Search

International audience; Collaborative robots (cobots) have the potential to augment the productivity and life quality of human operators in the context of Industry 4.0 by providing them with physical assistance. For this reason, it is necessary to define the relationship between humans and cobots and to study how the two agents adapt to each other. However, to the best of our knowledge, literature is still missing insight into how humans perceive and react to changes in the cobot behavior (e.g. changes in the learned trajectory and in the role the robot assumes). Specifically, a study of how humans adapt to changing roles and control strategies of collaborating robots is missing. To fill this gap, we propose a human study in which 16 participants executed a collaborative human-robot sawing task where the cobot altered between three different control strategies. We examinedhuman adaptation when cobot suddenly changed the control strategy from one to another, resulting in six experimental conditions. The experiments were performed on a setup involving Kuka LBR iiwa robotic arm. The results suggest that transition influences movement performance in the early stages and at steady state, subjects prefer to abandon modes that require more effort and they adapt faster to energy-demanding modes. Finally, for the specific task we studied, subjects tend to prefer collaborative modes to ones in which the robot assumes a fixed role.

International audience; Scheduling production is an important decision issue in the manufacturing domain. With the advent of the era of Industry 4.0, the basic generation of schedules becomes no longer sufficient to face the new constraints of flexibility and agility that characterize the new architecture of production systems. In this context, schedules must take into account an increasingly disrupted environment while maintaining a good performance level. This paper contributes to the identified field of smart manufacturing scheduling by proposing a complete process for assessing the robustness of schedule solutions: i.e., its ability to resist to uncertainties. This process focuses on helping the decision maker in choosing the best scheduling strategy to be implemented. It aims at considering the impact of uncertainties on the robustness performance of predictive schedules. Moreover, it is assumed that data upcoming from connected workshops are available, such that uncertainties can be identified and modelled by stochastic variables This process is supported by stochastic timed automata for modelling these uncertainties. The proposed approach is thus based on Stochastic Discrete Event Systems models and model checking techniques defining a highly reusable and modular process. The solution process is illustrated on an academic example and its performance (generecity and scalability) are deeply evaluated using statistical analysis. The proposed application of the evaluation process is based on the technological opportunities offered by the Industry 4.0.

Published in Advances in Production Management Systems. Artificial Intelligence for Sustainable and Resilient Production Systems. APMS 2021. IFIP Advances in Information and Communication Technology, vol 630 , pp. 73-80, Springer, Cham, 2021Part II - Hybrid Approaches for Production Planning and Scheduling; International audience; This paper proposes an approach for robust scheduling on parallel machines. This approach is based on a combination of robust mathematical and discrete event systems models which are iteratively called in order to converge towards a schedule with the required robustness level defined by the decision maker. Experimentations on a small instance (10 jobs and 2 unrelated machines) and a more complex one (30 jobs and 6 uniform machines) show that this approach permits to converge quickly to a robust schedule even if the probability distribution associated to the uncertainties are not symmetrical. The approach achieves a better rate of convergence than those of the literature’s methods.

International audience; When a human is interacting physically with a robot to accomplish a task, his/her posture is inevitably influenced by the robot movement. Since the human is not controllable, an active robot imposing a collaborative trajectory should predict the most likely human posture. This prediction should consider individual differences and preferences of movement execution, and it is necessary to evaluate the impact of the robot's action from the point of view of ergonomics. Here, we propose a method to predict, in probabilistic terms, the human postures of an individual for a given robot trajectory executed in a collaborative scenario. We formalize the problem as the prediction of the human joints velocity given the current posture and robot end-effector velocity. Previous approaches to solve this problem relied on the inverse kinematics, but did not consider the human body redundancy nor the kinematic constraints imposed by the physical collaboration, nor any prior observations of the human movement execution. We propose a data-driven approach that addresses these limits. The key idea of our algorithm is to learn the distribution of the null space of the Jacobian and the weights of the weighted pseudo-inverse from demonstrated human movements: both carry information about human postural preferences, to leverage redundancy and ensure that the predicted posture will be coherent with the endeffector position. We show in a simulated toy problem and on real human-robot interaction data that our method outperforms model-based inverse kinematics prediction, sample-based prediction and regression methods that do not consider geometric constraints. Our method is validated on a a collaboration scenario with a human interacting physically with the Franka robot.

Published in IFAC-PaperOnLine 54(1):1168-1173, 2021; International audience; Industry 4.0 reflects a new area for production workshops. This new concept aims to bring more flexibility and agility to the production shop floor. In this context, the scheduling problem is an important issue. In fact, a proposed schedule has to guarantee a high performance level able to take into consideration the several changes and perturbations occurring in the workshop. The proposed approaches usually aims to find a robust schedule capable of optimizing some criteria with considering a perturbation. With the new requirements of the workshop and the importance of decision making when implementing the schedule, it is essential to extend the robust scheduling problem to be adaptable to the needs of the decision maker in evaluating several properties. We propose, in this paper, a robust scheduling framework based on a robustness specification. The paper gives also the using of this framework in a decision making context.

International audience; This chapter presents a generic methodology when considering robustness in production systems of Industry 4. It is the first milestone for coupling Operations Research models for robust optimization and Discrete Event Systems models and tools for property checking. The idea is to iteratively call Operations Research and Discrete Event Systems Models for converging towards a solution with the required robustness level defined by the decision-maker.

Published in IFAC-PapersOnLine (51)11:758-763, 2018; International audience; This work is a positioning research paper for energy efficient building based on ICT solutions. Through the literature about the solutions for energy control of buildings during operational phase, a 3-layers model is proposed to integrate these solutions: first level consists in communication technologies, second level is about data modelling and third level is related to decision-making tools. For each level, key research topics and remaining problems are identified in order to achieve a concrete step forward. 1. CONTEXT AND PROBLEMATICS Through studies on ICT solutions for energy control of buildings, a 3-layers model is proposed to integrate these solutions and position a new way for energy efficiency. The building sector is the largest user of energy and CO 2 emitter in the EU, estimated at approximately 40% of the total consumption (Sharples et al., 1999). According to the International Panel on Climate Change (European Union, 2010), 30% of energy used in buildings could be reduced with net economic benefits by 2030. Such a reduction, however, is meaningless unless "sustainability" is considered. Because of these factors, healthy, sustainable, and energy efficient buildings have become active topics in international research; there is an urgent need for a new kind of high-technology driven and integrative research that should lead to the massive development of smart buildings and, in the medium term, smart cities. From a building lifecycle perspective, most of the energy (~80%) is consumed during the operational stage of the building (European Union, 2010) (Bilsen et al., 2013). Reducing building energy consumption may be addressed by the physical modifications which can be operated on a building like upgrading windows, heating systems or modifying thermic characteristics by insulating. Another possible path to reduce the energy consumption of a building is to use Information and Communication Technologies (ICT). According to the International Panel on Climate Change, a reduction of energy even greater than the 30% can be targeted by 2030 by considering ICT solutions. In support of this claim, some specialists believe that ICT-based solutions have the potential to enable 50-80% greenhouse gas reduction globally. In this respect, ICT innovation opens prospects for the development of a new range of new services highly available, flexible, safe, easy to integrate, and user friendly (Bilsen et al., 2013). This, in turn, should foster a sophisticated, reliable and fast communication infrastructure for the connection of various distributed elements (sensors, generators, substations...) that enables to exchange real-time data, information and knowledge needed to improve efficiency (e.g., to monitor and control energy consumption), reliability (e.g., to facilitate maintenance operations), flexibility (e.g., to integrate new rules to meet new consumer expectations), and investment returns, but also to induce a shift in consumer behaviour.

Published in IFAC-PapersOnLine, 51(7):271-276, 2018; International audience; Discrete Event Systems (DES) have already proved their efficiency to model and solve scheduling production problems in the field of manufacturing systems. Their main asset is the capacity to represent the dynamic behavior of workshops.The stochastic extensions of DES models offers, actually, interesting characteristics to take into consideration the perturbations in scheduling. In this paper, we propose to use a stochastic DES approach (using stochastic timed automata) to evaluate the robustness of schedules under perturbations. The perturbations that are considered are the machine failures with uncertain reparation duration.

International audience; The current Internet-of-things (IoT) hype, pushed by the unprecedented rate of the technological enablers’ innovation, is threatening to leave behind some major, not so obvious, unresolved issues. IoT platforms will extend existing enterprise information systems (EIS) infrastructures to encompass cross-domain sensing and actuating capabilities, thus introducing additional complexity and major risks to the implementation. Furthermore, IoT platforms are typically driven by models of the trivial complexity; they support very simple data structures and almost no business logic implementation. Finally, IoT systems are today managed centrally, which often means less openness, less flexibility and greater change management costs. In this article, we provide the overview of the scientific disciplines which could contribute to the resolution of the IoT implementation problem, namely requirements engineering, change management/continuous improvement, model-based systems engineering, system architecture design, interoperability and policy and regulatory aspects. Then, we identify the challenges of these contributions in the context of IoT and finally make an attempt to identify research directions which could have a significant impact. The discussion of the challenges and opportunities is illustrated by the proposed domain framework for implementation of open IoT ecosystems.

Published in IFAC-PapersOnline, 51(11):921-926, 2018; International audience; Six sigma is a well-assessed methodology for optimizing the performance of the enterprise through the reduction of the variability of their process outputs. Six sigma implementation consists of the development of the DMAIC phases to improve existent processes. Each phase of this methodology adopts statistical as well as qualitative tools to fix variability sources and assess the process to a six sigma level of quality. However, there is a lack of an implementation model on how to guide effectively the implementation of the methodology DMAIC. This is particularly true for SMEs that not have enough human resources for supporting the needed knowledge. To answer to this limitation of six sigma, the objective of this paper is to present a framework to support the implementation of six sigma approach in SMEs. This framework is formalized by a meta-model, which is based on different studies on six sigma, its phases and its tools. This framework aims to propose a solution to validate the passage between the DMAIC phases defining the requirements to be satisfied during each phase.

Published in Service Orientation in Holonic and Multi-Agent Manufacturing, Borangiu T., Trentesaux D., Thomas A., Cardin O. (eds). Studies in Computational Intelligence, vol 762, pp. 345-357, Springer, Cham; International audience; Industry 4.0 implies new scheduling problems linked to the optimal using of flexible resources and to mass customisation of products. In this context, first research results show that Discrete Event Systems models and tools are a relevant alternative to the classical approaches for modelling scheduling problems and for solving them. Moreover, the challenges of the industry 4.0 mean taking into account the uncertainties linked to the mass customisation (volume and mix of the demand) but also to the states of the resources (failures, operation durations,. . .). The goal of this paper is to show how it is possible to use the simulation based on statistical model checking for taking into account these uncertainties and for evaluating the robustness of a given schedule.

International audience; Global competitiveness has challenged manufacturing industry to rationalize different ways of bringing to the market new products in a short lead-time with competitive prices, high quality, and customization. Modern Product Development Process (PDP) has been requiring simultaneously collaboration of multiple groups, exchanging information from multiple perspectives within and across enterprise boundaries. However, semantic interoperability issues (misinterpretation and mistakes) in view of the information heterogeneity from multiple perspectives and their relationships. This research proposes a semantic reconciliation view to support the interoperable information relationships in product design and manufacturing. This view is part of the conceptual framework of an Interoperable Product Design and Manufacturing System (IPDMS). The semantic reconciliation method uses three approaches (Adjustment Context, Ontology Intersection, and Semantic Alignment) to provide support for the semantic information relationships across the product design and manufacturing. The method is applied in a rotational thin-wall plastic injected design and manufacturing and evaluated through the development of semantic rules responsible for the information mapping of sharing, conversion and translation. This semantic rules were modeled in Semantic Web Rule Language (SWRL) supported by Web Ontology Languages (OWL). Sequentially, the potential benefits and limitation of the method was discussed, contributing to the semantic information interoperability during the development of complex products.

Published in IFAC-PapersOnLine, 50(1):1225-1230, July 2017; International audience; This paper investigates the relevance of modeling workshop scheduling problems using a Discrete Event System (DES) approach based on timed automata (TA). To realize this study, the DES approach is compared with a classical approach based on Mixed Integer Linear Programming (MILP). In order to compare these two modeling approaches, an evaluation system is proposed based on: (i) a problem instances generation system using the classical three-fields notation of Graham (alpha|beta|gamma), (ii) a measurement system based on three criteria and associated metrics: complexity, genericity and scalability. This system covers the most common problems when dealing with flexible workshop scheduling. The results obtained by the application of the method is finally discussed.

International audience; Global competitiveness challenges manufacturing industry to rationalise different ways of bringing to the market new products in a short lead-time with competitive prices while ensuring higher quality levels and customisation. Industries need to effectively share heterogeneous information during Product Development Process (PDP) within and across their institutional boundaries to be competitive. However, problems with misinterpretation and mistakes have been identified during information exchange due to the semantic interoperability obstacles. Thus, this research proposes a systematic literature review to identify the main researches and the milestones reference works on semantic interoperability field. A rigorous methodology was conducted in different databases, covering the articles published in scientific journals from 2005 to 2015 as a preliminary study had indicated that the incidence of articles related to the subject was more frequent from the second half of the 2000s. The research structure consisted of four steps: Survey-searching, analysis and selection of recent researches; Categorization-categorization of the selected papers; References citation frequency analysis-the selected papers were analysed and the main researches and milestones references were identified; and Main researches critical analysis – the main researches were analysed for their contributions and limitations, their contributions and limitations, resulting in 14 selected scientific articles and 8 identified milestones references. It is evident that this field has interesting perspectives on future research opportunities on semantic interoperability of information issues across PDP, contributing to the new concepts of future factories.

Une nouvelle économie axée sur l’utilisation plus efficiente des biens a vu le jour au cours des dernières années : l’économie collaborative. Désormais présente et fleurissante dans de nombreuses sphères de l’économie française, cette nouvelle réalité a entraîné son lot de préoccupations. L’une d’elles, moins traitée en France, est le fait que ces plateformes constituent un terreau fertile pour les pratiques discriminatoires. Puisqu’elles requièrent généralement l’affichage d’une photo, d’un profil public et de nombreuses informations personnelles, certains préjugés et stéréotypes enracinés dans la conscience collective sont susceptibles d’entrer en jeu et d’entraîner des différences de traitements. Dans ce texte, l’auteur examine l’applicabilité et l’efficacité des mécanismes français de protection contre la discrimination face à ces risques. Autrement dit, ces mécanismes sont-ils adaptés à la nouvelle réalité proposée par les Airbnb, Uber et BlaBlaCar de ce monde ? A new economy based on the efficient use of goods has emerged in recent years : the sharing economy. Now flowering in many spheres of the French economy, this new reality has led to many concerns. One of them, less discussed in France, is the fact that these collaborative platforms provide a fertile ground for discrimination. Given the fact that they generally require the display of a photo, a public profile and many personal informations, prejudices and stereotypes rooted in the collective consciousness may come into play and lead to adverse treatments. In this text, the author analyzes the applicability and effectiveness of the protection against discrimination to these risks. In other words, are these mechanisms adapted to the new reality proposed by the Airbnb, Uber and BlaBlaCar of this world ?

Published in Procedia Engineering 180:812-82, 2017; International audience; An exhibition system is typically a physical open space made of several buildings devoted to performing demonstrations or social activities. Several examples of exhibition systems can be found word wide. The management of these spaces, even in presence of a limited number of activities, requires significant efforts: safety, welcome, merchandising, public services, etc. Nowadays, the new trend is to offer multiple, contemporary use of exhibition spaces, allowing exhibition events and regular social activities (commercial, cultural, etc.) to be dynamically intertwined, either to endeavor the huge economic investments behind infrastructures or to increase the attractiveness of the exhibition activities.This is the case of “Fiera del Levante”, one of the oldest exhibitions located in a fascinating central area on the seaside of the city of Bari - Italy. Here, multiple social scopes, such as multi-exhibitions and contemporary social activities related to public offices, restaurants and entertainment, live together. This is an almost unique setting, which makes Fiera del Levante (FdL) a complex system and makes the management of the exhibition a challenging co-management task.The paper deals with the development and demonstration of a complex-system modeling framework for the decision support of the co-management of Fiera del Levante exhibition system, taking into account the interaction both physical aspects (building assets, infrastructure, equipment, fittings, …) and intangible aspects (permanent and temporary events, management and business activities, etc) and, of course, considering as a main input the needs of the different stakeholders involved.

An approach for the representation, the retrieval and the reuse of design knowledge for product families design is proposed.All methods in literature rely on (at least) a partial representation based on natural language.The relation between the natural language and the ambiguity of the retrieval (and the consequent reuse) is highlighted.An anti-logicist approach to unambiguously represent the design knowledge.The impacts on the ambiguity of the retrieval and the reuse for the product family design are tested on an industrial case. The product family design is a design approach to meet the demand of customisable products. This paper deals with the knowledge representation, retrieval and reuse supporting the design stage of product families. Usually, the methods in the literature do not focus on the retrieve and the reusability of the knowledge. In other words, they do not ensure if a non-expert user can effectively retrieve and reuse the represented knowledge. To cope with this point, here, the aim is to apply an anti-logicist approach for the unambiguous design-knowledge representation to support the unambiguous retrieval and the automatic reuse of the knowledge during a product family design stage. The retrieval is unambiguous because the link between the knowledge models and the requirements is based on a syntax comparison, e.g. intervals of numbers, units of measure. An algorithm for the automatic reuse has been developed: provided an unambiguous definition of the new requirements related to the product family, the algorithm's outputs are the functional and physical definitions of all the products included in the product families, i.e. performances and CAD files. The case study is a family of components of the HVAC (heating, ventilating and air-conditioning) systems sector. Finally the advantages and issues of a potential industrial implementation are discussed.

International audience; Production planning and control and more generally taking a decision in the context of production systems often consider that input information are known, static and predictable. However, uncertainties on data and perturbations are recorded in the genetic of every production system. For instance, it is impossible to know exactly the level of the demand for a product, the availability of resources, etc. Dealing with this issue asks the question of the ability to take robust decisions against uncertainty (off-line) or the ability to be flexible (on-line). This paper proposes to analyze how Product Driven Systems – as reactive systems against unpredicted perturbations – can be part of operational research solution process against perturbations. Moreover, an overview of models and approaches for dealing with uncertainty in Operational Research is given and a first proposition is made for applying these elements into PDS as decision-making-against-perturbations engines.

Part 3: Interoperability and Systems Integration; International audience; The semantic interoperability of information has become increasingly important in Product Development Process (PDP) to support different phases during the product development. This article presents a proposal of an Interoperable Product Design and Manufacturing System (IPDMS) concept based on a set of engineering domain ontologies and sematic mapping approaches. The concept explores the potentials of semantic well-defined core-foundations in a Semantic Web ontology language. The formal core-foundations can be specialised to perform application view in Product or Manufacturing Model. The application view is used to support the information sharing between product design and manufacturing and verify the accordance with product requirements. A preliminary experimental has been realised, using a test case to share information from the design to manufacturing of an injection moulding plastic. As results, it was identified potential benefits and limitations. The mains contributions are: semantic interoperability during information sharing in PDP and analysis of inconsistencies in PDP.

International audience; This article presents an ongoing research project that aims to define a conceptual method to formally model the system requirements and their relationships in terms of transformation (translation, conversion and sharing) and traceability, based on an ontological approach. In this way, we are considering two hypothesis: (H1) Multiple phases of System Life Cycle can be supported by system requirements in a semantically interoperable manner; and (H2) System requirements formalization can ensure the comprehensibility and verifiability, reducing inconsistencies between different domains across the phases of System Life Cycle.

A draft architecture of a software platform is proposed to support software components interoperability in energy production systems. The scope of the platform is to address the sustainability of energy production and use efficiency. A first prototype of this platform is under development in the framework of the French ANR Plate-Form(E)3 project. Interoperability issues have been faced to design the platform. Best available candidate technologies for implementing the platform are also discussed. One architecture for facing the basic interoperability issues is also presented. A real industrial application case example is proposed to show the potential use of the proposed architecture.

Cette thèse présente plusieurs contributions dans le domaine de l’interaction physique homme-robot. En premier lieu, elle propose une méthode pour prédire la posture humaine pendant qu’un humain interagit physiquement avec un robot. Deuxièmement, elle décrit des algorithmes et des outils de simulation pour visualiser le score d’ergonomie humaine associé au mouvement d’un humain, en temps réel, même lorsque l’humain est physiquement couplé au robot. Troisièmement, la thèse fait progresser les connaissances sur la façon de contrôler et d’adapter le comportement du robot pendant la collaboration, grâce à des études expérimentales impliquant des humains et des robots dans des scénarios de comanipulation. La première étude examine les meilleures stratégies d’impédance pour que le robot puisse collaborer avec l’humain lors de la co-manipulation d’un tuyau dans une tâche d’insertion de précision, tandis que la deuxième étude examine comment les humains s’adaptent aux stratégies d’impédance changeantes d’un robot lors d’une tâche de sciage collaborative.; The thesis presents several contributions in the area of human-robot physical interaction. First, it proposes a method to predict human posture while a human interacts physically with a robot. Second, it describes algorithms and simulation tools to visualize the human ergonomics score associated with themovement of a human, in real-time, even when the human is physically coupled with the robot. Third, the thesis advances the knowledge on how to control and adapt the robot behaviour during collaboration, thanks to experimental studies involving humans and robots in comanipulation scenarios. The firststudy investigates the best impedance strategies for the robot to collaborate with the human during a co-manipulation of a pipe in a precision insertion task, while the second study investigates how humans adapt to changing impedance strategies of a robot during a collaborative sawing task.

The main objective of this thesis is to propose an approach to consider perturbations in the context of a production scheduling problem. The proposed approach emphasizes the importance of evaluating the performance of a scheduling in a disturbed environment. Production scheduling is an important and common problem in the world of industry and scientific research. In order to propose scheduling solutions, the suggested methods propose optimal scheduling to minimize or maximize a defined criterion. This objective has certainly been adopted for several years but no longer fully meets the dynamics of production workshops. Taking perturbations into account is therefore a major issue in the scheduling problem. To address this issue, after an introduction and state of the art, we propose, in the main body of the manuscript, the definition of the performance and appropriate approach to deal with the impact of perturbations in scheduling. Robustness is then identified as a suitable performance to assess this impact. Furthermore, specifying robustness allows to identify a formal framework to specify the different robustness problems that scheduling must face. To address the problem of scheduling robustness, we propose an evaluation process based on discrete-event system models. More precisely, stochastic timed automata and statistical model-checking are used to measure the robustness of a perturbed schedule. The evolution of production systems and their flexibility and agility constraints require that the evaluation approach must also be flexible and adaptable. The process proposed in this thesis allows not only its adaptation to several production workshop problems but also to take into account real decision situations in an industrial context. To illustrate the feasibility and applicability of this process, the UppAal SMC tool is used., L’objectif principal de cette thèse est de proposer une approche permettant de considérer les perturbations dans le cadre d’un problème d’ordonnancement de la production. L’approche proposée permet de mettre l’accent sur l’importance de l’évaluation de la performance d’un ordonnancement dans un cadre perturbé. L’ordonnancement de la production est un problème important et commun au monde de l’industrie et de la recherche scientifique. Dans le but de proposer des solutions d’ordonnancement, les méthodes proposées proposent des ordonnancements optimaux permettant de minimiser ou maximiser un critère défini. Cet objectif est certes adopté depuis plusieurs années mais ne répond plus tout à fait à la dynamique des ateliers de production. La prise en compte des perturbations est donc un enjeu majeur dans la problématique de l’ordonnancement. Pour répondre à cet enjeu, nous proposons, de définir la performance et l’approche à adopter pour traiter l’impact des perturbations sur l’ordonnancement. La robustesse est alors identifiée comme étant une performance adaptée pour évaluer cet impact. Plus encore, spécifier la robustesse permet d’identifier un cadre formel pour spécifier les différents problèmes de robustesse auxquels l’ordonnancement doit faire face. Pour traiter le problème de la robustesse d’un ordonnancement, nous proposons un processus d’évaluation basé sur les modèles de systèmes à événements discrets. Plus précisément, les automates temporisés stochastiques et le model checking statistique sont utilisés afin de mesurer la robustesse d’un ordonnancement perturbé. L’évolution des systèmes de production et leurs contraintes de flexibilité et d’agilité exige que l’approche d’évaluation soit également flexible et adaptable. Le processus proposé dans le cadre de cette thèse permet non seulement de s’adapter à plusieurs problèmes d’atelier de production mais également à prendre en compte des situations de décisions réelles dans un contexte industriel.