Mining data to design value: A demonstrator in early design
Year: 2017
Editor: Anja Maier, Stanko Škec, Harrison Kim, Michael Kokkolaras, Josef Oehmen, Georges Fadel, Filippo Salustri, Mike Van der Loos
Author: Bertoni, Alessandro; Larsson, Tobias; Larsson, Jonas; Elfsberg, Jenny
Series: ICED
Institution: 1: Blekinge Institute of Technology, Sweden; 2: Volvo Construction Equipment, Sweden
Section: Design Theory and Research Methodology
Page(s): 021-029
ISBN: 978-1-904670-95-7
ISSN: 2220-4342
Abstract
bility of data mining algorithms as decision support in early design stages of a complex product development project. The paper describes a scenario built in two-stages providing the rationale for the application of data science in engineering design. Furthermore, it describes a demonstrator where usage data are fed back to the early design stage and used to populate value models to reduce the uncertainty in engineering design decision making. The development of a new machine for construction equipment, a wheel loader, is the subject of the demonstration and machine learning algorithms are applied on a dataset built on machine performances and contextual and environmental data. The demonstrator allows the estimation of the fuel consumption of different design concepts and the analysis of the performance variations given by a change in a contextual or environmental variable. Finally, the demonstrator allows the visualization of how much the tested performances of a new design deviate from the original designers’ expectations.
Keywords: Early design phases, Case study, Data mining, Value Driven Design, Decision making