Journal of Mechanical Design

companion website

FEATURED ARTICLES

MACHINE LEARNING ALGORITHMS FOR RECOMMENDING DESIGN METHODS

11/7/2014 Authors: Mark Fuge, Bud Peters and Alice Agogino 
J. Mech. Des. 136(10), 101103 (2014) (8 pages)doi: 10.1115/1.4028102

Designers use specific methods to discern people’s needs and how to best create products or services that meet those needs. Choosing precisely the right method for a given problem is extremely difficult: it requires a deep understanding of the nature of the problem, knowledge of the vast array of design methods, and years of experience. This paper demonstrates that by collecting expert experience in the form of case studies, machine learning algorithms can help new designers pick better design methods and understand how methods are related to one another. Specifically, we show that looking at which methods designers use together can be more informative than just looking at the content of the method itself. In addition, you can use counts of which methods are used together to automatically cluster methods into groups that agree with human ratings; this means that you can study many more methods than could be done manually.

MACHINE LEARNING ALGORITHMS FOR RECOMMENDING DESIGN METHODS

For the Abstract and Full Article see ASME’s Digital Collection

SHARE: 

Featured Articles Subjects

Additive Manufacturing
Ancient design
Artificial Intelligence
Associate Editors
Awards
Bioinspired Design
Complex Engineered Systems
Compliant Mechanisms
Composites
Data Driven Design
Data Mining
Data-driven Design
Design Automation
Design Communities
Design Education
Design Fixation
Design Innovation
Design of Mechanisms and Robotic Systems
Design Optimization
Design Research
Design Theory
Design Theory And Methodology
Digital Twin
Direct Contact Mechanisms
Double-Blind Review Option
Dynamics
Editors' Choice Award
Energy
Engineered Materials And Structures
Ethics
Fluids
Gears
Generative Design
Guest Editorials
IDETC
Industry
Information Design
International Perspectives
JMD History
JMD Review Process
JMD Statistics
Kinematics
Leadership
Machine Learning
Manufacturing
Mechanisms
Mechanisms Robotics
Memoriam
Neural Networks
Optimization
Origami
Orthotics
Piezoelectric Actuators
Power transmission gearing
Product Development
Robotics
Simulation-based Design
Smart Structures
Special Issues
Sustainable Design