors include 1 –
in real-time. data from Stock, suppliers, and vendors, 2. Smart Contract
Irrespective of industry, these methodologies can be Policies, 3. Trust Threshold Level for each element involved
implemented with a few minor customisations. Many in the production process, 4. SVML and FAI for perfor-
industries, including leather, textile, biogas, fuel cells, mance comparison, modelling, and analysis at various
and two-wheelers, can incubate these methods in their stages of design. In addition to these critical form fac-
practices to achieve profitability in the short term. They tors, other methods used in training and processing data
maintain ongoing modifications include curtailing the and stages include classification, assembly, fitting, pol-
cost of productivity and minimise energy consump- ishing, testing, analysis, assembly, grouping, logging, and
tion during their production procedures. In some other schematic verification.
cases, these techniques were exclusively used in iden- Data shared between different stages with the ML
tifying the flaws, damages, and defects in the compo- algorithms will extract the features and help in decision
nents, tools, and product design. This aims to attain making. SC comes into the picture when the devices
sustainability without losing the market margin gain or equipment are observed of any faults or damages or
within the calculated time frame. Recent reviews show malfunction during any stage of the assembly until the
that the fault identification and testing scheme evalu- product is delivered into the commercial market.
ation can be made based on industry standards and As this process is technically considered stable, the
company policy, whereas in AAM, all the investors market fluctuation always prevails while fixing the prod-
need to abide by the SC policy. This method of test- ucts’ cost and commercial value. Thus, in this AAM, a
ing and evaluation includes all forms of cost, energy novel estimation technique called TTL is implemented in
consumption, and market fluctuations. This confirms the design itself. Through this estimation scheme, all the
an eco-friendly market share dividend among all the products are modelled based on their cost, energy usage,
participating agencies as per their investment per- market value, and profit margin of utilising them in the
centage and profit margin obtained during product design. TTL has the following features that have helped
delivery. the XYZ company to have a commercial profit margin
Qualitative evaluation of every deliverable plays a sig- of 13 percent within 60 days of their commercial vehicle
nificant role in design, development, and product assem- sales. Feature includes:
bly. Many techniques other than ML includes fuzzy logic;
the neural network can explore SMEs based on their (1) Data modelling and component analysis (Vendors,
production cost and infrastructure capabilities. Predic- suppliers, stocks)
tion analytics and modelling with these techniques aim (2) Loading, Unloading, distribution, and usage valua-
to offer better performance without considering any tion cost
market investments. Thus, ultimately, improved produc- (3) Consumer satisfactory index
tion cost, commercial value gain, and market valuation (4) Market valuation and SC legal policies
become void in these industries. In some worst cases, (5) Sale valuation and investors profit margin
the decision model lags in providing the required results (6) Storages and stock listing data
130 A. MANIMUTHU ET AL.

(7) Stock purchase valuation comparison focusing on their internal costing features and func-
(8) Market pricing vs. Purchase pricing analytics tions. The developed AAM tried to integrate all forms of
functions, procedures, values, and specifications of every
These data help fine-tune each component’s threshold tool and commodity. Further, the TTL value helps iden-
value after their potential utility during the assembly and tify and process every component’s functional attribute
delivery process. AAM with TTL in the industries uses with detailed identification, analysis, and market utility.
a margin variation function with a maximum deviation Thus, it indirectly facilitates the firm to get more visibility
of 2.3-3.2 percent. TTL derived from the production and and position itself in the competitive commercial vehicle
assembly process involves market investments and com- market. Besides, FAI and TTL designed and implemented
pany stock shares that are openly available for any kind are unique and novel where both relatively bring sustain-
of investors. It helps them to take part in legal policy with able industrial practices and solutions that any industry
smart contract blockchain. can quickly adapt in their workplace. The model devel-
Data is modelled and distributed among various sec- oped, solutions provided, and dataset training function-
tions for cross-referencing, stocking, and complex com- alities are entirely customisable based on the industry
putation purposes. requirement and infrastructure support. It will signifi-
cantly contribute to all forms of supply chain practices
and operations management with assured safety. Finally,
8. Implications
the secured smart contract usages act as a backup to meet
The design and research analysis offer significant con- any legal grounds in real-time.
tributions in operations and supply chain, specifically
automotive assembly processes.
8.2. Practical implications
The study has notable managerial contributions. The
8.1. Theoretical contributions
design provides use cases to the SMEs and large-scale
First, the study provides a use case of machine learn- industries irrespective of their domain. First, it supports
ing models and smart IoT devices and bridges two the incubation and use of smart IoT devices in industrial
indigenous methodologies that contribute to produc- processes through real-time data collection and process-
tivity improvement(O’Leary 2013; Yuanyu Zhang et al. ing using machine learning algorithms. Here data act
2019). When compared with the existing industrial prac- as a critical resource in consecutive processing stages in
tices, these smart techniques notably promote the stock the IoT-enabled smart assembly unit. Grouping, vectori-
value in the commercial market. Timing, data analytics, sation, feature extraction, and data analysis involve both
fault identification, and systematic assembly process were SVML and Federated learning at different stages. They all
made throughout the production and product delivery help leverage the information about the tools, software,
process. Targeting the commercial vehicle distribution components, and devices in the assembly process.
market, SMEs focus more on reducing the losses from Secondly, in the developed design, a novel element
purchasing and procuring raw materials. The process is called Trust Threshold Limit is used in all areas of
continued till the final product is delivered. Second, the data training under the Federated Artificial Intelligence
study provides procedural guidance to create a smart (FAI) integrated framework. It supports the industrial
industrial ecosystem that involves AI-enabled smart sen- automation and analytics that can be realistically mod-
sors and machine learning practices(Manimuthu et al. elled using the emerging smart techniques. Limit for
2021) for the best market valuations to each commodity. purchase and storage can be visually made available
As the design involves stock pricing and energy con- to the concerned stakeholders using this mode of data
sumption factors, all these data are modelled with their processing and analytics. Modelling and development
respective normalisation values obtained during train- involve machine learning models that help bring the
ing, analysis, and estimation. In this way, the marginal best market values for all the components involved in
error accumulated at every stage of data processing is the design process. This training and modelling help
significantly improved. investors and the third parties involved in the design to
Third, this is the first study that emphasizes the FAI closely monitor the product prices and their stock val-
and its role in the product assembly industry. The delib- ues in real-time. Combining blockchain, federated AI,
erations in managing the product complex system and and machine learning models helps foresee the com-
support vectors for data normalisation helps to find the ponent requirements, usages, procedural functionali-
best fit values for design and testing. Fourth, it discusses ties, and data-driven decision-making models, thereby
cost-saving functions in the available industrial practices reengineering the overall product tracking system in a
 INTERNATIONAL JOURNAL OF PRODUCTION RESEARCH 131

manufacturing firm. In terms of process control and roadmap for implementing a wide range of smart tech-
design optimisation, tracking, modelling and design ver- nologies for vehicle operations, control, and assembly
ification play a major role. Valuation and the process- performance valuation using data-driven modelling and
ing capabilities of the smart systems can be studied and analysis. The study involves IoT and supports vector
compared with the existing tools for optimised business machine learning for grouping, analyzing, and classify-
operations. ing tools, components, and other supporting goods in
Thirdly, the simulation and modelling provide the the assembly process. Irrespective of the market invest-
necessary resources for FAI analysis and the TTL estima- ments, these additive manufacturing strategies will help
tion. As the market risk is involved in all design stages, production and distribution processes and quantitatively
TTL serves as a better response index starting from load- assess the market risk and investors’ returns in real-time
ing goods to product delivery in real-time. The design (Alavian et al. 2020; Guo and Ryan 2021).
shows the importance of IoT devices and ML modelling The design proposed a novel TTL value that spotlights
mechanisms in bringing the best possible values for each the use of threshold limits in the purchase, production,
component subject to their market investment and com- and product delivery. TTL is combined with federated
mercial values. learning AI mechanism to propose a smart solution for
Fourthly, the study supports the practical usage of improving profit margins. Data normalisation, vector
TTL using ML models and data analytics methods in classification, analysis, and feature evaluations are criti-
real-time for equal distribution of resources. Energy dis- cally scrutinised throughout the process of AAM design.
tribution and parallel computation provisions engage IoT TTL and FAI help fix, finalise, and set the limit values
devices, smart sensors, and electronic controllers. Imple- for purchase, storage, usage, and distribution of goods
menting smart, innovative techniques in the industrial and services commercially, irrespective of market fluctu-
practices streamline the industrial procedures and orient ations. The legal policy for security and liability is taken
them towards profit-making from market investments, care of by smart contract as all the investors and stake-
with a focus on market risk and secure product com- holders are legally entitled under standards industrial
mercialisation in real-time. Thus, in this AAM design, operating procedures. Thus, TTL helps set the limit in
TTL and FAI advance understanding of the role of stock cost, energy usage, purchase, and processing options. It
pricing, market rate, commercial product valuation with assists in building the vehicle from scratch without the
intense ideation of practical usage of cost, energy, indus- worry of losses due to socio-economic crises or market
trial standards, and smart, innovative technologies for stock fluctuations. TTL assists for energy usage, pur-
building a better sustainable industrial ecosystem. Evalu- chase of raw material, transportation, and delivery. The
ation of the existing methods provides a standard perfor- local vendor selection is mapped, modelled, analyzed,
mance metrics as tabulated in Table 8 include energy con- and tabulated, and verified in the industrial environment
sumption and margin of wastages. This becomes more in real-time.
realistic while implementing the TTL in the actual busi-
ness practices. Minimum modifications with maximum
9.1. Limitations and future directions
potential in operations, supply-chain and production
factors in real-time resource assessment and manage- The study has its limitations and offers future research
ment can be achieved. Besides, AAM will help foresee avenues. First, the study does not involve any data stor-
the requirements, usages, procedural functionalities, and age with remote access. Thus, firms can consider using
data-driven decision-making models. Using smart tech- cloud storage for data processing and remote accessibility.
nologies such as ML, IoT and Blockchain in industrial All the software and codes are purchased from third-
practices, the companies may have better accountability party vendors. They cannot be trusted all the time for
and sustain competitive commercial market index value. data reliability. Hence the data used for processing may
Thus, AAM directly facilitates restructuring of the overall have a marginal variation of permissible range that could
firm’s standard operating procedures. contribute to other errors. SC formed for this design
scheme can only permit few extendable ranges of com-
plications that include natural disasters. However, it does
not involve defects in industrial design infrastructure, as
9. Conclusion
that is taken care of solely by the company alone.
The work provides evidence that the enhanced data The simulation scheme used in the design requires
collection, processing, and control procedures help in skilled labourers, and Graphical User Interface (GUI)
efficiently handling the data generated for the manu- cannot be customised as they are not locally developed
facturing procedures. The experimental study offers a for industrial practice. TTL uses FAI and SVML data
132 A. MANIMUTHU ET AL.

processing that any other ML technique can outperform. V. G. Venkatesh is an Associate Profes-
Data acquired and processed from the IoT-enabled smart sor in supply chain management with
sensors can have their latency in data delivery. EM Normandie Business School, France.
He is a Certified Supply Chain Profes-
In the current industrial setup with more than 60–70 sional (CSCP) from APICS-SCC, USA.
percent of automation in place, the incubation and His teaching and research interests are
deployment of the developed AAM can be a hassle-free in global procurement, logistics, sustain-
action. Finally, the developed design is not integrated ability, and technology applications. His
with the existing model and warrants its customised recent publications appear in reputable outlets such as Inter-
national Journal of Production Economics (IJPE), International
infrastructure. The developed model can be adapted in
Journal of Production Research (IJPR), Annals of Operations
manufacturing, warehousing, and distribution industries Research (ANOR). He is the visiting faculty to reputable institu-
with minor adjustments and customised control levels, tions across the globe teaching in executive MBA and corporate
data collection mode, processing, and analytical meth- programmes.
ods. New designs may integrate AAM with other existing Yangyan (Peter) Shi is a faculty member
industrial models by customising it based on infrastruc- of Macquarie Business School, Macquarie
ture, cost, and energy availability. A better alternative for University in Australia, and a member of
R packages can also be tested in the design. The devel- the Center for Supply Chain Management
oped AAM with TTL and FAI can set a new benchmark at the University of Auckland Business
School in New Zealand. He focuses on
for research on emerging and innovative smart technolo- the research of operations, logistics and
gies in real-time. supply chain management. His research
papers have been published in international operations and
supply chain management journals. He acknowledged that the
Acknowledgements grant for ‘research on the supply of market-oriented elderly care
services with involvements of financial institutions’ was sup-
Yangyan Shi acknowledged that the grant for ‘research on the ported by funding from the Academy of the Social Sciences in
supply of market-oriented elderly care services with involve- Australia.
ments of financial institutions’ was supported by funding from
the Academy of the Social Sciences in Australia. V. Raja Sreedharan is an Assistant Pro-
fessor in the Supply Chain Management,
BEAR Lab (Business, Economie et Actu-
Disclosure statement ariat), Rabat Business School, Univer-
site Internationale de Rabat, Morocco. He
No potential conflict of interest was reported by the author(s). graduated from the College of Engineer-
ing Guindy with a PhD in Lean Six Sigma
and has published works in peer reviewed
journals. His current research interests focus on managing the
Data availability statement
VUCA in the business operations, and process improvement
The data that support this study is available from the cor- for services industries. He also consults in the field of process
responding author, Professor Lenny Koh (S.C.L.Koh@she optimisation with industries in private sectors.
ffield.ac.uk) on request. Lenny Koh is the Founder and Direc-
tor of the Advanced Resource Efficiency
Centre (AREC) and Head of Communi-
cation, Partnership and Internationalisa-
Notes on contributors tion of Energy Institute at The University
A. Manimuthu is currently working as a of Sheffield, UK. Her work contributes to
Research Fellow, Cybersecurity Research advancing the understanding and resolu-
Centre (CYSREN) Nanyang Technologi- tion of complex supply chain with inter-
cal University (NTU), Singapore. Before disciplinary approaches across supply chain and information
joining NTU, he was invited to work systems to achieve sustainability, digital transformation, circu-
as Visiting Research Fellow at Robotics lar economy, carbon neutrality and net zero transformation.
and Automation Research Lab (ROAR), Her research are funded through several sources including EU,
Singapore University of Technology and ERDF, EPSRC, ESRC, BBSRC, NERC, STFC, Leverhulme Trust,
Design (SUTD), Singapore. He teaches courses like Artificial Innovate UK, GCRF, DEFRA, industry, government and so on.
Intelligence, Operation management, Data Analytics, Business With H index 69 and > 354 publications; she is world leader
Research Methods, and Embedded Process Automation. He and pioneer of 4 digital Cloud based software tools supported
has been a keynote speaker for 15+ events in various reputed by Microsoft (SCEnAT suites including SCEnAT, SCEnAT+,
research institutions globally. His current research interest SCEnATi and SCEnAT 4.0). She is on the Editorial Board of
focuses more on AI, Autonomous Vehicles, Cybersecurity, various high quality journals including International Journal
Machine learning, and Big Data Analytics. of Production Research. Her work transforms supply chain
 INTERNATIONAL JOURNAL OF PRODUCTION RESEARCH 133

towards resource sustainability; and infuses and digitises sup- Gonçalves, J. N., P. Cortez, M. S. Carvalho, and N. M. Frazão.
ply chain life cycle thinking (including LCA, AI, ML, DL and 2021. “A Multivariate Approach for Multi-Step Demand
blockchain) across sectors, technologies and systems. Forecasting in Assembly Industries: Empirical Evidence
from an Automotive Supply Chain.” Decision Support Sys-
tems 142: 113452.
ORCID Guo, G., and S. M. Ryan. 2021. “Sequencing Mixed-Model
Assembly Lines with Risk-Averse Stochastic Mixed-Integer
Arunmozhi Manimuthu http://orcid.org/0000-0003-4909 Programming.” International Journal of Production Research
-4880 1–18. https://www.tandfonline.com/doi/abs/10.1080/0020
V. Raja Sreedharan http://orcid.org/0000-0003-3601-8002 7543.2021.1931978.
Gupta, R., S. Tanwar, N. Kumar, and S. Tyagi. 2020. “Block-
chain-based Security Attack Resilience Schemes for Autono-
References mous Vehicles in Industry 4.0: A Systematic Review.” Com-
Alavian, P., Y. Eun, S. M. Meerkov, and L. Zhang. 2020. “Smart puters & Electrical Engineering 86: 106717.
Production Systems: Automating Decision-Making in Man- Hadian, H., S. Chahardoli, A. M. Golmohammadi, and
ufacturing Environment.” International Journal of Produc- A. Mostafaeipour. 2020. “A Practical Framework for Supplier
tion Research 58 (3): 828–845. Selection Decisions with an Application to the Automotive
Allam, Z., and Z. A. Dhunny. 2019. “On big Data, Artificial Sector.” International Journal of Production Research 58 (10):
Intelligence and Smart Cities.” Cities 89: 80–91. 2997–3014.
Andoni, M., V. Robu, D. Flynn, S. Abram, D. Geach, D. Jenkins, Ivanov, D., A. Dolgui, and B. Sokolov. 2019. “The Impact of
P. McCallum, and A. Peacock. 2019. “Blockchain Technology Digital Technology and Industry 4.0 on the Ripple Effect
in the Energy Sector: A Systematic Review of Challenges and and Supply Chain Risk Analytics.” International Journal of
Opportunities.” Renewable and Sustainable Energy Reviews Production Research 57 (3): 829–846.
100: 143–174. Ivanov, D., A. Dolgui, B. Sokolov, F. Werner, and M. Ivanova.
Baryannis, G., S. Validi, S. Dani, and G. Antoniou. 2019. “Sup- 2016. “A Dynamic Model and an Algorithm for Short-Term
ply Chain Risk Management and Artificial Intelligence: State Supply Chain Scheduling in the Smart Factory Industry
of the art and Future Research Directions.” International 4.0.” International Journal of Production Research 54 (2):
Journal of Production Research 57 (7): 2179–2202. 386–402.
Bhamra, R., S. Dani, and K. Burnard. 2011. “Resilience: the Ivanov, D., B. Sokolov, and A. Dolgui. 2014. “The Ripple
Concept, a Literature Review and Future Directions.” Inter- Effect in Supply Chains: Trade-off ‘Efficiency-Flexibility-
national Journal of Production Research 49 (18): 5375–5393. Resilience’ in Disruption Management.” International Jour-
Cioffi, R., M. Travaglioni, G. Piscitelli, A. Petrillo, and F. De nal of Production Research 52 (7): 2154–2172.
Felice. 2020. “Artificial Intelligence and Machine Learning Jabbar, R., M. Kharbeche, K. Al-Khalifa, M. Krichen, and
Applications in Smart Production: Progress, Trends, and A. K. Barkaoui. 2020. “Blockchain for the Internet of Vehi-
Directions.” Sustainability 12 (2): 492. cles: A Decentralized IoT Solution for Vehicles Commu-
Copeland, R., M. Copeland, S. Ahvar, N. Crespi, O. Shagdar, nication Using Ethereum.” Sensors (Switzerland) 20 (14):
and R. Durand. 2019. “Automotobe Virtiual Edge Commu- 1–27.
nicator (AVEC) with Vehicular Inter-Agent Service Orches- Jain, S., G. Shao, and S. J. Shin. 2017. “Manufacturing Data Ana-
tration and Resourcing (ViSOR).” Annales Des Telecom- lytics Using a Virtual Factory Representation.” International
munications / Annals of Telecommunications 74 (9-10): Journal of Production Research 55 (18): 5450–5464.
655–662. Keivanpour, S., D. Ait-Kadi, and C. Mascle. 2017. “Automobile
De Giovanni, P. 2020. “Blockchain and Smart Contracts in Sup- Manufacturers’ Strategic Choice in Applying Green Prac-
ply Chain Management: A Game Theoretic Model.” Interna- tices: Joint Application of Evolutionary Game Theory and
tional Journal of Production Economics 228: 107855. Fuzzy Rule-Based Approach.” International Journal of Pro-
Dolgui, A., D. Ivanov, S. Potryasaev, B. Sokolov, M. Ivanova, duction Research 55 (5): 1312–1335.
and F. Werner. 2020. “Blockchain-oriented Dynamic Mod- Khan, P. W., and Y.-C. Byun. 2020. “Smart Contract Centric
elling of Smart Contract Design and Execution in the Supply Inference Engine for Intelligent Electric Vehicle Transporta-
Chain.” International Journal of Production Research 58 (7): tion System.” Sensors 20 (15): 4252.
2184–2199. Kim, T., I. Y. Jung, and Y. Hu. 2020. “Automatic, Location-
Elverum, C. W., and T. Welo. 2016. “Leveraging Prototypes Privacy Preserving Dashcam Video Sharing Using Block-
to Generate Value in the Concept-to-Production Process: a chain and Deep Learning.” Human-Centric Computing and
Qualitative Study of the Automotive Industry.” International Information Sciences 10 (1): 1–23.
Journal of Production Research 54 (10): 3006–3018. Koh, L., A. Dolgui, and S. Sarkis. 2020. “Blockchain in Trans-
Erfurth, T., and J. Bendul. 2018. “Integration of Global Manu- port and Logistics – Paradigms and Transitions.” Interna-
facturing Networks and Supply Chains: a Cross Case Com- tional Journal of Production Research 58 (7): 2045–2062.
parison of six Global Automotive Manufacturers.” Interna- Koh, L., G. Orzes, and F. Jia. 2019. “The Fourth Industrial Rev-
tional Journal of Production Research 56 (22): 7008–7030. olution (Industry 4.0): Technologies Disruption on Opera-
Fenwick, M., and E. P. M. Vermeulen. 2019. “Technology and tions and Supply Chain Management.” International Jour-
Corporate Governance: Blockchain, Crypto, and Artificial nal of Operations and Production Management 39 (6/7/8):
Intelligence.” Texas Journal Business Law 48: 1–15. 817–828.
Giancaspro, M. 2017. “Is a ‘Smart Contract’ Really a Smart Idea? Kolvart, M., M. Poola, and A. Rull. 2016. “Smart Contracts.” In
Insights from a Legal Perspective.” Computer Law & The Future of Law and Etechnologies, edited by Kerikmae T
Security Review 33 (6): 825–835. and Rull A, 133–147. Switzerland: Springer.
134 A. MANIMUTHU ET AL.

Kong, Q., R. Lu, F. Yin, and S. Cui. 2021. “Blockchain- O’Leary, D. E. 2013. “Artificial Intelligence and big Data.” IEEE
based Privacy-Preserving Driver Monitoring for MaaS in the Intelligent Systems 28 (2): 96–99.
Vehicular IoT.” IEEE Transactions on Vehicular Technology Omohundro, S. 2014. “Cryptocurrencies, Smart Contracts, and
70 (4): 3788–3799. Artificial Intelligence.” AI Matters 1 (2): 19–21.
Kumar, S., B. S. Purohit, V. Manjrekar, V. Singh, and Parunak, H. V. D. 1996. “Applications of Distributed Artificial
B. K. Lad. 2018. “Investigating the Value of Integrated Intelligence in Industry.” Foundations of Distributed Artifi-
Operations Planning: A Case-Based Approach from Auto- cial Intelligence 2: 1–18.
motive Industry.” International Journal of Production Research Pokhrel, S. R., and J. Choi. 2020. “Federated Learning with
56 (22): 6971–6992. Blockchain for Autonomous Vehicles: Analysis and Design
Lacerda, A. P., A. R. Xambre, and H. M. Alvelos. 2016. “Apply- Challenges.” IEEE Transactions on Communications 68 (8):
ing Value Stream Mapping to Eliminate Waste: a Case Study 4734–4746.
of an Original Equipment Manufacturer for the Automo- Pournader, M., Y. Shi, S. Seuring, and S. C. L. Koh. 2020.
tive Industry.” International Journal of Production Research “Blockchain Applications in Supply Chains, Transport and
54 (6): 1708–1720. Logistics: a Systematic Review of the Literature.” Inter-
Liao, Y., F. Deschamps, E. D. F. R. Loures, and L. F. P. Ramos. national Journal of Production Research 58 (7): 2063–
2017. “Past, Present and Future of Industry 4.0-a Systematic 2081.
Literature Review and Research Agenda Proposal.” Interna- Raj Kumar Reddy, K., A. Gunasekaran, P. Kalpana, V. Raja
tional Journal of Production Research 55 (12): 3609–3629. Sreedharan, and S. Arvind Kumar. 2021. “Developing a
Lu, Y., X. Huang, K. Zhang, S. Maharjan, and Y. Zhang. 2020. Blockchain Framework for the Automotive Supply Chain:
“Blockchain Empowered Asynchronous Federated Learn- A Systematic Review.” Computers and Industrial Engineering
ing for Secure Data Sharing in Internet of Vehicles.” IEEE 157, doi:10.1016/j.cie.2021.107334.
Transactions on Vehicular Technology 69 (4): 4298–4311. Rane, S. B., and Y. A. M. Narvel. 2021. “Re-designing the
Macrinici, D., C. Cartofeanu, and S. Gao. 2018. “Smart Con- Business Organization Using Disruptive Innovations Based
tract Applications Within Blockchain Technology: A Sys- on Blockchain-IoT Integrated Architecture for Improving
tematic Mapping Study.” Telematics and Informatics 35 (8): Agility in Future Industry 4.0.” Benchmarking: An Interna-
2337–2354. tional Journal 28 (5): 1883–1908.
Magazzeni, D., P. McBurney, and W. Nash. 2017. “Validation Saberi, S., M. Kouhizadeh, J. Sarkis, and L. Shen. 2019.
and Verification of Smart Contracts: A Research Agenda.” “Blockchain Technology and its Relationships to Sustain-
Computer 50 (9): 50–57. able Supply Chain Management.” International Journal of
Manimuthu, A., and V. Dharshini. 2021. “Framework for Load Production Research 57 (7): 2117–2135.
Power Consumption in HAN Using Machine Learning and Samarakoon, S., M. Bennis, W. Saad, and M. Debbah. 2020.
IoT Assistance.” IEEE Design & Test 38: 102–108. “Distributed Federated Learning for Ultra-Reliable low-
Manimuthu, A., A. V. Le, R. E. Mohan, P. Veerajagadeshwar, Latency Vehicular Communications.” IEEE Transactions on
N. Huu Khanh Nhan, and K. Ping Cheng. 2019. “Energy Communications 68 (2): 1146–1159.
Consumption Estimation Model for Complete Coverage of Sayeed, S., H. Marco-Gisbert, and T. Caira. 2020. “Smart Con-
a Tetromino Inspired Reconfigurable Surface Tiling Robot.” tract: Attacks and Protections.” IEEE Access 8: 24416–24427.
Energies 12 (12): 2257. Shahbazi, Z., and Y. C. Byun. 2021. “Smart Manufacturing Real-
Manimuthu, A., and R. Ramadoss. 2019. “Absolute Energy Time Analysis Based on Blockchain and Machine Learning
Routing and Real-Time Power Monitoring for Grid-Connec- Approaches.” Applied Sciences 11 (8): 3535.
ted Distribution Networks.” IEEE Design & Test 36 (2): Sharma, P. K., N. Kumar, and J. H. Park. 2019. “Blockchain-
88–96. based Distributed Framework for Automotive Industry in a
Manimuthu, A., V. G. Venkatesh, V. Raja Sreedharan, and V. Smart City.” IEEE Transactions on Industrial Informatics 15
Mani. 2021. “Modelling and Analysis of Artificial Intelli- (7): 4197–4205.
gence for Commercial Vehicle Assembly Process in VUCA Singh, S. K., S. Rathore, and J. H. Park. 2020. “Blockiotintel-
World: a Case Study.” International Journal of Produc- ligence: A Blockchain-Enabled Intelligent IoT Architecture
tion Research 1–19. https://www.tandfonline.com/doi/abs/ with Artificial Intelligence.” Future Generation Computer
10.1080/00207543.2021.1910361. Systems 110: 721–743.
Min, H. 2010. “Artificial Intelligence in Supply Chain Man- Theorin, A., K. Bengtsson, J. Provost, M. Lieder, C. Johnsson,
agement: Theory and Applications.” International Journal of T. Lundholm, and B. Lennartson. 2017. “An Event-Driven
Logistics: Research and Applications 13 (1): 13–39. Manufacturing Information System Architecture for Indus-
Mishra, B., B. Mahanty, and J. J. Thakkar. 2021. “A Quan- try 4.0.” International Journal of Production Research 55 (5):
tifiable Quality Enabled Servitisation Model: Benchmarking 1297–1311.
Indian Automobile Manufacturers.” International Journal of Treleaven, P., R. G. Brown, and D. Yang. 2017. “Blockchain
Production Research 59 (9): 2667–2689. Technology in Finance.” Computer 50 (9): 14–17.
Mohanta, B. K., D. Jena, U. Satapathy, and S. Patnaik. 2020. Venkatesh, V. G., K. Kang, B. Wang, R. Y. Zhong, and A. Zhang.
“Survey on IoT Security: Challenges and Solution Using 2020. System architecture for blockchain based transparency of
Machine Learning, Artificial Intelligence and Blockchain supply chain social sustainability. Vol. 63.
Technology.” Internet of Things, 100227. https://doi.org/ Wang, S., L. Ouyang, Y. Yuan, X. Ni, X. Han, and F. Y.
10.1016/j.iot.2020.100227. Wang. 2019. “Blockchain-enabled Smart Contracts: Archi-
Nofer, M., P. Gomber, O. Hinz, and D. Schiereck. 2017. tecture, Applications, and Future Trends.” IEEE Transactions
“Blockchain.” Business & Information Systems Engineering on Systems, Man, and Cybernetics: Systems 49 (11): 2266–
59 (3): 183–187. 2277.
 INTERNATIONAL JOURNAL OF PRODUCTION RESEARCH 135

Wang, S., Y. Yuan, X. Wang, J. Li, R. Qin, and F.-Y. Wang. Data-Driven Approach.” International Journal of Production
2018. “An Overview of Smart Contract: Architecture, Appli- Research 58 (17): 5359–5379.
cations, and Future Trends.” 2018 IEEE Intelligent Vehicles Xu, L. D., E. L. Xu, and L. Li. 2018. “Industry 4.0: State of the
Symposium (IV), 108–113. art and Future Trends.” International Journal of Production
Wei, Z., Y. Feng, Z. Hong, R. Qu, and J. Tan. 2017. “Prod- Research 56 (8): 2941–2962.
uct Quality Improvement Method in Manufacturing Process Zhang, Yuanyu, S. Kasahara, Y. Shen, X. Jiang, and J. Wan. 2019.
Based on Kernel Optimisation Algorithm.” International “Smart Contract-Based Access Control for the Internet of
Journal of Production Research 55 (19): 5597–5608. Things.” IEEE Internet of Things Journal 6 (2): 1594–1605.
Xia, S., F. Lin, Z. Chen, C. Tang, Y. Ma, and X. Yu. 2020. “A Zhang, Yu, and J. Wen. 2017. “The IoT Electric Business Model:
Bayesian Game Based Vehicle-to-Vehicle Electricity Trad- Using Blockchain Technology for the Internet of Things.”
ing Scheme for Blockchain-Enabled Internet of Vehicles.” Peer-to-Peer Networking and Applications 10 (4): 983–994.
IEEE Transactions on Vehicular Technology 69 (7): 6856– Zheng, Z., S. Xie, H.-N. Dai, W. Chen, X. Chen, J. Weng, and
6868. M. Imran. 2020. “An Overview on Smart Contracts: Chal-
Xu, Z., and Y. Dang. 2020. “Automated Digital Cause-and-Effect lenges, Advances and Platforms.” Future Generation Com-
Diagrams to Assist Causal Analysis in Problem-Solving: a puter Systems 105: 475–491.