A : Methodolgie B : Problematique

The automotive industry is a vital Logistics risk assessment in the

sector for the Moroccan supply chain is considered as one
economy, and in the year 2020, of the important topics that has
the added-value of the attracted the attention of
automotive industry in Morocco researchers in recent years;
is higher than that of the Companies that struggle to
fertilizer. This sector is manage their logistical risks by
considered the first exporter of not putting in place resilient
the country. Our study focuses on strategies to mitigate them, may
the assessment of the pure suffer from significant financial
logistical risks in the Moroccan losses.
automotive industry.

1 : A decision support system for evaluating the logistical risks in Supply chains based on RPN factors and multi criteria de
Motivated by the security For several years, the automotive
requirements inherent in wireless industry and government
vehicular networks, this study agencies have sought ways to
proposes a secure and enhance road safety and traffic
lightweight authentication management. The evolution of
scheme for V2X communication. wireless communication systems
The methodology involves the has paved the way for the
use of Attribute-based Signature realization of vehicle-to-vehicle
(ABS) to achieve message (V2V) and vehicle-to-
authentication, privacy, and infrastructure (V2I)
integrity. The proposed solution communication. Referred to as
supports vehicles in securely intelligent transportation systems
receiving a secret key and (ITS), these wireless automotive
message from a roadside unit communication systems,
(RSU) or other vehicles. To specifically Vehicle-to-Everything
validate the security of the (V2X) communication, are critical
proposed solution, a formal for improving road traffic and
verification is conducted using ensuring the safety and privacy of
the Automated Validation of users. However, the challenge
Internet Security Protocols and lies in maintaining security
Applications (AVISPA) tool. requirements to uphold system
functionality and safeguard users'
safety and privacy.

2 : A Secure Vehicle to Everything (V2X) Communication Model for Intelligent Transportation System Commentaires impo
 This paper presents a multi-level One of the latest trends in EU
decision-making framework manufacturing research is to
targeting dynamic work balancing increase reconfigurability of
among human operators and production systems by
robot resources. The proposed introducing mobile dual-arm
mechanism exploits Artificial workers able to collaborate with
Intelligence techniques for humans in dynamic changing
generating efficient task environments. The flexibility of
sequences while ensuring the system is increased by the
optimal and collision-free path ability of the mobile robots to
generation for the assigned easily re-locate themselves inside
resources. The focus is given on the shopfloor for performing
the integration of multi-criteria various assembly operations.
decision-making mechanisms
with mobile robots’ low-level
planning modules for deriving
effective reconfiguration plans
considering the shopfloor's real-
time status.
3 : AI based combined scheduling and motion planning in flexible robotic assembly lines Commentaires importés
To this end, the purpose of this The COVID-19 crisis has severely
paper is to propose a leagility disrupted Moroccan automotive
evaluation framework using fuzzy production. This pandemic has
quality function deployment weakened the automotive supply
approach. chain; it faced a fall in demand
and reduction in sales.
Consequently, the automotive
industry developed their
production capabilities through
constant innovation in resource
reduction (leanness) while
responding rapidly to demand
changes (agility). A combination
of lean-agile supply chain leads to
obtain competitiveness in a time
and cost effective manner.
Successful implementation of
leagile supply chain requires
evaluation of criteria and
attributes.

4 : An integrated fuzzy QFD approach to leagile supply chain assessment during the COVID-19 crisis Commentaires import
 The proposed approach consists This paper deals with a topical
in the realization of an algorithm issue for researchers, namely the
to detect an object in the shape detection of objects in real-time.
of a triangle, square, and This is an important topic in the
rectangle, applied on traffic road field of computer vision, as the
signs for the automotive sector. detection and recognition of
Many detection methods can be objects from images require very
adapted to progress efficiently powerful algorithms. There are
and robustly, combining these different ways to detect objects,
different studies to find the best for example, based on either the
strategy. Besides, the use of color or shape of the objects.
numerous images in the test
phase can be used to increase
the accuracy of object detection.

5 : Applying Real-Time Object Shapes Detection to Automotive Traffic Roads Signs Commentaires importés
This paper deals with a topical The vision of Industry 4.0 impacts
issue for researchers, namely the the mechanism as well as its
detection of objects in real-time. benefits to the supply chain
This is an important topic in the system. The biggest challenge to
field of computer vision, as the achieve a robust, stable and
detection and recognition of efficient eco-system for Supply
objects from images require very chain system is to have an
powerful algorithms. There are architecture that initiates the
different ways to detect objects, framework for both the eco-
for example, based on either the system and predictive analysis.
color or shape of the objects.

6 : Architectural Framework for Industry 4.0 Compliance Supply Chain System for Automotive Industry Commentaires imp
 This paper explores different The first part of this paper
areas of the automotive industry presents a literature review-
like the vehicle itself, the based study of the most
designing and manufacturing emerging technology of the
sectors, the after-sales services. It century, artificial intelligence (AI)
provides solutions to make each and its various applications in the
of them ‘Intelligent’. automobile sector. In this era of
the fourth Industrial Revolution
(Industry 4.0), industries have
become all the more
sophisticated owing to the
intelligent technologies
employed to maximize
production, quality, and profits;
and minimize wastage, time and
cost of production. This paper
studies various aspects of AI and
related tools and techniques and
aims to employ them in the
automobile industry to make
modern vehicles smart, safe and
reliable. Simultaneously, it strives
to automate the drives, thereby
reducing manual labor, increasing
efficiency and relieving humans
7 : Artificial intelligence and advanced materials in automotive industry~ Potential applications and perspectives Commen
of performing mundane,
 This survey systematically automotive industry is
reviews existing research at the undergoing a profound digital
intersection of AI/ML and V2X transformation to create
communications, focusing on autonomous vehicles. Vehicle-to-
handover management, Everything (V2X) communications
proactive caching, physical and enable the provisioning of
computation resources transportation use cases for road
allocation, beam selection traffic and safety management.
optimization, packet routing, and At the same time, during the past
QoS prediction in vehicular decade, Artificial Intelligence (AI)
environments. We extract the and Machine Learning (ML) have
underlying AI/ML techniques, the been in the spotlight because of
training features, their their outstanding performance in
architecture and discuss several various domains, including
aspects regarding the intricacies natural language processing, and
of vehicular environments and computer vision. Considering also
ML. These aspects include time current standardization efforts,
complexity of the algorithms, towards incorporating AI and ML
quality of real-world vehicle as integral sub-systems of
traces, suitability of AI/ML beyond 5G and 6G networks,
techniques in relevance to the these technologies are
designated network operation considered very promising to
and the underlying automotive optimize user, control, and
use case, as well as velocity and management network functions,
positioning accuracy but also to support road safety
8 : Artificial Intelligence and Machine Learning as key enablers for V2X communications~ A comprehensive survey Comme
requirements towards the and even entertainment
This paper presents an Industry To implement an Industry 4.0
4.0 framework, named BiDrac, Framework in an ongoing
that integrates computing, industrial manufacturing process
communication and control to prepare it for complex
under an Industrial Cyber- Artificial Intelligence use cases is
Physical​ System (ICPS) ecosystem. not an easy task. The automotive
It combines Artificial Intelligence industry is undergoing a large
and Industrial Internet of Things transformation due to a variety
(IIoT) inside the Industry 4.0 of disruptive factors through the
paradigm for the Predictive introduction of CASE (Connected,
Maintenance of an Automotive Autonomous, Shared, Electrified)
Paint Shop Process inside an technologies. Production
Industrial Cloud (IC) as an open improvements will enable future
platform that provides a development of CASE
community of cloud-based capabilities.
solutions to enable the
development of new digital
solutions.

9 : BiDrac Industry 4.0 framework~ Application to an Automotive Paint Shop Process Commentaires importés
 Hence, this study investigates the In traditional industries, such as
extent to which incumbents the automotive industry,
innovate through the design of AI- incumbents must draw on big
embedded systems-either via data and artificial intelligence (AI)
explorative or exploitative technologies by designing AI-
strategies-by focusing on the case embedded systems integrated
of the automotive industry. It into their end products. While
employed a sequential such systems are predominantly
explanatory mixed-method presented as paving the way for
design and a knowledge search new knowledge explorative
theoretical framework. approaches, traditional industry
incumbents may face challenges
integrating such disruptive
technology in their optimized
new product development
processes.

10 : Big data as an exploration trigger or problem-solving patch~ Design and integration of AI-embedded systems in the au
The objective of this chapter is to The rising demand for digitization
present the opportunities and and the increased focus on
challenges of Industry 4.0 in the efficiency and cost-efficient
automotive supply chain. We also productivity are boosting the
propose a blockchain-based automotive industry as well as
automotive supply chain model other industries’ growth. Recent
and study how it can fulfill our technological advances such as
needs in terms of transparency, industrial automation, IoT, cyber-
security, and traceability of physical systems, artificial
operations with the challenges intelligence, and blockchain lay
that come with it. the foundation for significant
advancements in the automotive
industry. The increased demand
for customized products in the
automotive market as well as
transparency regarding the
product’s life cycle challenges the
current automotive supply chain
model.

11 : Blockchain-Based CPS and IoT in the Automotive Supply Chain Commentaires importés
 This article discusses the The automotive industry is
modelling of an industrial process constantly seeking performance
“cataphoresis” of a large to remain competitive in a
automobile manufacturer via constantly changing market.
neural networks in order to
control and optimize it.

12 : Control of the “Cataphoresis” Process by Artificial Intelligence Commentaires importés

Within this context; as an The development of artificial
automotive company; we have intelligence (AI) in the
provided efficient and secure AI automotive sector is a major
models that can be used in the contributor to the increasing
autonomous driving industry, and number of intelligent
we have regularly investigated transportation systems.
and provided the necessary Especially, with the rapid
security attributes to ensure that emergence and evolution of
the AI models used in the machine learning and artificial
autonomous driving industry are intelligence (AI) capabilities the
equipped with the necessary number of AI models used in the
security attributes. For this automotive industry has been
purpose, we have performed a D- developed.
patching attack on the Yolo v3
model (Mujahid, 2021) to
compromise its integrity.

13 : Cybersecurity Assessment Construction of Artificial Intelligence Commentaires importés

This paper presents a theoretical Deep neural networks
framework based on deep neural revolutionized the field of
networks to predict the outcome artificial intelligence by
of quality assurance tests using successfully advancing the
vehicle configuration and process progress in complex problems.
data. One such complex challenge in
the automotive industry is the
prediction of quality assurance
tests in the vehicle assembly
process.
14 : Deep Learning based Predictive Testing Strategy in the Automotive Industry Commentaires importés
 This paper proposes a novel With smart sensors and
design framework that uses embedded drivers, today's
Federated learning-Artificial automotive industry has taken a
intelligence (FAI) for decision- giant leap in emerging
making and Smart Contract (SC) technologies like Machine
policies for process execution and learning, Artificial intelligence,
control in a completely and the Internet of things and
automated smart automobile started to build data-driven
manufacturing industry. The decision-making strategies to
proposed design introduces a compete in global smart
novel element called Trust manufacturing.
Threshold Limit (TTL) that helps
moderate the excess usage of
embedded equipment, tools,
energy, and cost functions,
limiting wastages in the
manufacturing processes.

15 : Design and development of automobile assembly model using federated artificial intelligence with smart contract Co
 In this context, in collaboration Object recognition is among the
with an Automotive components most important subjects in
manufacturer and FST faculty of computer vision, it has
sciences and technologies of undergone a huge evolution
Tangier (UAE University) they during these last decades, but in
have taken the initiative to the last years artificial
develop an object recognition intelligence has seen the
and self-correction system for a appearance of Deep Learning,
winding machine that requires a and through the efforts of
good accuracy of location of the researchers, Deep Learning is
needle using Deep Learning having great success, its
which is the purpose of this applications have touched on
paper. This report summarizes different fields, such as robotics,
the work done within this industry, automotive.
Company concerning the
development and
implementation of a system for
automating and self-correcting
the location of the needles of
winding machines using artificial
vision with Deep learning.

16 : Detection and Control System for Automotive Products Applications by Artificial Vision Using Deep Learning Commen
 This study is therefore designed The latest boom of artificial
to find out which general use intelligence (AI) has left the
cases exist for AI within the information management
context of car manufacturing and community in strong need of
which ones might be the most structure-providing, high-level
promising ones to pursue at this overview works. Such works are
early stage. We conducted a supposed to allow both
Delphi study with 39 experts in researchers and practitioners to
25 different globally scattered keep track of that steep
organizations over one and a half development across the
years. technology's numerous possible
application domains. So it is
among other things that AI is said
to incorporate enormous
potential for reducing the
operational costs of car
manufacturers all over the globe.
Nevertheless, many of them are
still struggling with adopting it at
a large scale just because of a
lack of knowledge on if and
where to apply it.

17 : How can artificial intelligence enhance car manufacturing~ A Delphi study-based identification and assessment of gen
Intelligent methods can support The automated design of
engineers in repetitive tasks and production systems is a young
give them more opportunity to field of research which has not
focus on work which requires been widely explored by industry
their core competencies. This nor research in recent decades.
paper presents a novel artificial Currently, the effort spent in
intelligence methodology that production system design is
automatically generates initial increasing significantly in the
production system configurations automotive industry due to the
based on real industrial scenarios number of product variants and
in the automotive field of body-in- product complexity.
white production. The hybrid
methodology reacts flexibly
against data sets of different
content and has been
implemented in a software
prototype.

18 : Hybrid Artificial Intelligence System for the Design of Highly-Automated Production Systems Commentaires importés
 We apply several techniques; the In this paper, we aim to improve
first is pre-processing, which the performance of Deep
includes the conversion of color Learning (DL) by creating a robust
space RGB, then equalization and and efficient Convolutional
normalization histogram of the Neural Network (CNN) model.
image dataset according to This CNN model will be subjected
Computer Vision (CV) tools. The to detecting and recognizing
second is devoted to Artificial traffic signs in real-time.
Intelligence (AI), which requires
the right choice of a neural layer
such as the convolution layer or
dropout layer, with powerful
optimizers such as Adam and
activation functions such as ReLU
and SoftMax. Also, we use the
technique of augmentation
dataset which characterizes by
the function of batch size for
each epoch.

19 : Improved Deep Learning Performance for Real-Time Traffic Sign Detection and Recognition Applicable to Intelligent T
Next, an evaluation index system The reasonable location of a
is constructed based on five charging station is crucial for
location influencing factors: land promoting the rapid
cost, construction costs, road development of the new energy
traffic flow, power grid automobile industry. This paper
conditions, and the surrounding initially establishes the location
environment. Numerical studies model with the goal of
show that both grey incidence minimizing the total social cost,
decision and grey target theory employing a genetic algorithm for
have the advantages of ease of the solution.
operation, low requirements for
data collection and processing
when selecting the optimal
location.

20 : Location of electric vehicle charging stations~ A perspective using the grey decision-making model Commentaires imp
 This paper aims to apply different Key performance indicators are
machine learning methods, tools for management, decision
namely support vector support, and forecasting; they
regression, optimized support reflect the strategy and vision of
vector regression (using a genetic the company in terms of
algorithm), random forest, objectives and allow always
extreme gradient boosting, and staying in step with the
deep learning to predict the expectations of the stakeholders.
overall equipment effectiveness
as a case study. We will make use
of several configurations of the
listed models in order to provide
a wide field of comparison. The
data used to train our models
were provided by an automotive
cable production industry.

21 : Machine learning for KPIs prediction~ a case study of the overall equipment effectiveness within the automotive indu
This paper presents a research In the context of
methodology using a reference internationalization, the supply
model called GRAILOG to capture chain (SC) becomes complex with
decisions and related KPI’s to a profusion of decisions to take.
control the SC. Then, the SC modeling and performances
methodology PPTechIP is have been widely discussed by
described and applied to researchers. The emergence of
accompany the company on the new technologies impacts the
relevant digital transformation to running and key performance
improve performances. PPTechIP indicators (KPI) of SC. There are
is based on a set of radars several models of SC but none of
divided into different decision them are oriented towards SC
levels and SC functions based on operations management, given
the GRAILOG model. The importance of Industry 4.0
calculated potential of progress technologies.
helps in decision-making.

22 : Methodology combining industry 4.0 technologies and KPI’s reliability for supply chain performance Commentaires im
 To achieve this, a design of In this paper, the authors study
experiment has been performed. the applicability of artificial
The main functions of the neural networks (ANN) for the
proposed model on this article modeling of a particular welding
are: simulate the process for process widely used in the
operators training purposes, automotive industry: pulsed gas
improve the performance of the metal welding process (GMAW-
welding process by identifying P). To apply this artificial
regions which are insensible to intelligence technology, it is
variations on input parameters, necessary to introduce input and
and finally increase flexibility of a output data (stimulus and
robotic welding cell. answers) to the network.

23 : Neural Modeling for Automotive Welding Applications Commentaires importés

LCA is a valuable tool for such Within the automotive industry,
assessments; however, this there are efforts to replace glass
method is time-consuming, fiber composites with greener yet
involving numerous details and lightweight natural fibers to
potential complexities. potentially reduce environmental
Recognizing the challenges, the impacts. Determining if these
study turns to artificial replacements are
intelligence as a promising environmentally friendlier and
discipline capable of easily the extent of their emissions
predicting intricate inquiries. The reduction throughout the
researchers employ machine vehicle's life cycle necessitates
learning to compare and predict the gold standard of a life cycle
greenhouse gas emissions assessment (LCA) based on
associated with the substitution greenhouse gas emissions.
of materials in automotive parts.
Notably, the study differs in its
approach by processing very
limited input data, a departure
from the typical extensive
machine learning datasets.
Despite the common deterrent
posed by limited data in solving
such problems, it allows the
testing of various artificial
intelligence algorithms and input
24 : Prediction of greenhouse gas emissions reductions via machine learning algorithms~ Toward an artificial intelligence-
matrices to swiftly predict
 One crucial approach to achieve Recent developments in
this is predictive maintenance maintenance modeling, fueled by
(PdM). Since modern vehicles data-based approaches such as
come with an enormous amount machine learning (ML), have
of operating data, ML is an ideal enabled a broad range of
candidate for PdM. While PdM applications. In the automotive
and ML for automotive systems industry, ensuring the functional
have both been covered in safety over the product life cycle
numerous review papers, there is while limiting maintenance costs
no current survey on ML-based has become a major challenge.
PdM for automotive systems.
Consequently, the authors
surveyed and categorized papers,
analyzed them from an
application and ML perspective,
identified open challenges, and
discussed possible research
directions.

25 : Predictive maintenance enabled by machine learning~ Use cases and challenges in the automotive industry Commen
To support planning and decision- In many industries, there is a high
making in manufacturing probability that production lines
systems, new analytical might produce extreme
approaches must emerge to quantities, far away from what
enable the transformation and would be expected and/or
representation of the available desired considering the
data. The paper presents and production and market
discusses the use of prediction conditions. This impacts
models, specifically linear considerably on product costs
regression based on supervised and respective margins.
machine learning algorithms.
Multiple linear regression models
can be trained for different time
periods, allowing better control
of costs and serving as decision-
making tools for subsequent
periods.

26 : Prescriptive Cost Analysis in Manufacturing Systems Commentaires importés

 The paper aims to study the The Industry 4.0 technologies are
application of Industry 4.0 transforming the current industry
technologies in managing the into a smart industry, and the
production flow in the application of lean manufacturing
automotive industry. The tools has concurrently reduced
technologies considered include waste and improved efficiency.
Internet of Things (IoT), artificial However, it is unclear which of
intelligence (AI), cloud the Industry 4.0 technologies is
computing, machine learning, best suited for managing
security, Big data, Blockchain, production flows.
Deep learning, Digitization, and
Cyber-physical system (CPS). The
study evaluates how these
technologies contribute to
production efficiency and
reduced lead times.

27 : Production Flow Management Based on Industry 4.0 Technologies Commentaires importés

The paper introduces a novel In the manufacturing sector, the
framework for real-time ability to predict key
monitoring and prediction of performance indicators (KPIs),
manufacturing system KPIs, specifically Overall Equipment
focusing on production and Efficiency (OEE), is crucial for
maintenance metrics. The process optimization and
approach employs One- informed decision-making.
Dimensional Convolutional Despite various machine learning
Neural Networks, providing methods applied for KPI
accurate time series forecasting forecasting, real-time predictive
for KPIs with self-adaptive capabilities remain underutilized.
models that evolve in response to
data changes. The framework is
implemented and tested in an
advanced automotive industry
setting.
28 : Real-Time KPI Forecasting with 1D Convolutional Time Series for Enhanced Manufacturing Efficiency Commentaires im
These cars, which can be driven For the past twenty years, the
without the intervention of a automotive industry and research
driver, use several sensors and organizations have been aiming
artificial intelligence technologies to put fully autonomous cars on
simultaneously, allowing them to the road.
detect the environment, merge
the obtained information,
analyze it, decide on an action,
and implement it.

29 : SECURITY STUDY of ROUTING ATTACKS in VEHICULAR AD-HOC NETWORKS (AUTONOMOUS CAR) Commentaires impo
 Assurance frameworks for vehicle Automated Driving Systems
systems have traditionally been (ADSs) in road vehicles, which can
underpinned by standards. Some undertake the dynamic driving
commonly adopted standards will task without requiring a human
provide barriers to the market driver, are on the verge of large-
deployment of ADSs, particularly scale deployment.
the inherent capability of ADSs to
have their driving functionality
changed using Machine Learning
(ML) during in-service operation.

30 : Standards relevant to automated driving system safety~ A systematic assessment Commentaires importés
Introducing and integrating an AI- The success of data-driven
enhanced methodology in maintenance is strongly
reliability-centred maintenance dependant on effective use of AI
study of complex production and multi-structured data
systems leads to reducing failure sources.
rates and optimizing availability.
In manufacturing enterprises,
information about machine
failures and expert knowledge
are often stored in digital shift
books (DSB).

31 : Text mining for AI enhanced failure detection and availability optimization in production systems Commentaires impo
 This review outlines the recent The electrification of mass
advances in EVs and related transportation is hailed as a
infrastructure, mainly from solution for reducing global
artificial intelligence (AI), which greenhouse-gas emissions and
makes EVs a more attractive dependence on unsustainable
consumer option. The application energy sources. The annual sales
of AI in improving EVs, facilitating of electric vehicles (EVs) has
EV charging stations, and EV continued to rise since 2011, with
integration with the smart grid is a global sale of EVs of 2.1 million
critically analyzed and reviewed. in 2019. This increase in sales is
mainly due to continued
improvement in the cost and
performance of commercial EVs,
increased EV options available to
consumers, and environmental
awareness. However, despite the
positive outlook, EVs still face
major challenges that hinder
their rapid and widespread
adoption: limited driving range,
long charging times, and a lack of
sufficient charging infrastructure.

32 : The role of artificial intelligence in the mass adoption of electric vehicles Commentaires importés
 In this paper, a framework is Data is everything - at least this is
presented for processing sensor one of the main messages of the
data of machining processes with ongoing industrial revolution.
variable cycle times in an Manufacturing companies all
unstable environment. over the world are expanding
Traditional and novel AI their digital infrastructure and
algorithms are tested on the data knowledge on data analysis in the
of a vulcanization process from hope of increasing their KPIs with
the automotive industry, namely the help of artificial intelligence
from tire manufacturing’s curing (AI). Although several well-
phase. The process in question designed data-driven solutions
consists of several subprocesses, are available, the most crucial
and the quality of curing is mostly part, data preparation is still not
dependent of the status of a fully supported.
specific type of machine tool.
Conventional methods (e.g.,
examining the cured product
manually) are currently used for
failure recognition, however the
examination is only feasible after
a long delay due to the extreme
level of heat, which leads to
unnecessary and unwanted scrap
production.
33 : Tool failure recognition using inconsistent data Commentaires importés
 This paper presents aspects of Vehicle software architectures
ongoing research at the BMW have been evolving over the last
Research Department regarding a twenty years to support data-
conceptual design for vehicle driven functionalities. Several
software architectures in the enterprises from different
automotive industry. We discuss domains are currently focusing
the principles of a modern data on improving their data
architecture with particular architectures by re-defining the
emphasis on the data-centric underlying data models to enable
mindset. We also explore the core support for analytics and
current challenges and possible artificial intelligence. Moreover, a
working points as the foundation common desire to add clear data
to move from siloed data provenance and explicit context
towards a so-called AI factory. impulses the field of semantics
and knowledge graphs.
Nevertheless, in the automotive
industry, the scenario of
connected vehicles implies extra
complexity. Vehicle data has an
enormous variety, making it
essential to develop and adopt
standards.

34 : Towards a Data-Centric Architecture in the Automotive Industry Commentaires importés

The conceptual models used have This article presents a multi-
been formulated based on a method simulation-based tool
review of the state of the art that reflects in detail the network
research, which helped select the structure of the automotive
adequate modeling for supply chain and will be used to
simulation paradigms, mainly support decision-making. It
ABM (Agent-based modeling), DE allows examining the complex
(discrete event), and SD (System system behavior considering
dynamics) modeling, and mutual interactions among nodes
integrating PLM (Product Life and evaluating the upstream and
Cycle Management) approach. downstream connectivity
The first part of the paper mechanisms in product and
identifies the needs for supply chain design.
connectivity mechanisms in auto
SCM and modeling of this
complex system. In sequence, the
simulation modeling as a
business decision tool is
discussed, and the method
conception is presented.
35 : Towards a mixed method model and simulation of the automotive supply chain network connectivity Commentaires
 Therefore, there is an urgent Intelligent Connected Vehicle
need for lightweight and real- (ICV), as the product of the deep
time solutions for intra-vehicle integration of the automobile,
communication security. In electronics, information, and
addition, it should be noted that artificial intelligence in the era of
no matter where an attack is 5G advanced and beyond, has
launched, the attacker must become the pursuit of the
control the corresponding automotive industry. To improve
Electronic Control Unit (ECU) ICV security, traditional methods,
through intra-vehicle such as in-vehicle gateway
communication if he wants to firewall, encryption, and
control the ICV's behavior. authentication, become the
Therefore, intra-vehicle potential choice to protect the
communications can be regarded communication between vehicles
as the last line of defense for ICV and the outside world. However,
security. With this in mind, we traditional methods cannot be
conduct a comprehensive and directly applied to ICV due to the
new technical investigation about extremely limited computing and
the security issues of intra- storage resources of intra-vehicle
vehicle communications, focusing communication and the high real-
on the vulnerabilities, attacks, time requirements of the packets
and countermeasures. transmitted in the vehicle.
Additionally, we investigate the
activities and contributions of
standardization organizations
36 : Towards secure intra-vehicle communications in 5G advanced and beyond~ Vulnerabilities, attacks and countermeas
and highlight the new challenges
As a step towards this direction, Automobile manufacturers
in this paper, we propose a around the world are increasingly
dynamic trust management deploying Industrial Internet of
model suitable for industrial Things (IIoT) devices in their
environments. We also propose factories to accompany the
to change the traditional Industrial Revolution 4.0. Security
centralized architecture of IIoT and privacy are the main
networks in automotive plants limitations to the integration of
into a hybrid architecture based the Internet of Things (IoT) into
on a set of new industrial industrial processes. Therefore, it
relationship rules. is necessary to protect industrial
data contained in IIoT devices
and keep them confidential.

37 : Trust Management in Industrial Internet of Things Commentaires importés

 In the present research, we The overall equipment
implemented different Machine effectiveness (OEE) is a
Learning algorithms, namely; performance measurement
Support Vector Machine, metric widely used. Its calculation
Optimized Support Vector provides managers the possibility
Machine (using Genetic to identify the main losses that
Algorithm), Random Forest, reduce machine effectiveness
XGBoost, and Deep Learning to and then take the necessary
predict the estimated OEE value. decisions to improve the
The data used to train our situation. However, this
models were provided by an calculation is done a-posterior,
automotive cable production which is often too late.
industry.

38 : Using Machine Learning for Predicting Efficiency in Manufacturing Industry Commentaires importés
The article shows an approach to In high-tech production,
virtual in-line inspection, companies often deal with the
predicting the product's manufacture of assemblies with
functionality. An artificial neural quality requirements close to the
network (ANN) fed with product technological limits of
characteristics and process data, manufacturing processes.
as well as the resulting functional
fulfillment of the product, is
trained for virtual function
prognosis.

39 : Virtual In-line Inspection for Function Verification in Serial Production by means of Artificial Intelligence Commentaire
 C : Resultat

Our main objective for this study

is to assess the logistical risks
which will allow us to put in place
proactive and predictive resilient
strategies for their mitigation.

The security analysis

demonstrates that the proposed
authentication scheme
successfully meets the security
goals of V2X communication. The
formal verification using the
AVISPA tool provides assurance
in terms of security.
Furthermore, the performance
evaluation reveals that the
proposed protocol incurs a lower
computational cost compared to
previous protocols. This signifies
an improvement in
authentication and
communication efficiency,
making the proposed scheme
more suitable for practical
applications in wireless vehicular
networks.
The developments have been
tested in a case study from the
automotive industry.

As a result, ‘order guidance’ has

been taken as the most
important capability of
automotive supply chain. ‘E-
fulfilment logistic’ has been
considered as the most
important enabler to gain supply
chain leagility.
This approach could be
interesting, can be established in
the industrial sector, for the
conception or development of
embedded systems such as
Advanced Driver Assistance
Systems (ADAS) integrated into
autonomous vehicles. The
objective of this approach is to
help systems based on artificial
intelligence for road safety, to
have a high degree of accuracy
with negligible errors for
detecting traffic road signs.

It was found to be acceptable

and useful for adopting the
architecture for a synchronized
supply chain system to Industry
4.0 as well as provisions many
disruptive innovations, which is
quite useful for both social and
economic viewpoints.
The second part briefs us about
advanced materials like polymer
composites, carbon fiber, and
high-strength steel. It discusses
the various factors that make
these intelligent materials
feasible in the automobile
industry and discusses the
methods and areas of
applications of these advanced
materials in the automobile
industry.
The survey results shed light on
the potential of integrating AI
and ML in V2X communications,
addressing specific challenges in
vehicular environments. The
review encompasses various
aspects such as handover
management, proactive caching,
resource allocation, beam
selection, packet routing, and
QoS prediction, providing insights
into the underlying techniques
and considerations for their
application in optimizing network
functions and supporting diverse
automotive use cases.

Several real use cases of the

proposed platform have been
included in the results section to
illustrate the potential of the
proposed framework.
A quantitative analysis of 46,145
patents from the top 19
traditional companies to identify
AI and non-AI patents revealed
that firms primarily rely on
knowledge exploitation when
designing and integrating AI-
embedded systems, surprisingly
fostering innovativeness.
Complementary qualitative
insights reveal that big data and
AI technologies are integrated
into the industrialization phase of
new vehicle development, per a
creative problem-solving patch.
Notably, this study's findings
reveal the technical and
organizational challenges limiting
data-driven innovation, thereby
paving a way for more
technologically original
innovation with big data and AI.

The results of this study aim to

shed light on the potential
benefits and obstacles associated
with Industry 4.0 in the
automotive supply chain.
Additionally, the proposed
blockchain-based model is
presented as a solution to
enhance transparency, security,
and traceability in automotive
supply chain operations.
The objective is to obtain the
desired quality results at the
lowest cost.

The results show that these

methods are effective in
protecting the model, but they
require additional costs to be
considered. Finally, To prevent
this attack, we proposed a set of
countermeasures that can be
used to detect the input testing
dataset.

The insight gained by those

predictions can be used to design
highly individual and dynamic
quality assurance tests. By
applying the framework testing
efficiency can be increased by
15%.
This research highlights the use
cases of AI in decentralized
Blockchain with smart contracts,
the company's trading policies,
and its advantages for effectively
handling market risk assessments
during socio-economic crises. The
developed model, supported by
real-time cases, incorporated
cost functions, 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 manufacturing.
Customization and graphical user
interface with cloud integration
are some challenges of this
model.
The convolutional neural
networks have become advanced
methods for classification and
detection of objects over the last
five years. At present, they work
better than conventional image
processing method set, on many
image classification data sets.
Most of these datasets are based
on the notion of concrete classes.
In this paper, we present a new
set of image classification data as
well as object detection data,
which should be easy for humans
to solve, but its variations are
difficult for CNN. The
classification performance of
popular CNN architectures is
evaluated on this dataset and
variations of this dataset may be
of interest for future research.
As a result, we were able to
identify 20 different high-level
use cases for AI along the entire
car manufacturing process. Our
panelists have completely ranked
and assessed those 20 use cases
within two different dimensions,
i.e., their estimated business
value and their realizability.
Besides being the first study to
provide such an overview at one
glance and to give such
quantitative insights on that
steeply emerging topic, four use
cases from that list have never
been discussed in connection
with car manufacturing within
the scientific literature until now
and can therefore be considered
as completely new in that regard.

The results showcase the

development of a hybrid
methodology that addresses the
challenge of automated design of
production systems, specifically
in the automotive field of body-in-
white production. The
implemented software prototype
demonstrates the practical
application of this artificial
intelligence methodology for
generating initial production
system configurations based on
real industrial scenarios.
The results obtained are very
satisfactory, with the prediction
at the average equal to 98%. This
level of accuracy encourages the
integration of this approach in
Intelligent Transportation
Systems (ITS) in the automotive
sector.

The results highlight the

significance of employing the
established location model and
the genetic algorithm solution for
minimizing the total social cost.
The evaluation index system,
based on key influencing factors,
provides a comprehensive
approach to determine the
optimal location for charging
stations, crucial for the
advancement of the new energy
automobile industry.
The result shows that the
configuration in which we used
cross-validation technique, and
we performed a duly splitting of
data, provides predictor models
with better performances.

This complete method was

applied to PSA which is entering
the Industry 4.0 era. The
obtained results provide several
interesting insights into PSA's
control indicators. Big Data,
augmented reality, collaborative
robots and Cloud Computing are
receiving a lot of attention from
PSA and are considered as
priorities for the control of its
processes.
The study aims to explore the
applicability of artificial neural
networks (ANN) in modeling the
pulsed gas metal welding process
(GMAW-P) in the automotive
industry. The proposed model is
designed to simulate the process
for operator training, enhance
welding performance by
identifying insensitive regions to
input parameter variations, and
increase the flexibility of a
robotic welding cell.

Even though this method

deviates from the conventional
approach and requires further
discussions and testing, it
demonstrates a promising and
straightforward means of
predicting the accurate
greenhouse gas savings
associated with these materials
quickly and before the design
phase of auto parts.
Following that, we identify open
challenges and discuss possible
research directions. We conclude
that (a) publicly available data would
lead to a boost in research activities,
(b) the majority of papers rely on
supervised methods requiring
labelled data, (c) combining multiple
data sources can improve
accuracies, (d) the use of deep
learning methods will further increase
but requires efficient and
interpretable methods and the
availability of large amounts of
(labelled) data.

Data from September 2020 to

June 2021 of a first-tier supplier
of the automotive industry was
used, considering two different
training periods in two groups of
production lines for a specific
product. Results of R2 and MAE
validate and show the relevance
of the proposed models. The
accuracy of these models
depends on the artificial
intelligence techniques and the
training periods.
The results of this study indicate
that RFID, IoT, CPS, and AI are the
most applied Industry 4.0
technologies in production lines
for minimizing production time
and ensuring on-time delivery.

where the results demonstrated high

predictive accuracy, enabling better
data utilization from the production
floor and improved understanding of
the plant's status.

Thus, we are at the dawn of a

revolution in the world of
transport and mobility, leading us
to ensure the movement of the
autonomous car in a safe
manner. In this paper, we
examine certain attacks on
autonomous cars such as the
denial of service attack, as well as
the impact of these attacks on
the last two levels of vehicle
autonomy.
New standards specific to ADSs
are being adopted, but some of
these also present challenges for
ML-enabled system changes.
Then there are emerging cross-
sector AI and ML standards that
will have implications for the
automotive industry and
regulators. This paper
summarises an in-depth
assessment of standards that are
anticipated to play an important
role in the safety assurance of
ADSs. Summarised findings are
provided, including key themes,
covering issues such as the use of
ADS safety cases that are
strengthened by adopting
complementary standards,
ensuring safety assurance models
cover the whole ADS lifecycle,
and the use of a systems
engineering approach that treats
safety as a dynamic control
problem and appropriately
integrates software engineering
This paper introduces a
transferable and scalable AI-
enhanced methodology for DSB
in the automotive industry, which
enhances Overall Equipment
Efficiency (OEE) by optimizing
availability through reducing the
Mean Failure Detection Time
(MFDT). Experimental
investigations in the use-case
suggest an OEE increase by over
5%.
Finally, future trends and
prospects in the area are
discussed.
Therefore, a more sophisticated
and complex approach is
required to increase the quality
score. A combination of
mathematical methods is
proposed combining t-SNE
feature representation,
convolutional neural network,
and linear programming
optimization. The model highly
relies on the tool’s continuous
degradation characteristics. The
threshold for the given binary
classification is set by maximizing
the accuracy of the detection
model. The main contribution of
the research is the method of
inconsistent sensor data
manipulation which supports a
unique combination of AI models
for early failure recognition.
The ongoing research emphasizes
a conceptual design for vehicle
software architectures in the
automotive industry, focusing on
the principles of a modern data
architecture with an emphasis on
the data-centric mindset.
Challenges and potential working
points are explored to pave the
way for transitioning from siloed
data to what is referred to as an
"AI factory."

Finally, the proposal delimitation

and future uses in the
automobile industry are given.
Finally, we present some
discussions on next-generation
intra-vehicle communication
technologies and new solutions
to protect intra-vehicle
communications in the future.

The performance evaluation in

this work is done in two parts. In
the first part, we compare our
proposed architecture with the
traditional architecture of the
plant's IIoT network. The results
of this comparison show that our
architecture is more suitable to
simplify trust management of IIoT
devices. In the second part, we
demonstrated the ability, the
adaptiveness, and the resiliency
of our proposed trust model
against behavioral changes of IIoT
nodes in malicious environments.
The results show that Deep
Learning and Random Forest are
more accurate and present
better performance for the
prediction of overall equipment
effectiveness in our case study.

Through the preventive

identification of defective
products before the final
assembly step, components can
be recovered and returned to
serial production. By optimizing
the parameters of the ANN,
incorrect classifications are
reduced, and the practical
applicability is ensured. The
approach is demonstrated in an
industrial application in the
automotive industry.