Search

International audience; The development of product-service innovation projects within the context of a company is not yet supported by clear theories and methodologies. Our objective is to analyze innovation and idea generation for such projects from the fuzzy front end to the selected design concept, assessing their potential to be successfully developed and launched on the market. We present a protocol study, using which data derived from 19 innovation projects of five types and conducted by 86 students are analyzed. Sixty-one variables are observed, thus generating 700 data vectors. Bayesian network learning is used to explore conditional inferences among these variables. We examine conditional probabilities between the innovation process means and the significant results produced for the company, modulated by the influence of contextual variables. A number of surprising findings are drawn about the link between problem setting and problem solving processes, the importance of certain contextual variables, and the potential discrepancies between the apparent and produced results of innovative projects. Conducted analyses imply the need for novel innovation evaluation frameworks.

In a recycling scheme such as the recycling of End of Life Vehicle (ELV) materials, the different stakeholders are expected to configure the operation process and the product distribution to maximize their own profits. However, these individual optimizations in many cases do not correspond to the maximal profit for the whole chain. In addition, the financial autonomy must be carefully studied beforehand and controlled so as to ensure an economically viable recycling chain. This paper proposes a cost-benefit parameterized model for the ELV glazing recycling scheme both for the network structure and the material flow sizing. The model is mapped onto real ground data from French industrial partners, and a linear optimization model is built to simulate the conditions of costs and benefits situations for the upcoming glazing recycling chain in 2015 in France.Copyright © 2012 by ASME

In the New Product Development processes, there are usually multiple players interacting through coupled design activities with conflicting design objectives. We use Set-based design approach and Constraint Satisfaction Problem solving techniques to deal with the multiplayer, multi-objective design problem. In order to observe the states of the players during the design process, we develop satisfaction and progress indicators and combine them into a wellbeing indicator. In this paper we simulate different player behaviors on a two-player, multi-objective engineering design problem of a hollow cylindrical cantilever beam. During the simulation cases, at each simulation iteration automatic players are prioritized regarding their wellbeing states for proposing constraints in order to find a set of consistent solutions that improve their satisfaction states. Therefore during the simulation process, while solution space converges to a single solution, epistemic uncertainty is reduced and players’ satisfaction states are improved in wellbeing equilibrium.

In a country like France, electricity consumption devoted to domestic lighting represents nearly a fifth of the total energy consumption of a building. The use of electric lighting is influenced by several factors such as the building’s structural characteristics, the activities of its occupants, the lighting equipments, and the level of natural light. Designers do take into account, in their energy models, the influence of occupants on the building’s overall energy consumption. However, these models still have some drawbacks regarding the comprehension of real “occupants’ energy behaviors” which play an important role in the discrepancies between predicted and real energy consumptions. The behavioral factors behind occupants’ usage trends of energy are still not thoroughly explored. Therefore, it is assumed that a better comprehension of these behaviors and consumption mechanisms could lead to the identification of technical solutions and energy saving potentials, thus resulting in a more robust building design. The present paper aims to provide an insight into domestic lighting usages. The main objective is to explore the key factors (socio-demographic, economic, technical and behavioral) responsible for the disparities in lighting consumption between one household and another. For this purpose, an experiment is performed concurrently to the proposal of a lighting usage model. A micro level investigation protocol is elaborated and used to conduct in-depth studies on the usage patterns of electric lighting. The survey is conducted on a sample of 8 French households. The methodology for constructing the experimental protocol, its deployment, as well as the results obtained and their analysis are presented in this paper. The need for further qualitative and quantitative studies to better understand the usage trends of electric lighting is discussed.

Vous desirez developper un nouveau service, nous proposons dans cette fiche une methode structuree en sept etapes qui vous permettra de le concevoir pas a pas. Concevoir un service se dit « bâtir une offre de service ». Nous proposons dans cette fiche de le realiser en sept etapes : Definition du client et de ses valeurs Analyse fonctionnelle de l’offre de service Conception du bouquet de services elementaires Conception du processus de realisation Conception de la mise en scene ( front office , back office support de realisation) Conception de l’infrastructure ( back office de production et processus supports) et chiffrage Validation du service

Votre entreprise concoit, produit et vend des produits depuis toujours. Or, vous vous interrogez sur l’interet qu’il y aurait a utiliser votre savoir-faire pour vendre des services, ou tout au moins a introduire plus de services autour de vos produits. Vous vous interrogez en fait sur la pertinence de faire evoluer votre modele economique dominant. Cette fiche a trois objectifs : Connaitre les definitions d’un service versus celle d’un produit et savoir representer un service. Connaitre les differences de nature entre un produit et un service. Connaitre les avantages de passer d’un produit a un service.

International audience; Computation techniques have provided designers with deeper understanding of the market niches that were neglected before. Usage contextual information has been studied in marketing research since last century, however little research in design engineering focuses on it. Therefore, in this paper, we analyzed the relations between usage context information and the design of products. A usage coverage model is established to integrate users and their expected usage scenarios into product family assessment. We map the user's individual capacity together with a given product into the usage context space. The overlapping between required usage and feasible usage can be measured. Based on this mechanism, several usage coverage indices are proposed to assess the compliance of a given product family to the expected set of usage scenarios to be covered. The original method is demonstrated on a scale-based product family of jigsaws in a redesign context. Constraint Programming Technique is applied to solve the physics-based causal-loops that determine usage performances in a set-based design approach. Designers can rely on the results to eliminate redundant units in the family or modify the configuration of each product. The contribution of the paper is to provide an inter-disciplinary point of view to assessing the composition and configuration of a product family design.

International audience; A methodological framework to design, assess and manage food traceability systems (TS) is proposed. The services delivered for the multiple beneficiaries of the TS are listed and featured by a series of high-level performance criteria. We also propose a library of modular technical solutions to guide designers in choosing appropriate traceability solutions. Again, at this technical level, practical performance criteria are provided for daily traceability control. This performance system may be used in a design methodology as well as for auditing a TS. Based on this model, we develop an Information System that we apply to a poultry processing company. Few approaches propose practical ways to design, as well as audit and daily control a food traceability system (TS). Indeed, most of the existing literature is written by practitioners and not based on systematic and repeatable industrial engineering approaches. In this paper, we provide to food operators methodological tools to comply with traceability requirements, designing blocks and performance systems. The proposed model is a set of conceptual and methodological tools to design, assess and manage a food traceability system. Three complementary viewpoints of a TS are addressed. The first one is functional since it focuses only on what TS are expected to do, this is also the domain of expected performances. This stage has resulted in three deliverables: identifying the beneficiaries of a food TS, defining the services provided by this system and building quantifiable criteria to assess these services. The second point of view, which constitutes the subject of the present paper, is technical in the sense that it is dealing with lower-level functions and with processes which constitute the structural solution of a TS. Most of these (technical) functions apply to the information used to totally or partially reconstruct the lifecycle of traced products. In order to have a precise idea of this information, the third perspective of our research, described as informational, aims at building a generic traceability data model. Its purpose is to accurately define and characterize the different data to be taken into account within the framework of a food TS. A computer platform based on this model has been developed and implemented within a French food processing company.

International audience; Usage Context-Based Design (UCBD) is an area of growing interest within the design community. Usage context is the set of scenarios in which a product (or service) is to be used, including the environments in which the product is used, the types of tasks the product performs, and the conditions under which the product is purchased and operates. It is proposed in this work that in choice modeling for usage context-based design, usage context should be a part of the primary descriptors in the definition of a customer profile, in addition to the socio-demographic attributes for modeling customers' heterogeneity. As customers become more technology-savvy and market-educated, current choice modeling methods in engineering design could greatly benefit from exploiting the rich contextual information existing in product usage. In this work, we propose a choice modeling framework for Usage Context-based Design (UCBD) to quantify the impact of usage context on customer choices. We start with defining a taxonomy for UCBD. By explicitly modeling usage context's influence on both product performances and customer preferences, a step-by-step choice modeling procedure is proposed to support UCBD. Two case studies, a jigsaw example with stated preference data and a hybrid electric vehicle example with revealed preference data, demonstrate the needs and benefits of incorporating usage context in choice modeling.

Comment faire concretement pour rendre votre entreprise plus eco-innovante ? Six axes de progres en eco-innovation ou conception durable ( cf . fiche sur le sujet) peuvent faire l’objet d’engagements strategiques, de definitions d’indicateurs et d’objectifs annuels. A vous de mettre en œuvre les pistes de travail que vous estimez prioritaires, en fonction de vos objectifs et de l’interet que porte votre direction a la demarche d’eco-innovation. Nous proposons dans cette fiche de travailler a la fois sur differents fronts : preferer une vraie demarche de progres a du « green washing » ; developper la pratique des outils et methodes ; effectuer une veille sur les reglementations et les normes ; se positionner par rapport aux reglementations et normes (reactivite ou pro-activite).

Une enquete recente sur le management de l’innovation dans les grandes entreprises francaises confirme la difficulte des entreprises a mesurer quantitativement la performance de l’innovation. Or, cette mesure est indispensable pour justifier et motiver la rentabilite des investissements consentis en RD efficacite de la transformation (dont le taux de transfert vers le marche) ; taux d’effort Recherche et Developpement et contribution a la valeur ajoutee de l’entreprise ; efficacite des processus Recherche et Developpement (au sens du controle de gestion) ; efficacite du recours a l’ open innovation ; qualite de l’innovation percue par le client. Pour l’ensemble des indicateurs proposes, l’objectif poursuivi est d’identifier dans votre entreprise ceux qui sont utilises de facon reguliere et systematique, ainsi que leur capacite a mesurer reellement ou pas – du point de vue de l’entreprise – la performance de l’innovation.

Le processus de generation des idees, de selection et d’elimination progressive tout au long du cycle de decision et des projets de developpement, est generalement decrit et compare a un entonnoir , que ce soit dans la litterature des projets de developpement et de lancement de produits nouveaux qu’au sein des entreprises. Une enquete recente sur le management de l’innovation dans les grandes entreprises francaises ( cf . bibliographie dans la rubrique « Pour aller + loin ») nous a permis de proposer cinq modeles d'entonnoirs qui ont leurs propres caracteristiques et raisons d’etre. Nous les avons baptises : l’entonnoir « normalise, mais peu fertile » ; l’entonnoir « normalise mais a forte attrition » ; l’entonnoir des « heros » ; l’entonnoir « mixte (ou double dispositif) » ; l’entonnoir « agile et permeable ». Nous proposons que vous utilisiez ce modele comme un outil de reflexion sur votre propre processus d’entreprise. A vous de trouver le modele d’entonnoir vous correspondant et de tirer les conclusions adequates. De plus, la fiche « Comment retrouver l'agilite et la creativite des origines » complete cet outil en suggerant qu’une fois qu’une entreprise est a un stade donne, elle tend et cherche generalement a evoluer vers un stade ulterieur d’entonnoir. Vous pourrez alors mieux accompagner, voire accelerer les evolutions probables de votre processus de management des idees.

Comme explique dans la fiche « Comment selectionner des idees et les transformer en innovations effectives ? », une enquete recente sur le management de l’innovation dans les grandes entreprises francaises nous a permis de proposer cinq modeles d'entonnoirs pour representer les differentes dynamiques de gestion d’idees et de projets. Dans les trois premiers, il existe un lien fort entre la selection des idees, l’engagement des projets et le processus budgetaire. A partir de l’observation historique des entreprises, nous avons vu que les differentes formes d’entonnoir s’inscrivent dans un processus naturel et logique d’evolution cyclique en cinq stades successifs presentes dans cette fiche : la start-up, permeable et agile spontanement ; l’entreprise en croissance qui a tendance a standardiser les processus ; la reaction salutaire a la normalisation : celle de heros qui poussent a bout de bras les innovations ; la prise de conscience de la necessite de mettre en place des dispositifs favorisant l’innovation « hors cadre » ; l’organisation de l’agilite.

International audience; The eco-design of complex industrial systems becomes a major issue for the concerned companies. Life Cycle Assessment (LCA) in particular is more adapted to "classical" products such as consumer goods. Performing LCA of such systems requires some thoughts to ensure significant results, for example concerning data granularity and quality. This article proposes a Life Cycle Assessment of an Alstom Grid AC/DC conversion substation for the aluminium industry. This LCA integrates first answers to the previous limitations. Very interesting results permit to orient the eco-design strategy of the company. Moreover they lead to imagine original ways to configure the system. Thus a first and simple parametric LCA model is proposed: four different electrical sources are used to analyze the sensitivity of the design factors to the environmental impacts of the substation. Based on design of experiments, a more advanced model of such a tool would permit to identify the best configuration in terms of environmental performance, costs and reliability.

International audience; A View of Design (and JMD): The French Perspective A quick glance on history of science and engineering shows that the creation and invention of technical products has held a great appeal in France. A tradition of inventiveness, engineering and entrepreneurship has developed over the past several centuries; some famous mechanical and industrial pioneers, for instance, Gustave Eiffel, Louis Renault, André Citroën, André Michelin, Marcel Schlumberger, Louis Blériot, William LeBaron Jenney are alumni of the Ecole Centrale Paris. Technical realizations like the TGV train or Concorde airplane also illustrate this tradition of design. The French research in design is anchored in the mechanical discipline. In the late 1980's, a design community was progressively constituted through the desire to merge mechanics with computer science...

Research highlights? CBN introduces the conditional independence relationships among attribute variables. ? CBN provides an effective approach to classify the failure rate rank of products. ? CBN increases the classification accuracy. ? CBN makes an acceptable balance between classifier complexity and performance. To identify the product failure rate grade under diverse configuration and operation conditions, a new conditional Bayesian networks (CBN) model is brought forward. By indicating the conditional independence relationship between attribute variables given the target variable, this model could provide an effective approach to classify the grade of failure rate. Furthermore, on the basis of the CBN model, the procedure of building product failure rate grade classifier is elaborated with modeling and application. At last, a case study is carried out and the results show that, with comparison to other Bayesian networks classifiers and traditional decision tree C4.5, the CBN model not only increases the total classification accuracy, but also reduces the complexity of network structure.

Models of the engineering design process and of the development process nowadays present similar forms in the engineering design literature and interact in a similar way. These models are often presented as generic, in order to be used in a wide area of applications. This interaction is however not unproblematic, and in this publication we present some important issues and challenge the generic aspect of these models. In order to increase clarity, we have divided the publication into two parts. In Part I, the generally accepted engineering design and development process models are presented. The fundaments of the development model and the motivations behind its current form are highlighted. In Part II, the consequences in the form of severe shortcomings resulting from the interaction of the engineering design and development process models are highlighted. These shortcomings do not disappear when the systematic design process model is applied with alternative development process models. The implications for the further development of methodologies supporting design and development models are discussed.

In this paper, we investigate a special case of new product development (NDP), that of radical innovation. When a company desires to go outside a current saturated market, it is necessary to offer a new product which is radically different from the competition. If successful, the offering will create a new market that the company can expect to dominate a least for a while until the competition strikes back. A radical NPD has several characteristics that are quite unique and distinct from other types of NPD. This paper aims at making a synthesis of these specificities. From this set of specificities, the ability of the current methodologies to deal with radical innovation is discussed.

In the New Product Development processes, there are usually interacting multiple actors with multiple design objectives. Design objectives of the actors can be contradictorily related and design modifications related to some objectives may generate negative impacts on the other actors. Therefore conflicts may occur. We present a multi-actor design platform that is capable of preventing potential conflicts with ensuring the satisfaction and progress of the actors at a certain level. An indicator of wellbeing is developed in order to indentify the satisfaction and progress states of the actors. We use Set-based design approach and Constraint Satisfaction Problem solving techniques to deal with the multi-actor, multi-objective design problem. We apply our collaborative design approach to a multi-actor multiobjective engineering design problem of a hollow cylindrical cantilever beam with a load applied at the unsupported end.Copyright © 2011 by ASME

International audience; The integration of environmental concerns into the product design process has highlighted a new problem that arises when confronted with complex systems. Indeed environmental assessment methodologies like Life Cycle Assessment (LCA) become in this case particularly heavy to implement. Considering aluminium electrolysis substations as a complex industrial system, we propose a new eco-design methodology based on a Lean Six Sigma approach. Including the environmental parameter as the fourth dimension of the Quality, Costs, Time triangle this methodology has the advantage to cover and systematize the entire eco-design process. It answers to most of the limits raised in our study and allows managing a part of the complexity that appears in particular during the goal and scope definition and the inventory phases of LCA. An application of aluminium electrolysis substations is mentioned.

So as to create innovative car silhouettes, we propose in this technical report a model based on an Interactive Genetic Algorithm using an encoding of a design solution by a Fourier analysis approach. This model permits the designer to browse through generations of car profiles from an initial population of existing silhouettes. By qualitatively assessing each individual, the user converge towards solutions complying with his/her requirements and so potentially create novelty. We describe here tests for assessing the efficiency of this innovative design platform. These tests are based on a similarity matrix, a similarity measure being the perceived distance between two cars silhouettes. The results show a really satisfactory behavior of the model and open perspectives thanks to its flexible and extensible aspects.

Usage Context-Based Design (UCBD) is an area of growing interest within the design community. A framework and a step-by-step procedure for implementing consumer choice modeling in UCBD are presented in this work. To implement the proposed approach, methods for common usage identification, data collection, linking performance with usage context, and choice model estimation are developed. For data collection, a method of try-it-out choice experiments is presented. This method is necessary to account for the different choices respondents make conditional on the given usage context, which allows us to examine the influence of product design, customer profile, usage context attributes, and their interactions, on the choice process. Methods of data analysis are used to understand the collected choice data, as well as to understand clusters of similar customers and similar usage contexts. The choice modeling framework, which considers the influence of usage context on both the product performance, choice set and the consumer preferences, is presented as the key element of a quantitative usage context-based design process. In this framework, product performance is modeled as a function of both the product design and the usage context. Additionally, usage context enters into an individual customer’s utility function directly to capture its influence on product preferences. The entire process is illustrated with a case study of the design of a jigsaw.Copyright © 2010 by ASME

International audience; Once a design concept has been chosen and parameterised, the embodiment design stage consists of choosing materials and dimensions to ensure a |good matching| with the expected performances. In this context of preliminary design stages, several approaches exist, which correspond to slightly different complexities and issues and must consequently be used at different moments. We consider in this paper three families of approaches: 1) exploring design (parametric) dimensioning under uncertainty (through constraint programming techniques, representations of feasible design points or Pareto frontiers); 2) robust design and multidisciplinary optimisation; 3) design for reliability. We advocate and state in this paper that these approaches must be used in that order of increasing complexity. Indeed, applying an approach allows one to quickly figure out inadequacies with performance specifications or initial allowable bounds of design parameters and then to backtrack or to refine the design issue before proceeding to the next stage or approach. We illustrate that phenomenon by successively applying the three approaches on a dimensioning issue of a two-member truss structure. We clearly show that the successive optimal designs obtained are notably different, but that the optimal point obtained in a given approach is used to explore its surroundings within the next approach.

special Issue on "Engineering Emotional Design (EED) and Kansei Engineering (KE); International audience; Starting from a primary perceptual evaluation of a set of car dashboards, we propose to build a Bayesian network (BN) between perceptual attributes and design attributes. Two types of learning processes may be considered: supervised BN when the prediction on a targeted attribute must be optimised and unsupervised BN otherwise. These two types of BNs are considered along three design simulation scenarios: the direct scenario which consists of the prediction of a design change impact on customer perceptions, the inverse scenario for fixing design characteristics so as to result in an expected customer perception, and a more realistic combined scenario.

Throughout these years, with the high demand for system safety, operational efficiency and life cycle cost control, decision support systems that balance among such requests for complex system maintenance were brought forward. This paper illustrated a newly developed influence diagram decision model which integrates current states, decision actions, decision objects and cost analysis to facilitate the maintenance strategy decision making. At first, the network structure of the influence diagram is proposed. Then the conditional probabilities and the utility values of each kind of costs were elaborated. At last, the procedure for implementing maintenance decision influence diagram model was demonstrated based on a case study. With the input of current states, this model could provide the recommendations and relevant supportive information for the purpose of cost budgeting. In order to verify the model's reliability, costs from different decisions are listed for comparison.

International audience; Accurately capturing the future demand for a given product is a hard task in today's new product development initiatives. As customers become more market-savvy and markets continue fragment, current demand models could greatly benefit from exploiting the rich contextual information that exists in customers' product usage. As such, we propose a Usage Coverage Model (UCM) as a more thorough means to quantify and capture customer demand by utilizing factors of usage context in order to inform an integrated engineering design and choice modeling approach. We start by presenting the principles of the UCM model: terms, definitions, variable classes and relation classes so as to obtain a common usage language. The usage model exhibits the ability to differentiate between individuals' product performance experiences. With Discrete Choice Analysis, individuals' performance with a given product is compared against that of competitive products, capturing individual customers' choice behavior and thereby creating an effective model of product demand. As a demonstration of our methods, we apply our model in a case study regarding the general task of cutting a wood board with a jigsaw tool. We conclude by presenting the scope of future work for the case study and the contribution of the entire current and future work to the field as a whole.

International audience; So as to create innovative car silhouettes, we propose in this paper a model based on an Interactive Genetic Algorithm using an encoding of a design solution by a Fourier analysis approach. This model permits the designer to browse through generations of car profiles from an initial population of existing silhouettes. By qualitatively assessing each individual, the user converge towards solutions complying with his/her requirements and so potentially create novelty. We describe here tests for assessing the efficiency of this innovative design platform. These tests are based on a similarity matrix, a similarity measure being the perceived distance between two cars silhouettes. The results show a really satisfactory behavior of the model and open perspectives thanks to its flexible and extensible aspects.

With the application of maintenance management system, the daily failure data of airplane components has been collected in great number. The challenge faced currently is how to predict the failure probability of certain components and carry out optimal maintenance strategy (when and how) to reduce the lifecycle cost of the airplane. In this paper, we have presented a proactive maintenance management framework as well as the entire maintenance work flow which guides the establishment of maintenance system in detail. Furthermore, we brought out the procedures of building prognosis Bayesian networks and decision Bayesian networks. The prognosis network which is based on discovered associations from failure data can provide reliable predictions on component failure probability; while the decision network could help to make maintenance decisions balancing among resource availability, repair effect, cost and product risk.

International audience; We propose in this paper a car silhouettes design model based on an Interactive Genetic Algorithm. The design solutions are indirectly encoded by a Fourier analysis approach. Thus after creating an initial population, the designer is able to evaluate himself generations of car profiles, in order to converge towards an individual which meets its requirements. An automatic test has already been conducted, but it is not sufficient to validate the model; we propose here 6 new tests. The two first ones are based on the user satisfaction towards the created design solution. The data were collected during a short subjective evaluation workshop. The 4 other tests use a similarity index (that quantify the perceived distance between 2 individuals) to process the same data and validate the model. These results reveal a really satisfactory behaviour and open perspectives on the understanding of the user perception.

In product platform development the technology selection issue has seldom been studied. This article tries to fill this gap by introducing an original Technology Selection Method (TSM) to help designers to manage technology selection within the design of a product family. This method is based on technological coverage and functional verification, including the behaviour of the future product family. Practically, the proposed approach uses a multi-objective analysis based on dimensional analysis theory to choose the best technology available. The developed approach has three major points of interest. First, dimensional analysis theory is considered as a specific type of multi-objective optimisation approach which aggregates attributes using a weighting method based on the laws of physics. This manner of viewing dimensional analysis theory provides a scientific coherence to the weighting process which is not obtainable with other multi-objective methods. Second, the principle of similarity included in dimensional analysis theory is extended and allows a range of technologies and range of functions to be compared in a unique design space. Third, the dimensional analysis provides a powerful simulating tool for studying different kinds of behaviours and interactions between attributes. Consequently, dynamic aspects can be analysed and robustness analyses can be performed. The interest of such a method is highlighted through a case study involving a family of small excavators. We suggest that the potential scope of the approach is broad and our aim is to demonstrate the entire scope of the approach in future research.

National audience; Face au manque d'outils CAO permettant de prendre en compte le ressenti émotionnel des designers dans les premières phases du processus de conception, nous proposons dans cet article un modèle de conception de profils de voiture par la technique des Algorithmes Génétiques Interactifs. Ce modèle permet notamment d'encoder des solutions de conception (profils de voiture en 2D) par transformation de Fourier, et de manipuler des générations d'individus, dans le but de converger vers une solution de conception remplissant les objectifs fixés. Afin de valider ce modèle, un séminaire d'évaluation subjective a été mené ; dans ce cadre, deux tests ont été réalisés sur un panel d'utilisateurs, et avec deux attributs sémantiques (sympathique et sportif). Ces deux tests montrent un comportement satisfaisant du modèle. Une analyse plus fines des résultats par un indice de similarité (mesurant la différence perçue entre deux profils) confirme ces résultats et fournit un grand nombre de données sur la perception des utilisateurs. De fortes corrélations existent ainsi dans le cadre de profils sportifs, tandis que l'attribut sympathique est très différemment perçu par les utilisateurs.

International audience; Pairwise comparison methods are convenient procedures for providing a sound weight vector from a set of binary comparisons between elements to be rated. In such procedures, each decision maker is asked to separately consider pairs of elements which are not necessarily independent from each other. For this reason the votes collected are liable to contain inconsistencies. In this paper we are providing a selective indicator that focuses on the inconsistencies the decision group is willing to correct in conformity with its vote strategy.

In this paper, we present a project with the world leader of office furniture and workplace solutions. The goal is to build a decision-making tool combining a need model and a solution model to be used by designers and sales people to simulate different planning scenarios having different contributions to company performance. This decision-making tool is based upon a model of need (for a new workplace design project) and a solution need (roughly describing a solution sketch). We describe the ontology of the tool and illustrate with a case study about a decision making process in a limited perimeter of lighting system.

The issue of this paper is the use of perceptual evaluations of products as a simulation platform for improving decision making in the design process of a new product. In previous work, we proposed bayesian kansei models using unsupervised learning of relationship between perceptual evaluation of a family of similar products and technical characteristics of these products. We showed the main advantages of such bayesian models compared to other techniques in that they are not opaque to the designer and they can be easily used in compound analysis/synthesis scenarios. The objective of this paper is to show the limits of such unsupervised bayesian kansei models and to propose the use of complementary supervised bayesian kansei models. We show that these models are more accurate to carry out a local optimization of a design.Copyright © 2007 by ASME

Three families of methods coexist for managing (i.e. representing and propagating) uncertainty of product data during the preliminary design, namely: fuzzy methods, probabilistic methods and Constraint Programming (CP) methods. CP methods over reals are, up to now, the less frequently used approaches, but they are worth further study for use in design engineering thanks to a number of good properties and recent significant advances. They may be roughly considered as a collection of methods that are sophisticated evolutions of interval analysis. The objective of this paper is to assess four of these major methods; namely the {Hull, Box, 3B-weak, 3B}-consistency methods in the context of the preliminary design of mechanical products where large variable domains are considered and a representation of the remaining consistent design space turns out to be of practical interest to support the designers’ understanding and decision making. A measure for comparing the level of consistency of the methods is then proposed in the context of engineering design. It consists of a pairwise comparison of the overlapping part of the remaining design spaces for a given splitting grain size. Numerical results are established for an example of a combustion chamber design with 6 variables and 12 constraints. Next, a sensitivity analysis of the consistency of the previous methods is performed in regards to a variable splitting grain size. Experiments have been performed on a research platform including up-to-date NCSP methods. The paper concludes that NCSP methods are not easy-to-use and that the designer must be aware of a number of concepts so as to select the best choices in the resolution strategies.

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.

National audience

The elementary conceptual design loop is made of understanding and measuring the customer need, proposing product concepts and propagating the customer need along the product deployment, and finally assessing concepts in a multi-criteria way in regards to the need. Many methods from various disciplines (industrial/design engineering, psychophysics, multi-criteria decision analysis, marketing, artificial intelligence, statistical analysis) presently contribute partially to support this design loop. But few methods propose an integrated and coherent framework to deal with this elementary conceptual design loop. In this article, we propose such a methodology which is a coherent combination of classical methods in psychophysics, marketing and decision theory, namely multidimensional scaling, semantic differential method, factor analysis, pairwise comparison and Analytical Hierarchy Process. Our approach provides designers with a tool which helps the definition of the semantic part of the need, it rates and ranks the new product prototypes according to their proximity to the “ideal product”, and it underlines the particular semantic dimensions that could be improved. To illustrate our approach, we have performed users’ tests and applied our methodology to the design of table glasses. For clarity, each stage of the methodology is presented in detail on this particular example.

Design space exploration during conceptual design is an active research field. Most approaches generate a number of feasible design points (complying with the constraints) and apply graphical post-processing to visualize correlations between variables, the Pareto frontier or a preference structure among the design solutions. The generation of feasible design points is often a statistical (Monte Carlo) generation of potential candidates sampled within initial variable domains, followed by a verification of constraint satisfaction, which may become inefficient if the design problem is highly constrained since a majority of candidates that are generated do not belong to the (small) feasible solution space. In this paper, we propose to perform a preliminary analysis with Constraint Programming techniques that are based on interval arithmetic to dramatically prune the solution space before using statistical (Monte Carlo) methods to generate candidates in the design space. This method requires that the constraints are expressed in an analytical form. A case study involving truss design under uncertainty is presented to demonstrate that the computation time for generating a given number of feasible design points is greatly improved using the proposed method. The integration of both techniques provides a flexible mechanism to take successive design refinements into account within a dynamic process of design under uncertainty.Copyright © 2005 by ASME

International audience; As teachers and researchers belonging to two sister French engineering schools, we are convinced that the processes of value creation and management are essential in today's teaching of industrial engineering and project managers. We believe that such processes may be embedded in a three-part curriculum composed of value management and innovation management courses linked to the practice of group projects so as actually to acquire value and innovation spirit, reflexes and professional skills. After cogitation about the need for such knowledge and skills and a listing of an ideal value and innovation curriculum, we present our solution for a short and efficient training on value creation and management in engineering.

Uncertainty is inherent to product design since a design process starts with needs defined semantically and ends with precise product and process technical specifications. Managing uncertainty of product and process data is essential to conceptual design since data are still imprecise and to concurrent engineering since it guaranties data consistency at any time and an optimal paralleling of development and production activities [1–2]. After noting the poverty of tools and techniques in conceptual design of mechanical systems, the design process is presented as a process of topological and dimensional variability narrowing. Notably, a propagation of the reduction of design variable domains must occur towards the performance domains as soon as possible and conversely, so as to respect the principles of data consistency and simultaneous engineering. Then, one briefly shows that probabilistic simulations and fuzzy inference techniques can respond in a static way to the sole management of the dimensional variability. A state-of-the-art of constraint programming (CP) techniques especially over continuous domains is performed. One believes that these latter techniques are the most promising ones, in the long term, for supporting a breakthrough to conceptual design and concurrent engineering practices.

The elementary conceptual design loop is based upon understanding and measuring the customer’s need, proposing concepts of products and propagating the customer’s need along the product deployment, and finally assessing concepts in a multi-criteria way with regards to the need. Many methods from various disciplines (industrial/design engineering, psychophysics, multi-criteria decision analysis, marketing, artificial intelligence, statistical analysis) presently partially contribute to support this design loop. But few methods propose an integrated and consistent framework to deal with this elementary conceptual design loop. In this article, we propose such a methodology which is a consistent combination of classical methods in psychophysics, marketing and decision theory, namely multidimensional scaling, semantic differential method, factor analysis, pairwise comparison and Analytical Hierarchy Process. Our approach provides designers with a tool which helps them define the semantic part of the need, it rates and ranks the new prototypes according to their proximity to the “ideal product”, and it stresses the particular semantic dimensions that could be improved. To illustrate our approach, we have performed users’ tests and applied our methodology to the design of table glasses. For clarity, each stage of the methodology is presented in details on this particular example.Copyright © 2004 by ASME

Constraint Programming (CP) is a promising technique for managing uncertainty in conceptual design. It provides efficient algorithms for reducing, as quickly as possible, the domains of the design and performance variables while complying to the engineering and performance constraints linking them. In addition, CP techniques are suitable to graphically represent 3D projections of the complete design space. This is a useful capability for a better understanding of the product concept’s degrees of freedom and a valuable alternative to optimization based upon the construction of an arbitrary preference aggregation function. Unfortunately, one of the main impediments for using Constraint Programming on industrial problems of practical interest is that constraints must be represented by analytical equations, which is not the case of hard mechanical performances — such as meshing and finite element computations — that are usually obtained after lengthy simulations. We propose to use metamodeling techniques (MM) to generate approximated mathematical models of these analyses which can be employed directly within a CP environment, expanding the scope of CP to applications that previously could not be solved by CP due to the unavailability of analytical equations. We show that there is a tradeoff between the metamodel fidelity and the resulting CP constraint tractability. A strategy to find this compromise is presented. The case study of a combustion chamber design shows amazingly that the compromise is to favor the simplest and the coarsest first-order response surface model.

Often, the development of Computer Aided Process Engineering (CAPE) for Manufacturing Engineering Departments (MED) leads to isolated Databases (DB) because of the past vertical orientation of system design (one database — one application). This usually caused a tremendous amount of data redundancy, heterogeneous manufacturing DB and, as a result, a configuration of manufacturing design that was far from optimal. Our objective here is to overcome those difficulties, by proposing a new Information System scheme, based on the construction of two models with interactions in two directions: the “Business Model” is a representation of business processes according to different views of the company: Function, Activity, Milestone... and at different levels of detail: Conceptual, Logical and Physical. The second one consists in modeling the trade objects (common and specific) with their mutual interactions. Eventually, this project should lead to a standardization of process specification (resource and operation stated references) with special emphasis on the data handled during the engineering stage.

The present paper describes an original contribution to handling imprecision in multi-criteria group decision making. Based on Monte Carlo simulation, this approach allows the generalization of any deterministic pairwise comparison method. Furthermore, it extends the practical interest of the pairwise comparison concept to a wider variety of inputs. For instance, it can provide a useful aid in concurrent engineering.

International audience; We propose an original method to synthesize the dimensions of a planar mechanism (linkage) whose function is to generate a trajectory shape. Most graphical and analytical synthesis methods for path generators require specifying the desired trajectory in a non-functional manner, by a list of points rather than a pure shape. Concerning the dimensional optimization methods, they turn out to be slow and their convergence depends on the initial solution. Alternatively, we propose a case-based approach (i.e., couples of trajectories and dimensions of a given structure mechanism) using a neural network. The first stage consists in the generation of a huge case number through kinematic simulations, for random values of dimensions, and in a learning process of the neural network. In the second stage, of utilization, the neural network instantaneously makes it possible to obtain an approximate solution of the synthesis problem, which is an interpolation of close cases. We show on the four-bar linkage example the good quality of the synthesized solutions, for a tiny size of the network. Next, these solutions may be used as judicious initial solutions for a conventional dimensional optimization.

Metamodels are effective for providing fast-running surrogate measures of product or system performance. Instantaneous performance estimation can allow designers to pursue design alternatives interactively, a methodology different from traditional case-based or optimization-based design. Metamodel approximations are generally based on continuous functions, and therefore they can be poor at fitting discontinuous responses. Many engineering models produce functions that are only piecewise continuous, due to changes in modes of behavior or other state variables. This paper shows that a state-selecting metamodel approach can provide an accurate approximation for piecewise continuous response functions. The method is applied to a desk lamp performance model.

In this paper, different approaches are presented in order to automatically estimate the relevance of design performances in terms of aglobaI satisfaction. The first part describes a functional needs expression approach: a bottom-up satisfaction model from lower-Ievel functions to the global satisfaction, taking into account data imprecision and uncertainty at each level. The second part introduces a virtual prototyping design platform, which consists in a primary simulation of some solution performances in order to propagate them through the functional model. Redesign, optimization and risk management capabilities can be imagined, since it is possible to measure the influence of the structural parameter choices over the measure of the global satisfaction and its uncertainty.