AbstractWhen designing complex systems, multiple people contribute to the process of information collection in support of decision making. In this paper, we study information collection in the Issue Resolution Decision Support (IRDS) framework. We assess the difficulties associated with uncertainty in the often scarce data when implementing the framework in a company and map out how the data sources are scattered across the organization. We study the elicitation process and propose to leverage sensitivity analysis to better allocate data collection efforts.

Decision-makers often rely on heuristics and experience to make complex decisions in the industrial context. Often, integrating implicit or expert knowledge as well as uncertainties can lead to decisions that are not necessarily the best ones. Moreover, in engineering design, the decision-making approaches focus on the product itselfand do not investigate the necessary effort that is needed to gather additional data in order to devise more precise decision-making models. In our research, we propose to integrate this estimation of additional effort needed for data gathering and decision-making refinement in order to support design teams. This research has been conducted in collaboration with a major car manufacturing company, and in particular in the development process through Modeling and Simulation. The objective is to propose a decision-making model that integrates data-gathering estimation, hence integrating also the estimation ofpostponing one decision. A decision problem model based upon expected utility combined with the value of information theory is proposed to address this issue. The model has been developed and tested on 4 case studies. We define a decision support framework by integrating the model into a tool and by proposing roles in the decision-making process. We finally present its application on a concrete example.

Nowadays, simulation technologies are becoming indispensable to support the design process, and notably during the verification and optimization phases. However, when building a simulation architecture that has to provide the systems architects with answers, simulations architects have to correctly specify the corresponding models required. In a MBSE concept, we propose to extend the concept of model of intention, to verify the specification of these models, to model suppliers by providing them with a Model Request Package. Models of Intention will allow an executable verification of the simulation architecture before the integration of the final models of realization.

The design process can be considered as series of decisions supported by modeling and simulation (M&S). Current developments aim at supporting this decision making with regard to increasing resources committed in the M&S process. To understand possible decision support, we conducted an empirical study in a car manufacturing company to map out the decision-making process during the development phase. A qualitative data analysis was performed to understand the difficulties and the needs expressed by decision makers. Industrial preliminary observations have shown that decisions regarding design issues are often postponed, causing iterations, and time and cost overruns in the development process. The study revealed that decisions are escalated to upper hierarchical levels as complexity and uncertainty increase and as the tradeoffs become impactful. A lack of knowledge about the M&S performance and limits, a lack of clarity due to design ambiguity, and uncertainty are more likely to cause iterations and delay. In addition, decision makers and stakeholders are sometimes unadvised of the influence of the decision under consideration on subsequent decisions and on the profit. These findings are interesting as they shed light in terms of decision supported needed in the future.

Integration, Verification and Validation (IVV) practices in simulation-based design helps reduce inconsistencies in multidisciplinary systems, i.e. those combining multiple mechanics, structural, hydrodynamic or other complex components. In current multidisciplinary simulation model development processes, subsystem simulation models are usually Verified and Validated (VV in such scenarios, each system is verified and validated separately – mechanical, structural, hydrodynamic, etc. However, many problems may arise during the actual integration of these modular subsystem simulations at the Original Equipment Manufacturer (OEM) level, which increases the risk of late inconsistencies such as interface mismatches or other interoperability-based problems. To address this problem, the present work aims to reduce late inconsistency detection through ensuring early stage collaborations between the different suppliers and the OEM by proposing a clear simulation model request. Our approach is illustrated with an industrial case study showing how a Model Request Package that contains the Model Identity Card (MIC) and Model of Intention (MoI) concepts, helps reduce the knowledge gap and inconsistencies between OEMs and model suppliers.

Today, one of the major challenges in full-vehicle model creation is to get domain models from different experts while detecting any potential inconsistency problem before the Integration, Verification, Validation, and Qualification phase. To overcome such challenges, the conceptual design phase has been adapted to the current model development process. For that, the system engineers start to define the most relevant system architecture by respecting quality and time constraints. Next, the simulation model architects design the delivered system architecture in a more formal way with a modeling and simulation point of view to support the integration of domain-level simulation models in a consistent fashion. Finally, the model architects negotiate with different simulation model providers with the aim of specifying vehicle- and domain-level simulation models and their interface connections. To improve knowledge sharing between the mentioned actors, we propose a model identity card (MIC) for classifying simulation model knowledge, including input/output parameters, method, and usage specifications. The fundamental concepts that form the basis of all simulation models are identified and typed for implementation into a computational environment. An industrial case study of the engine-after-treatment model is used to show how MICs and the integrated model design phase might be used in a given scenario. A validation protocol is conducted through a heuristic observation to estimate the rate of model rework and ambiguity reduction.

Modeling a complex system implies the integration of different simulation models in various fields of expertise. These models should communicate with each other to simulate the behavior of the whole system. In this multidisciplinary context, the actors involved in the modeling process should deal with three main problems. Firstly, in order to reduce ambiguity, they need a common vocabulary and format to describe their models in a less informal way. Secondly, in order to reduce the cost of lately correction, any potential incompleteness and inconsistency problems related to the models should be identified in the early phases of creation and integration of models. Thirdly, the characterization of simulation models should allow actors to reuse existing models more efficiently. In this poster, we propose a common framework called Model Identity Card (MIC) to specify and characterize simulation models contents and interfaces. This new concept is implemented in arKItect (a MBSE tool) to facilitate the knowledge sharing between different actors. It allows users to reduce time to get a correct model by checking the completeness and consistency of their models throughout the modeling process. An industrial test-study in automotive industry is presented to illustrate the interest of the proposed approach.

Integration and coordination of engineering analysis model is a vast development field in the context of complex product development. Engineers' siloed way of working in combination with lack of efficiency in current model development process may cause inconsistency based on model interfaces, human errors, miscommunication between teams and misinterpretations. In lean terms, this may create multiple wastes, including waiting, overproduction leading to excess inventory, unnecessary processing and may be the most harmful: defects (e.g., incorrect models) with rework consequences. Hence, product manufacturing companies must establish effective processes to add value throughout the multidisciplinary distributed modeling environment. The goal of this paper is to propose a pull-control model development process, providing model architecture integration and coherent control in early design phase. This paper proposes also an appropriate reuse strategy; this allows for utilizing plug-and-play type modular product models managed through a single-source of authority concept. A pull-control development process helps prevent potential rework arising from inconsistencies related to definitions, know-how and stakeholders communication at an early stage of the design process. Also, the proposed black box models reuse strategy helps reduce human-related error such as lack of domain knowledge, experience and misinterpretations. The proposed method is used to identify and visualize potential improvement in terms of increased model transparency and reuse when transforming from the present to the suggested future modeling strategy. The research has been conducted by synthesizing findings from a literature review, in combination with observations and analysis of current analysis model development practices within the automotive OEM Renault in France.

Complex product development processes are evolving towards simulation driven design which leads to many heterogeneous computational models and design teams that interact with each other. However, this interaction creates a bottleneck for communication and models reuse throughout the design process because, very often, the model provider (i.e. analysts) and model users (i.e. designers) do not have the same level of understanding. In addition, the tools such as PDM (Product Data Management) or SDM (Simulation Data Management) consider the numerical models as black-box documents and they cannot access or link parameters and variables of models. The poverty of semantics in terms of simulation logics and design leads to a lack of interoperability between the contributing disciplinary simulation components, herein called numerical models. To reinforce this semantics, it is necessary to create a semantically-rich model characterization support to reduce knowledge gap between model provider and user, and to achieve a higher level of reuse. This work aims to introduce the first necessary step, herein, creation of domain ontology for formally characterizing reusable numerical model. Based on this common vocabulary, in automotive context, a Model Identity Card (MIC) is developed as an intermediate support which characterizes a model into five attributes; Physical Object, Interface, Methods, Means Usage, Validation and Verification. The MIC is illustrated with a Vehicle Thermic Comfort model example and a computer interface is developed to collect a series of representative MICs in a database.

The major concern of past and present industry is to find how to make and supply the products for fulfilling customer needs with minimal cost and time. With this motivation, Renault recently decided, in order to improve efficiency and reduce costs, to re-design their distributed and heterogeneous thermal comfort simulation model’s activities in order to have a complete simulation environment. The purpose of this work is to represent the preliminary steps in achieving this goal in building a change process model which aid in V model requirement elicitation phase. So, in this work, we propose an extended V model based on Goal Oriented Requirement Engineering (GORE) for complex system’s requirement elicitation. Is the existing approach ineffective? What really are the research issues?

INSIGHTVolume 16, Issue 4 p. 30-32 Special Feature Creating a Common Vocabulary to Support the Exchange of Numerical Models between Suppliers and Users in a Complex System Design Göknur Sirin, Göknur Sirin goknur.sirin@ecp.fr Search for more papers by this authorBernard Yannou, Bernard YannouSearch for more papers by this authorEric Landel, Eric LandelSearch for more papers by this author Göknur Sirin, Göknur Sirin goknur.sirin@ecp.fr Search for more papers by this authorBernard Yannou, Bernard YannouSearch for more papers by this authorEric Landel, Eric LandelSearch for more papers by this author First published: 23 June 2015 https://doi.org/10.1002/inst.201316430AboutPDF ToolsRequest permissionExport citationAdd to favoritesTrack citation ShareShare Give accessShare full text accessShare full-text accessPlease review our Terms and Conditions of Use and check box below to share full-text version of article.I have read and accept the Wiley Online Library Terms and Conditions of UseShareable LinkUse the link below to share a full-text version of this article with your friends and colleagues. Learn more.Copy URL Share a linkShare onFacebookTwitterLinkedInRedditWechat No abstract is available for this article. Volume16, Issue4December 2013Pages 30-32 RelatedInformation

Systems verification requires first to model the system to be verified, then to formalize the properties to be satisfied, and finally to describe the behaviour of the environment. This last point, known as the proof context, is often neglected. It could, however, be of great importance in order to reduce the complexity of the proof. The question is then how to formalize such a proof context. This article review a language, named CDL (context description language), that is proposed for expressing formal specifications of an execution context, including attachment of properties to specific regions in this context. We show that such contexts can be translated into timed automata, and can then be integrated into a timed model checker. Our contribution is a report on several experiments that they have carried out on software from the aviation and military industries.

Resume. Pour ameliorer les pratiques dans le domaine de la validation formelle de modeles, nous explorons un axe de recherche dans lequel nous formalisons la notion de « contexte de preuve » integrant la description du comportement de l’environnement interagissant avec le modele et les proprietes a verifier dans ce contexte. L'article presente le langage CDL (Context Description Language) propose a l’utilisateur pour la description des contextes de preuve. Ceux-ci sont exploites, actuellement dans nos travaux, par une technique de verification de type model-checking avec la mise en œuvre d’observateurs. Dans une approche Ingenierie Dirigee par les Modeles (IDM), les modeles de contextes sont transformes en modeles d’automates temporises puis en codes exploitables par l’outil OBP/IFx (Observer-Based Prover). Ce travail a donne lieu a plusieurs experimentations industrielles comme la validation formelle d’un protocole de communication avionique pour l’AIRBUS A380. Dans cet article, nous decrivons l’application de notre approche pour la validation d’un modele de controleur de systeme aerien concu par THALES. L’article rend compte de la mise en œuvre du langage CDL et d’un retour d’experience.

The assessment of mechanical performances, in the automotive engineering domain, is mainly at present time, the result of late finite element analysis processes (FEA) which remain computationally expensive, limiting their use to the analysis of a limited number of design alternatives. But, in the conceptual design stage, the quality depends on the comprehension and on the exploration capabilities of the design space. This paper describes a strategy for building and more systematically exploring mechanical conceptual models, in the case of non-trivial expected mechanical performances. This strategy consists of a series of consistent stages: simplification of the parameterized structural model, choice of a subset of determining design parameters, computation of a limited number of approximate models of performances (metamodels obtained after a design of experiments and a model fitting) and a concept exploitation stage (deterministic exploration, optimization, non-deterministic exploration). This strategy has been successfully applied to assess vibro-acoustic performances of an automotive sub-frame. In this example, we show that designers have obtained useful information from the graphical and the numerical exploitation of this conceptual model. Moreover, this is now possible to take vibro-acoustic performances into account since the determining stage of envelope volumes allocation for sub-systems, a stage that is necessary in concurrent engineering for the automotive architecture deployment. Before, acousticians were not even able to negotiate with architects for a given volume allocation in regards to the possible consequences on the performances of which they were in charge.

This paper deals with the preliminary design of large complex mechanical products. It gives a new way to lead tradeoffs between architectural constraints and mechanical performances. While topology optimization and structural optimization have been widely developed in the last two decades [1], they are still not adapted to take into account the problem of volume allocation in the preliminary design of large and complex mechanical systems as automotive vehicles and aircrafts. New approximate criteria to assess the compliance of architectural constraints and important mechanical performances like contributions to noise are proposed. These criteria are assembled within a sole performance metamodel that embeds most of decision issues usually discussed in the preliminary design stage and which allows to encompass the strict traditional volume allocation process. Through the use of Pareto techniques, we show that it is convenient and relevant to explore the design space by allowing great variation of the design morphology. The trade-offs are made possible since the stage of volume allocation and automotive architects can be helped by tools for leading quantitative negotiations in preliminary design.

Product decomposition speeds up design stages, allows concurrent engineering and improves product quality. In the conceptual design stage, design teams focus separately on each subsystem and develop physical models (principle solution) which must meet some independent design requirements. But these requirements at the subsystems levels must be synthesized so as to meet at best the system performance targets. The paper applies a target cascading strategy to the optimization of the road noise performance of an automotive system. It focuses on how vibro-acoustic performance targets are specified and aggregated at a subsystem level (the front subframe) and how the current vibro-acoustic performances are assessed and optimized.

Dans le domaine de la conception mécanique en ingénierie véhicule, l'évaluation des performances mécaniques d'un concept est à l'heure actuelle principalement le résultat d'un processus tardif d'analyse et de simulation d'un modèle Eléments Finis, coûteux en temps de calcul et ne permettant donc d'évaluer qu'un faible nombre de configurations candidates. Or, en phase de préconception, la qualité dépend beaucoup de la compréhension et de l'exploration de l'espace de conception. Dans cet article, nous envisageons une stratégie de construction de modèles conceptuels consistant à passer par un premier modèle simplifié pour finalement aboutir à un métamodèle, fonction mathématique d'estimation rapide des performances. Cette stratégie a été appliquée au cas de l'évaluation de la prestation acoustique d'un berceau de voiture. La prise en compte de critères de performances vibro-acoustiques est maintenant rendue possible dès la phase déterminante d'allocation de volumes enveloppes des sous-systèmes, phase nécessaire en ingénierie concourante pour le déploiement d'une architecture automobile. Nous montrons que les concepteurs ont retiré des informations utiles de l'exploitation de ce modèle conceptuel.

To develop methods of predicting radiated noise that can be used in design offices, it is essential to formulate hypotheses to simplify the general problem. The expression of the acoustic presssure autospectrum for quasiplane surfaces in far field is used, obtained from the Green’s function, which depends on: the Green’s function, the cross-spectral density function of the normal acceleration on the body. The spatial distribution of the vibrational field on the vibrating body is, in an industrial context, very difficult to obtain, as it would require too many measurement points. It is therefore necessary to approximate it, and to this end it is assumed that: the vibrating body can be approximated by a combination of elementary surfaces; these surfaces are thin plates; the energy of each mode is uniformly distributed in a given frequency band. Once the vibrational field has been characterized in this way, it is possible to deduce the acoustic radiated pressure. A number of validation tests were carried out, which demonstrated close agreement between the computed and experimental data in the two following cases: a baffle plate and a box.