Êtes-vous un étudiant de l'EPFL à la recherche d'un projet de semestre?

Travaillez avec nous sur des projets en science des données et en visualisation, et déployez votre projet sous forme d'application sur GraphSearch.

Découvrez-en plus sur les applications Graph.

Concept

Iterative design

Iterative design is a design methodology based on a cyclic process of prototyping, testing, analyzing, and refining a product or process. Based on the results of testing the most recent iteration of a design, changes and refinements are made. This process is intended to ultimately improve the quality and functionality of a design. In iterative design, interaction with the designed system is used as a form of research for informing and evolving a project, as successive versions, or iterations of a design are implemented. History Iterative design has long been used in engineering fields. One example is the plan–do–check–act cycle implemented in the 1960s. Most New product development or existing product improvement programs have a checking loop which is used for iterative purposes. DMAIC uses the Six Sigma framework and has such a checking function. Object-Oriented Programming Iterative design is connected with the practice of object-oriented programming, and

À propos de ce résultat

Cette page est générée automatiquement et peut contenir des informations qui ne sont pas correctes, complètes, à jour ou pertinentes par rapport à votre recherche. Il en va de même pour toutes les autres pages de ce site. Veillez à vérifier les informations auprès des sources officielles de l'EPFL.

Publications associées

Chargement

Personnes associées

Chargement

Unités associées

Chargement

Concepts associés

Chargement

Cours associés

Chargement

Séances de cours associées

Chargement

Résumé

Source officielle

À propos de ce résultat

Publications associées (11)

Publications associées

Chargement

Personnes associées

Chargement

Unités associées

Chargement

Concepts associés (1)

Concepts associés

Chargement

Cours associés (11)

Cours associés

Chargement

Séances de cours associées (10)

Séances de cours associées

Chargement

Safe non-smooth black-box optimization with application to policy search

Andreas Krause, Ilnura Usmanova

For safety-critical black-box optimization tasks, observations of the constraints and the objective are often noisy and available only for the feasible points. We propose an approach based on log barriers to find a local solution of a non-convex non-smooth black-box optimization problem minf0(x)minf0(x)\min f^0(x) subject to fi(x)≤0,i=1,…,mfi(x)≤0,i=1,…,mf^i(x)\leq 0, i = 1,\ldots, m, at the same time, guaranteeing constraint satisfaction while learning with high probability. Our proposed algorithm exploits noisy observations to iteratively improve on an initial safe point until convergence. We derive the convergence rate and prove safety of our algorithm. We demonstrate its performance in an application to an iterative control design problem.

PMLR2020

Chargement

Human- and machine-centred designs of molecules and materials for sustainability and decarbonization

Sokseiha Muy, Yang Yu

Data-driven approaches based on high-throughput capabilities and machine learning hold promise in revolutionizing human-centred materials discovery for sustainability and decarbonization. This Review examines the strengths and limitations of different traditional and emerging approaches to demonstrate their inherent connection and highlight the evolving paradigms of materials design. Breakthroughs in molecular and materials discovery require meaningful outliers to be identified in existing trends. As knowledge accumulates, the inherent bias of human intuition makes it harder to elucidate increasingly opaque chemical and physical principles. Moreover, given the limited manual and intellectual throughput of investigators, these principles cannot be efficiently applied to design new materials across a vast chemical space. Many data-driven approaches, following advances in high-throughput capabilities and machine learning, have tackled these limitations. In this Review, we compare traditional, human-centred methods with state-of-the-art, data-driven approaches to molecular and materials discovery. We first introduce the limitations of human-centred Edisonian, model-system and descriptor-based approaches. We then discuss how data-driven approaches can address these limitations by promoting throughput, reducing cognitive overload and biases, and establishing atomistic understanding that is transferable across a broad chemical space. We examine how high-throughput capabilities can be combined with active learning and inverse design to efficiently optimize materials out of millions or an intractable number of candidates. Lastly, we pinpoint challenges to accelerate future workflows and ultimately enable self-driving platforms, which automate and streamline the optimization of molecules and materials in iterative cycles.

NATURE PORTFOLIO2022

Chargement

Development of a collaborative conceptual desing support system

Duck Young Kim

Most of the life cycle cost of a product can be pre-determined by the end of the conceptual design phase, which begins with the establishment of a function structure by decomposing an overall function into a set of sub-functions. However, today's products have become complex and accordingly entail intricate function structures that consist of a number of sub-functions; each of them is elaborated by its own design principles. Because of the existence of many choices for design principles of a sub-function, many combinations are possible to satisfy the requirements of the overall function. It is therefore necessary to narrow down the large set of combinations to a manageable size while minimizing the loss of compatible (feasible and preferable) combinations in order to prepare a set of design concept alternatives for the design concept selection phase. Furthermore, since the design concept elaboration process requires a large body of intensive and multidisciplinary knowledge, it is practically performed through the collaboration of different design teams having their own tastes, their own individuality and different design identities about a design; in other words, design teams design their subfunctions according to their own local criteria even though they take into account the overall functionality and global criteria. Another difficulty during the conceptual design stage is that information about the design problem is limited and imperfect due to incompleteness, vagueness and uncertainty while the impact of decisions is so crucial. Considering the above characteristics and complexities of collaborative conceptual design problems, this thesis first introduces three heuristics: (i) pair-wise comparison, (ii) unified domain approach and (iii) domain set ordering heuristic to extract the set of compatible combinations of design principles from all possible combinations considering global design constraints and each design team's local criteria. Subsequently this thesis presents a robust decision support method to select design concepts taking into account multiple participants and their global criteria in order to minimize the need for critical refinement and serious modifications of the design at the subsequent design phase. Finally, the functionality of a COllaborative COnceptual DEsign support system (CO2DE) based on the proposed methods is demonstrated with a case study of a robotic vacuum cleaner design. CO2DE can speed up the overall product design process by reducing design iterations and help the company gain competitive advantage in the long run.

EPFL2006

Chargement

MGT-409: D. Thinking: real problems, human-focused solutions

This course aims to engage students into multidisciplinary collaboration to tackle real world problems with a human centered approach. It will also provide knowledge about the role of design in innovation.

CS-489: Experience design

As we move towards a design economy, the success of new products, systems and services depend increasingly on the excellence of personal experience. This course introduces students to the notion and practice of experience design following a hands-on, studio-based approach.

BIO-695: Image Processing for Life Science

Registration details will be announced via email. It takes place from September to December & intends to teach image processing with a strong emphasis of applications in life sciences. The idea is to enable the participants to solve image processing questions via workflows independently.

Prototype

vignette|Photographie d'un prototype BMW : la BMW Nazca C2 Dans le domaine de l'industrie et plus généralement de la recherche et développement (R&D), un prototype est selon la définition de l'OCDE ,