ASME Journal Accepting Submissions for a Special Issue on Machine Learning for Engineering Design

The ASME Journal of Mechanical Design is now accepting submissions for a special issue concentrating on machine learning for engineering design. Authors who are interested in having their work included in the special issue, which is slated for publication in November 2019, should submit their papers electronically via the ASME Journals Connect page on ASME.org by Feb. 1, 2019.

Modern machine learning (ML) techniques are transforming many sectors — including the areas of transportation and healthcare — by revealing patterns in data, producing autonomous systems that mimic human abilities and supporting human decision-making. Although engineering design researchers have increasingly used ML techniques to tackle problems ranging from materials design to uncertainty quantification in high-dimensional problems, many questions remain unanswered. These questions include how to best employ ML for new design applications that are not well-supported by current ML practice or tools; how to leverage the unique aspects of engineering design in creating new ML approaches; and how to share benchmark problems or datasets that can gauge ML progress in design.

This proposed special issue of the ASME Journal of Mechanical Design is intended to provide a compilation of fundamental scientific and mathematical contributions addressing those three topics. The guest editors of the special issue are particularly seeking manuscripts that highlight the intersection between engineering design and ML and take a broad view involving multiple design problems.

The special issue is expected to address a variety of subjects including fundamental advances in unifying prior engineering and design knowledge with ML techniques; techniques for understanding and supporting human designers, including computational creativity for engineering design, supporting conceptual design, and blending human or organizational information into ML models; and principled ML-based approaches for computational design support, including leveraging and managing uncertainty, learning from multiple representations of design, transfer learning for cross-domain or cross-physics design problems, and techniques for design using limited data. Other areas to be covered include the challenges associated with using ML models for engineering design, such as calibration and validation of ML-based models, and addressing security, privacy, and cyber resilience/reliability; the implications of ML for engineering design education; and the creation and distribution of testbeds and datasets that can support future research in the area.

The editors for this special issue of the ASME Journal of Mechanical Design are Jitesh H. Panchal, Purdue University, USA, panchal@purdue.edu; Mark Fuge, University of Maryland, USA, fuge@umd.edu; Ying Liu, Cardiff University, UK, liuy81@cardiff.ac.uk; Samy Missoum, University of Arizona, USA, smissoum@email.arizona.edu; and Conrad Tucker, Pennsylvania State University, USA, ctucker4@psu.edu.

Manuscripts should be submitted by Feb. 1, 2019 at https://journaltool.asme.org/home/JournalDescriptions.cfm?JournalID=12&Journal=MD, with a note on the cover page that the paper is intended for the special issue, “Machine Learning for Engineering Design.” Early submission is encouraged. Authors should also email the journal editor, Prof. Wei Chen, at editor@asmejmd.org, to inform her that the paper is intended for the special issue.

For more information on the ASME Journal of Mechanical Design, visit www.asmejmd.org. To learn more about the ASME Journal Program, visit http://asmedigitalcollection.asme.org/journals.aspx.