Artificial intelligence (AI) has become an essential tool for manufacturers seeking to optimize their production processes, reduce costs, and improve product quality. However, the complexity of the underlying mechanisms of AI systems can render it difficult for humans to understand and trust AI-driven decisions. Explainable AI (XAI) is a rapidly evolving field that addresses this challenge, providing human-understandable explanations of AI decisions. Based on a systematic literature survey, We explore the latest techniques and approaches that are helping manufacturers gain transparency in the decision-making processes of their AI systems. In this survey, we focus on two of the most exciting areas of XAI: ontology-based and semantic-based XAI (O-XAI, S-XAI, respectively), which provide human-readable explanations of AI decisions by exploiting semantic information. These latter types of explanations are presented in natural language and are designed to be easily understood by non-experts. Translating the decision paths taken by AI algorithms to meaningful explanations through semantics, O-XAI, and S-XAI enables humans to identify various cross-cutting concerns that influence the decisions made by the AI system. This information can be used to improve the performance of the AI system, identify potential biases in the system, and ensure that the decisions are aligned with the goals and values of the manufacturing organization. Additionally, we highlight the benefits and challenges of using O-XAI and S-XAI in manufacturing and discuss the potential for future research, aiming to provide valuable guidance for researchers and practitioners looking to leverage the power of ontologies and general semantics for XAI.

Thanks to the advent of robotics in shopfloor and warehouse environments, control rooms need to seamlessly exchange information regarding the dynamically changing 3D environment to facilitate tasks and path planning for the robots. Adding to the complexity, this type of environment is heterogeneous as it includes both free space and various types of rigid bodies (equipment, materials, humans etc.). At the same time, 3D environment-related information is also required by the virtual applications (e.g., VR techniques) for the behavioral study of CAD-based product models or simulation of CNC operations. In past research, information models for such heterogeneous 3D environments are often built without ensuring connection among different levels of abstractions required for different applications. For addressing such multiple points of view and modelling requirements for 3D objects and environments, this paper proposes an ontology model that integrates the contextual, topologic, and geometric information of both the rigid bodies and the free space. The ontology provides an evolvable knowledge model that can support simulated task-related information in general. This ontology aims to greatly improve interoperability as a path planning system (e.g., robot) and will be able to deal with different applications by simply updating the contextual semantics related to some targeted application while keeping the geometric and topological models intact by leveraging the semantic link among the models.

Flexible manufacturing plays an important role in Industry 4.0 for developing the factory of the future and requires enhanced planning, scheduling, and control. The quick and effective adaptation in the production line in response to customers' requirements or face of unwanted situations will promote considerable flexibility in manufacturing. CHAIKMAT is a research project funded by the French National Agency of research that aims to add flexibility and transparency to manufacturing through trustful automatic decision-making. The project proposes a human-centric AI approach that investigates whether an available set of machines can perform a specific production process and then provides human experts with meaningful explanations of how the decision process is conducted. A hybrid predictive model, comprising of both semantic reasoning and machine learning system will help in real-time decision making through the automated analysis of two sources of information: a stream of machine-monitoring data describing the current state of the production line and a common-sense knowledge graph (MCSKG) that is modelled based on machine capability and process planning ontology model. Furthermore, this hybrid predictive model will also be able to explain its prediction so that the user can fully comprehend the rationale behind such a decision. In this paper, we will describe the architecture of the proposed system along with a detailed plan for verification. The paper also presents the state-of-the-art of AI applications in flexible manufacturing to establish how CHAIKMAT project aims to apply some of the novel AI methodologies to circumvent the existing gaps.

In light of the complexity of unfolding disasters, the diversity of rapidly evolving events, the enormous amount of generated information, and the huge pool of casualties, emergency responders (ERs) may be overwhelmed and in consequence poor decisions may be made. In fact, the possibility of transporting the wounded victims to one of several hospitals and the dynamic changes in healthcare resource availability make the decision process more complex. To tackle this problem, we propose a multicriteria decision support service, based on the Analytic Hierarchy Process (AHP) method, that aims to avoid overcrowding and outpacing the capacity of a hospital to effectively provide the best care to victims by finding out the most appropriate hospital that meets the victims’ needs. The proposed approach searches for the most appropriate healthcare institution that can effectively deal with the victims’ needs by considering the availability of the needed resources in the hospital, the victim’s wait time to receive the healthcare, and the transfer time that represents the hospital proximity to the disaster site. The evaluation and validation results showed that the assignment of hospitals was done successfully considering the needs of each victim and without overwhelming any single hospital.

The choice of the best location to install the mobile blood collection center is one of the most significant challenges of the blood supply chain. To date, there is a strong need to mobilize potential donors and maximize the amount of collected blood in order to meet the demand. However, the traditional mobile blood collection process is non-digitalized, and there is a huge amount of information that should be considered to enhance the decision-making process. Understanding data more effectively is a key factor to provide insight in order to support decision makers in choosing the most appropriate site to install the mobile truck in order to target as many donors as possible. Accordingly, an ontology-based cyber-physical social system (CPSS) is essential to improve the mobile blood collection process and digitalize it. In this paper, we present Mobile Blood Collection Ontology (MBCO), a modular suite of ontologies that formalize the mobile blood collection process. To do so, we use Basic Formal Ontology (BFO) as a top-level ontology that facilitates the integration of the different ontological modules and also some existing domain ontologies. The proposed ontology will be exploited later by a data-driven decision support system to find out the best location for the mobile center.

Knowledge engineering has a vital role in advancing the semantic web, in which ontologies play a key role in data interoperability and integration. One of the key issues in ontology engineering is how to handle the subsequent updates in the ontologies. A number of concerns need to be considered while working on the ontology change, such as, tracking ontology versions and heterogeneity issues. Ontology change management has been partially addressed by different researchers in overlapping research areas. However, a concrete description of the problem and its related concerns are still not available in the literature. Our work aims to present an overview of ontology change management and its concerns. We point up the need for modularization in ontology change management based on its advantages in the context of ontology reuse from different contextual viewpoints. For this purpose, we propose a protege plugin for reusing OWL modules, and allowing a safe/clean manual integration and reuse of different ontology modules.

This work deals with the simulation of complex manipulation tasks in virtual environments. Validating such complex tasks, possibly to be performed under strong geometric constraints, requires considering task and path planning jointly. The contribution of this work focuses on using task-related information at the path planning level. We propose an ontology-based approach to a) model the 3D environment where the simulated task is executed, based on an original multi-level environment model involving higher abstraction level data than the purely geometric models traditionally used, and b) automatically define path planning queries for the primitive actions of a task plan, together with task-related geometric constraints on these queries. This approach allows the improvement of the state of the art from two points of view. First, our joint task and path planning approach allows the improvement of path planning through better semantic control of the path planning process. Second, if compared to hard-coded geometric constraints, the proposed ontology-based approach introduces a more flexible way of defining geometric constraints through an inference process, and can be adapted to different applications of manipulation tasks.

Managing complex disaster situations is a challenging task because of the large number of actors involved and the critical nature of the events themselves. In particular, the different terminologies and technical vocabularies that are being exchanged among Emergency Responders (ERs) may lead to misunderstandings. Maintaining a shared semantics for exchanged data is a major challenge. To help to overcome these issues, we elaborate a modular suite of ontologies called POLARISCO that formalizes the complex knowledge of the ERs. Such a shared vocabulary resolves inconsistent terminologies and promotes semantic interoperability among ERs. In this work, we discuss developing POLARISCO as an extension of Basic Formal Ontology (BFO) and the Common Core Ontologies (CCO). We conclude by presenting a real use-case to check the efficiency and applicability of the proposed ontology.

Disaster response is a highly collaborative and critical process that requires the involvement of multiple emergency responders (ERs), ideally working together under a unified command, to enable a rapid and effective operational response. Following the 9/11 and 11/13 terrorist attacks and the devastation of hurricanes Katrina and Rita, it is apparent that inadequate communication and a lack of interoperability among the ERs engaged on-site can adversely affect disaster response efforts. Within this context, we present a scenario-based terrorism case study to highlight the challenges of operational disaster command and response. In this article, which is based on the French emergency response doctrine, we outline a semantics-based common operational command system that is designed to guarantee an efficient information flow among ERs. Our focus is on offering to all ERs, a real-time operational picture of the situation in order to enable multilevel coordination among firefighters, police, healthcare units, public authorities, and other stakeholders. Our approach consolidates information to promote timely sharing of data among ERs. The proposed system is based on an ontology that has been developed to represent the different types of knowledge on the part of ERs, providing a shared vocabulary that covers a variety of interoperability concerns.

Disaster response requires the cooperation of multiple emergency responder organizations (EROs). However, after-action reports relating to large-scale disasters identity communication difficulties among EROs as a major hindrance to collaboration. On the one hand, the use of two-radio communication, based on multiple orthogonal frequencies and uneven coverage, has been shown to degrade inter-organization communication. On the other hand, because they reflect different areas of expertise, EROs use differing terminologies, which are difficult to reconcile. These issues lead to ambiguities, misunderstandings, and inefficient exchange of data and information among those involved, which can impede the response process and slow decision making. We, therefore, hypothesize that promoting semantic interoperability across ERO information systems might improve information exchange among stakeholders and thereby allow a more coherent response to the disaster. We propose an ontology-based messaging service on the basis of the Emergency Data Exchange Language (EDXL) standards. The parties involved will continue to use the terminologies to which they are accustomed, but the system will resolve inconsistencies and thereby enhance mutual understanding among EROs by ensuring semantic translation of the exchanged information. The evaluation of the semantic translation demonstrated the effectiveness and accuracy of the proposed service.

La gestion opérationnelle de situations de crise nécessite, selon l’importance et l’étendue de la crise, la mobilisation rapide et la coordination des différents services de secours. Malheureusement, cette coordination interservices est un exercice très délicat du fait de la diversité des acteurs intervenant sur le terrain et de l’hétérogénéité des différentes organisations. Aujourd’hui, il y a un manque de coordination, l’information n’est que très peu partagée entre les acteurs opérationnels et la communication n’est pas formalisée. Ces inconvénients conduisent au dysfonctionnement des réponses aux situations de crise. Afin de mieux répondre aux situations de crise, nous proposons POLARISC, une plateforme interopérable de coordination interservices pour la gestion opérationnelle de catastrophes visualisant en temps réel le théâtre des opérations. L’objectif de POLARISC est d’aider à la décision quel que soit le niveau de commandement. Pour atteindre ces objectifs, le premier enjeu de cette thèse est de garantir une interopérabilité sémantique entre les différents acteurs métiers pour assurer l’échange et le partage des informations. À cet égard, l’idée est de formaliser sémantiquement les connaissances des acteurs métiers de la gestion opérationnelle à l’aide des ontologies. En effet, nous proposons une approche fédérée qui représente les données, les services, les processus et les métiers de chaque acteur. Nous avons modélisé les connaissances des acteurs de secours en développant une ontologie modulaire (POLARISCO) comportant un module ontologique pour chaque acteur de secours et intégré ces derniers pour proposer un vocabulaire partagé. L’utilisation des ontologies de haut niveaux et des ontologies intermédiaires, respectivement « Basic Formel Ontology » et « Common Core Ontologies », facilitent l’intégration de ces modules et de leurs mappings. Le deuxième enjeu est d’exploiter ces ontologies afin de diminuer l’ambigüité et d’éviter la mal interprétation des informations échangées. Par conséquent, nous proposons un service de messagerie appelé PROMES transformant sémantiquement le message envoyé par un acteur émetteur selon le module ontologique de l’acteur destinataire. En effet, PROMES se base sur l’ontologie POLARISCO et sert à enrichir sémantiquement le message pour éviter tout type d’ambiguïté. Le fonctionnement de PROMES est basé principalement sur deux algorithmes ; un algorithme de transformation textuelle, et par la suite, un algorithme de transformation sémantique. Ainsi, nous avons instancié l’ontologie POLARISCO avec des données réelles de la réponse aux attaques terroristes de Paris en 2015 afin d’évaluer l’ontologie et le service de messagerie. Le troisième et dernier enjeu est de proposer un service d’aide à la décision multicritère qui permet de proposer des stratégies d’évacuation des victimes après le lancement du plan blanc. L’objectif est de trouver les structures hospitalières les plus adaptées à l’état de la victime. Le choix de l’hôpital le plus approprié dépend de la durée du transport, et surtout de la disponibilité des ressources matérielles et humaines, de façon à prendre en charge les victimes le plus rapide que possible. Notre étude comprend deux étapes : la première étape consiste à développer un module ontologique qui associe à chaque pathologie les ressources indispensables pour une meilleure prise en charge des victimes selon leurs états. La deuxième étape consiste à développer un algorithme qui permet de vérifier la disponibilité des ressources nécessaires, calculer le temps d’attente pour que la victime soit prise en charge dans chaque hôpital et par la suite choisir l’hôpital le plus approprié

Mobile blood collection provides an opportunity for increasing the number of donors to the fact that the variety of locations facilitates the donation by reducing the distance traveled by donors. In that sense, blood transfusion service in collaboration with blood collection institutions, in order to meet demand, uses a mobile blood collection to maximize the amount of collected blood. However, the traditional blood collection process is non digitalized, which may limit decision-making in the choice of the best location of the mobile center, resource management, and the waiting time that sometimes is too long for donors. This paper addresses the mobile blood collection process by proposing its digitalization through a Cyber-Physical Social System (CPSS) inspired by the centric network approach. The objective of this CPSS is to make the mobile blood collection process more intuitive and set up a data-driven decision.

In the context of globalization and knowledge management, information technologies require an ample need of unprecedented levels of data exchange and sharing to allow collaboration between heterogeneous systems. Yet, understanding the semantics of the exchanged data is one of the major challenges. Semantic interoperability can be ensured by capturing knowledge from diverse sources by using ontologies and align these latter by using upper-level ontologies to come up with a common shared vocabulary. In this paper, we aim in one hand to investigate the role of upper-level ontologies as a mean for enabling the formalization and integration of heterogeneous sources of information and how it may support interoperability of systems. On the other hand, we present several upper-level ontologies and how we chose and then used basic formal ontology (BFO) as an upper-level ontology and common core ontology (CCO) as a mid-level ontology to develop a modular ontology that defines emergency responders’ knowledge starting from firefighters’ module for a solution to the semantic interoperability problem in emergency management.

Up to now, the world has witnessed how inadequate communication capabilities can adversely affect disaster response efforts. There are various Emergency Responders (ERs) that potentially must work together towards a successful resolution of the disaster. However, the different terminologies and technical vocabularies that are being exchanged between the ERs may lead to a misunderstanding and lack of semantic integrity. Yet, understanding the semantics of the exchanged data is one of the major challenges. The purpose of this work is to define the complex knowledge of the ERs by proposing a common and modular ontology shared by all the stakeholders so as to come up with a common shared vocabulary in order to ensure semantic interoperability between ERs. In this paper, we present POLARISCO and we discuss how it was developed using Basic Formal Ontology as an upperlevel ontology and Common Core Ontology as a mid-level ontology to define each module.