previous day
all days

View: session overviewtalk overview

08:00-09:00 Session ME: Accueil + Inscription


Location: Salle des pas perdus (1er Étage / 1st Floor)
09:00-10:00 Session ME-1: PLENIÈRE 5 : Modelling and Simulation of Aircraft at Bombardier

 PLENARY 5 -  Modelling and Simulation of Aircraft at Bombardier

Michel Rioux (ÉTS, Canada)
Location: A-1600
Fassi Kafyeke (Bombardier Aerospace, Canada)
Modelling and Simulation of Aircraft at Bombardier
SPEAKER: Fassi Kafyeke

ABSTRACT. Bombardier develops and produce business jets, regional aircraft and now, with the CSeries, single-aisle mainline jets. Aircraft have long development times requiring intensive capital investments. Numerical simulations have been identified as a key ingredient to support compressed development schedules and provide high-fidelity analyses upstream in the design process, leading to higher performance vehicles and lowering the engineering costs associated with wind tunnel, rig and flight testing. The presentation will highlight some examples of simulations in aerodynamics, structures and systems. The biggest use of simulation at Bombardier is the aerodynamics, where advanced algorithms and supercomputers are used to investigate high-speed wings, high-lift systems, aircraft icing and engine/airframe interactions. Although current simulations provide high levels of fidelity, challenging aerodynamics problems remain to be solved. Current HPC hardware trends puts the burden on software developers and researchers to develop software/algorithms that are massively parallel, scalable and computationally efficient.

10:00-10:20Pause café ( Salle des pas perdus )

Coffee Break ( Salle des pas perdus )

10:20-12:00 Session ME-2-1: SS – Pilotage des systèmes de production de soins en milieu incertain

SS – Pilotage des systèmes de production de soins en milieu incertain [FR]

Issam Nouaouri (LGI2A University of Artois, France)
Location: A-1300
Benhajji Noura (LGIMP, France)
Daniel Roy (ENIM, France)
Didier Anciaux (ufr mim, France)
SPEAKER: Daniel Roy

ABSTRACT. Les systèmes hospitaliers doivent faire face à de nouveaux enjeux. L'objectif primordial de tout système hospitalier est l'amélioration de la qualité des soins tout en apportant des gains significatifs en termes de productivité et d'efficacité. Une organisation efficiente des activités de soins s'avère alors nécessaire. Des méthodes et approches scientifiques issues du domaine du génie industriel peuvent apporter des solutions à la problématique de gestion efficience des systèmes hospitaliers. Dans le présent travail, nous proposons un système multi-agents centré-patient pour la planification et le contrôle du flux des patients. Le système proposé est un système d’aide à la décision destiné aux cadres de santé en vue d’améliorer leur fonctionnement et gérer les situations d’urgence.

Julio Montecinos (École de Technologie Supérieure, Canada)
Satyaveer S Chuhan (Concordia University, Canada)
Mustapha Ouhimmou (École de Technologie Supérieure, Canada)

ABSTRACT. In this study, we propose a new approach to balance the loads in a network of walk-in clinics (WCs) by dispersing patients based on patients’ preferences, emergency and geographical locations. WCs accept patients without an appointment and they are flexible in serving last-minute patients and non-severe emergencies. WCs are usually geographically distributed to be close to patients. Having limited resources, WCs manage the number and severity of patients they can handle. In some cases, the geographical distribution is not enough to allow natural charge balance, leaving WCs with an excess of demand in some zones and with few patients in others, which forces health services to cease. This implies that patients sometimes do not receive healthcare quickly, they leave the facilities disappointed and medical staff loses efficiency in their work. In this article, we propose a centralized system that both reduces long waiting times and balances the charge for several WCs within a network, thus preserving patients’ freedom to choose a clinic at the same time.

Bruno Houessou (École de Technologie Supérieure, Canada)
Mickael Gardoni (École de Technologie Supérieure, Canada)
Michel Rioux (École de Technologie Supérieure, Canada)

ABSTRACT. The purpose of this article is to analyze the adoption of the concepts of ISO 55000 standard in a hospital biomedical context. It focuses on a case study with a qualitative content analysis of life cycle management practices of a CT scanner (Computed Tomography scanner) in a Quebec public hospital. The results of this analysis show that there are consubstantial paradigms in hospital settings which are like ''modelators'' that distinguishe the approach of application of the ISO 55000 standard concepts for the management of hospital biomedical physical assets.

Dorsaf Daldoul (LACS ENIT / LGI2A University of Artois, Tunisia)
Issam Nouaouri (LGI2A University of Artois, France)
Hanen Bouchriha (LACS ENIT, Tunisia)
Hamid Allaoui (LGI2A University of Artois, France)

ABSTRACT. Emergency Departments (ED) is considered a complex system with random demands. The overcrowding and the limited resources affect ED patients' waiting time. The goal of this paper is to optimize the number of the human and material resources to reduce the average total patients' waiting time under uncertainty of patients' arrival. We focus particularly on medical staff (physicians and nurses) and beds in emergency department. We consider simultaneously four patients’ queues. We propose a stochastic mixed-integer programming model that is solved by a sample average approximation (SAA) approach. To solve this model, we use the solver ILOG CPLEX Optimization Studio. The program has been tested on a set of instances. Numerical results show that the patients' waiting time decreased by optimizing the number of the human and material resources.

10:20-12:00 Session ME-2-2: Simulation en foresterie

Simulation en foresterie [FR]

Catalin Riesta (FPInnovations, Canada)
Location: A-1302
Samir Haddad (FPInnovations, Canada)
Catalin Ristea (FPInnovations, Canada)
Application of optimization in forest value chain planning
SPEAKER: Samir Haddad

ABSTRACT. ForestPlan is an innovative mixed integer linear programming optimization model that can evaluate optimal harvest, allocation and manufacturing plans considering mill performance, fibre availability, supply constraints and product demand. The model links the market demand for lumber products with the supply of logs from forest operations, in order to allow tactical planning decisions to be driven by value rather than cost. ForestPlan assists planners to determine which cut blocks to harvest, which mix of log grades to recover from each cut block, where to allocate logs, what operating levels to run at sawmills, and which production campaigns to run at each mill (what products to produce from which log sorts combinations), in order to maximize the overall network profit. Once a feasible optimal plan is found, users can quickly run “what-if” scenarios to determine the impact of adding or removing constraints, such as more manufacturing capacity, less demand for certain products, predefined flow between specific sources and destinations, changing market prices, the value of additional log purchases, and harvesting a different log profile. In a recent implementation with a large Canadian forest products company, ForestPlan identified potential margin increases by optimizing log sorting in the cut blocks (including in which cut blocks is profitable to sort logs by mill destination, and in which is not), and allocating “right sized” wood to mills in order to maximize production and overall network profits. ForestPlan was also used to investigate the expected payback of capital investments at sawmills, which is a novel use of this type of decision support models.

Vincent Monbourquette (FPInnovations, Canada)
Fattane Nadimi (FPInnovations, Canada)
Catalin Ristea (FPInnovations, Canada)
Application of simulation in wood yards

ABSTRACT. Wood yards are a critical component of supply chains that distribute logs from harvest sites to sawmills. Using a commercial flow simulation software, FPInnovations has developed discrete event simulation models for better wood yard efficiency, which has been applied to different types of yards. In this study, the simulation platform has been adapted to the particular case of log sort yards, where suitable log sorts need to be produced in order to match the demand profile. Log sorts are an important wood fibre characteristic of the forest-to-mill value chain. Sort yards operations have many different types of activities, including truck/barge unloading, spreading loads on the ground, log grading and sorting, moving sorted logs into log bunks, consolidating bundles and moving them to shipping dock/area. Due to the complexity, timing and interconnectivity of these processes, it is very difficult and expensive to evaluate by traditional methods the impact of changes in operational parameters such as yard layout, logistics, and resource capacities. To address the specific complexities of log sort yards, FPInnovations has developed a tool to support decision makers. The simulation model is able to identify productivity and efficiency gains by quickly analyze “what-if” scenarios and quantify the impact of changes. In a recent industrial implementation, a simulation model of a log sort yard was developed to evaluate “what if” questions, such as different yard layout configurations, number and location of log bunks, number of log sorts per truck load, and number of yard machines. Simulation model results indicate that reducing the number of log sorts per truck load would decrease the number of wheel loader trips by up to 26%, and installing a new type of log bunks would potentially increase capacity of a bottlenecked machine by 42%.

Louis-Alexandre Lapointe Pelletier (École de technologie supérieure, Canada)
Mikael Rönnqvist (Université Laval, Canada)
Mustapha Ouhimmou (École de technologie supérieure, Canada)

ABSTRACT. The distribution of materials is often affected by undesired congestion in the supply chain. To tackle this problematic, the use of simulation models may help to have a better understanding of the causes, their effects on the system and to test alternative solutions. A simulation model has been developed in collaboration with a forestry company in the province of Quebec, Canada, to assess the wood chips distribution. One of the major objectives is the reduction of the waiting time at the pulp mills. One particular aspect that has an impact is the unloading process and thus, has to be included in the model. The simulation-based scenarios have been created considering different types of haulage, the reorganization of the processes, the adding of equipment and routing in the system.

10:20-12:00 Session ME-2-3: Gestion de l'innovation

Gestion de l'innovation [FR]

Jean Renaud (INSA Strasbourg, France)
Location: A-1350
Ahmed Cherifi (ÉTS, Canada)
Mario Dubois (ÉTS, Canada)
Mickael Gardoni (ÉTS, Canada)
Jean Renaud (INSA.Strasbourg, France)
Rémy Houssin (INSA.Strasbourg, France)
Innovation in eco design by using Eca Triz method
SPEAKER: Ahmed Cherifi

ABSTRACT. To assist design teams in creativity session or small businesses in the development of their product, we propose a methodological approach that will drive potential users, eco-innovative concepts of product, space or process. An application of TRIZ inventive principles adapted, named Eca Triz complete reflection for the choice of eco-innovative solution. We have adopted an approach using a qualitative evaluation matrix including parameters related to the ease of use of the product related to the organizational preparation for the appropriation of an eco-design approach, in addition to the standard factors of eco-efficiency. The applicability of this method is justified by the many contradictions in the choices in a study of the life cycle and can help designers and companies to choose an approach to attain a satisfying level of eco-design for the resources invested in it. An application, based on inventive principles, will be adapted and completed.

Bruno Daille-Lefèvre (INRS, France)
Jacques Marsot (INRS, France)
Amandine Demarziani (ENSAM, France)
Alain Etienne (ENSAM, France)

ABSTRACT. Le concepteur d’un équipement de travail doit respecter des obligations réglementaires dictées par la directive européenne dite « Machines ». Aux vues des accidents du travail dus encore aujourd’hui à des causes imputables aux machines, cette injonction n’est pas suffisante pour garantir la santé et la sécurité des opérateurs. Il est donc nécessaire d’aider les concepteurs dans l’application de cette démarche de prévention intégrée à la conception

Un précédent article, présenté au MOSIM 2014, montrait les propositions de l’INRS : deux outils destinés à la spécification des besoins, basé sur l’analyse fonctionnelle, et au suivi de projets de conception, adapté de l’AMDEC (analyse des modes de défaillances, de leurs effets et de leurs criticités).Cet article montrait également l’importance de développer d’autres outils pour guider le concepteur tout au long de la démarche de création d’une machine.

Suivant cette recommandation, nous présentons ici la genèse et l’analyse d’une nouvelle proposition d’outil d’aide à la conception intégrant la notion d’usage. Cette notion d’usage de la machine était déjà au cœur des deux premiers outils, avec la conviction suivante : pour être sûre, une machine doit en premier lieu être adaptée au travail réel de l’opérateur. L’Usage de la machine, le Travail de l’opérateur, répondront ici, dans le cadre de la conception de machines, au concept générique de tâches de travail.

Patrick Mbassegue (École Polytechnique de Montréal, Canada)
Ahmed Cherifi (École de Technologie Supérieure (ETS), Canada)
Mickael Gardoni (École de Technologie Supérieure (ETS), Canada)
Écoinnovation pour les PMO: caractérisation d'un modèle basé sur la gestion des connaissances
SPEAKER: Ahmed Cherifi

ABSTRACT. La prise en compte de la consommation effrénée des ressources et des rejets générés, autant de la part des sociétés que des organisations et entreprises a mis en orbite les préoccupations environnementales. La transformation attendue au niveau du produit, des procédés, du marketing ou de l’ensemble de l’organisation se fait à partir des ressources que l’on combine, que l’on associe : ce qui conduit à générer de nouvelles connaissances. Pour pouvoir les limites identifiées, il est nécessaire d’établir les rapports qui doivent prévaloir entre éco-innovation et gestion des connaissances. Les designers doivent élargir leurs connaissances environnementales, développer la co-création et s’ouvrir au génie citoyen dans un contexte multidisciplinaire. Afin d’assister des équipes de conception en séance de créativité ou des PMO dans le développement de leur produit, nous proposons une démarche méthodologique qui permettra de conduire les utilisateurs potentiels, à des concepts éco-innovants de produit, d’espace ou de processus. Comme point de départ, une matrice qualitative globale et multicritère permettra la hiérarchisation des impacts. Une application des principes inventifs TRIZ adaptée complétera la réflexion pour le choix de solution éco-innovante.

Patrick Mbassegue (ecole polytechnique, Canada)
Ahmed Cherif (École de Techologie Spérieure, Canada)
Mickael Gardoni (ETS, Canada)
L'éco-innovation pour les petites et moyennes entreprises (PME): caractérisation d'un modèle basé sur la gestion des connaissances
SPEAKER: Ahmed Cherif

ABSTRACT. La prise en compte de la consommation effrénée des ressources et des rejets générés, autant de la part des sociétés que des organisations et entreprises a mis en orbite les préoccupations environnementales. D’un côté les entreprises et particulièrement les petites et moyennes entreprises (PME) doivent répondre à des impératifs de marché de plus en plus concurrentiels et de l’autre elles doivent y arriver en respectant diverses contraintes environnementales : ce qui nécessite de travailler à différents niveaux, entres autres sur le produit, les procédés, les intrants, les processus de transformations. Étant donné que la PME dispose de ressources limitées, elle doit donc innover, soit trouver des solutions nouvelles en termes de pratiques, de savoirs et de savoir-faire pour s’assurer d’être en adéquation avec les contraintes que lui imposent les nouveaux environnements de marché. Pour pouvoir faire face à ces nouveaux défis, la PME doit maîtriser les aspects relatifs autant à l’éco-innovation qu’à la gestion des connaissances. Il est donc nécessaire d’établir les rapports qui doivent prévaloir entre éco-innovation et gestion des connaissances. Pour soutenir la PME à relever ce défi, nous proposons une démarche méthodologique basée sur une combinaison des aspects critiques de gestion de connaissances et d’éco-innovation.

10:20-12:00 Session ME-2-4: Uncertainty and Risk Modeling

Uncertainty and Risk Modeling [ENG]

Aissani Djamil (Unité de Recherche LaMOS (Modélisation et Optimisation des Systèmes), Algeria)
Location: A-1340
Rebaiaia Mohamed-Larbi (Département de génie-mécanique Université Laval, Canada)
Ait-Kadi Daoud (Département de génie-mécanique Université Laval, Canada)
Jamshidi Afshin (Département de génie-mécanique Université Laval, Canada)
Some application results for components importance measures of binary coherent systems

ABSTRACT. Reliability Important measures have been developed to provide critical information on the degree of participation of any component to lead a system to be not vulnerable. Several classes of them have been proposed to handle this problem in accordance of risk and uncertainty assessment. The most useful between these important measures are Birnbaum, Fussel-Vasely and Risk Reduction Worth. Birnbaum measure for example can determine which components to be removed or to make it stronger to conduct a system to be more robust and optimal. This paper proposes a combination of solutions based on reliability estimation using enumerating approaches and some distinctive importance measures. It demonstrates the suitability of importance measures factors for determining the risk contribution that can affect the functioning of a system. A practical example has been proposed to testify the feasibility of these important measures on systems’ networks and results are displayed in terms of a list of preference which classifies the components according to their values generated from the application of any one of the importance measures.

Rebaiaia Mohamed-Larbi (Laval University, Canada)
Ait-Kadi Daoud (Laval University, Canada)
Notes on the Mean Residual Life Function of Coherent Systems: Ap-plication for a Reliability System’s Structure

ABSTRACT. Several researchers have used the mean residual life (MRL) ageing properties to describe the deterioration or the improvement of system’s life such as in engineering, economics and biomedical applications. The MRL function is studied at the design stage to provide solutions for estimating the lifetime that remains to a device before its failure or before its renewal. This paper discusses several properties of the MRL and obtains its mathematical expression in terms of the survival function and the failure rate for certain classes of the most usual probability distributions. Applications concern reliability structure of coherent systems by presenting the case of series systems

Hesham Maghrabie (Concordia University, Canada)
Andrea Schiffauerova (Concordia University, Canada)
Yvan Beauregard (École de technologie supérieure - Université du Québec, Canada)

ABSTRACT. The intended use of MCDA within decision making context is to leave DM(s) with a better understanding about a given decision situation. Although different MCDA methods exist, the practice of combining different methods has become familiar in order to handle noticeable deficiencies in some methods. A new decision analysis approach is proposed in this research in order to address the uncertainty aspects within a complex structure. The novelty of this approach lies in the establishment of a new grey based MCDA decision structure, which integrates the grey systems theory with Analytical Network Process (ANP) and Preference Ranking Organization METHod for Enrichment Evaluation II (PROMETHEE II). The intended purpose of the proposed methodology is to tackle several sources of uncertainty within such a complex system.

Tlilane Lydia (Unité de recherche LaMOS, Algeria)
Allaoua Hanane (Unité de Recherche LaMOS, Algeria)
Benouaret Zina (Unité de Recherche LaMOS (Modélisation et Optimisation des Systèmes), Algeria)
Aissani Djamil (Unité de racherche LaMOS, Algeria)
Quelques méthodes d'approximation des probabilités de ruine et application au modèle de Lundberg

ABSTRACT. Dans la gestion du risque en assurance, la probabilité de ruine représente une mesure très importante à évaluer. Dans ce travail, nous nous intéréssons à quelques résultats et les différentes méthodes utilisés dans l'estimation des probabilités de ruine où nous présentons plusieurs expressions en temps fini et infini. En particulier, nous détaillons l'application de ces approches dans le cas du modèle de Lundberg.