Recent technological advancements in hardware and software have transformed the design landscape of complex engineered systems such as heterogeneous materials, advanced manufacturing processes, biomedical devices, autonomous and intelligent systems, and aerospace structures. In this transformed landscape, uncertainty quantification (UQ) plays a significant role that begins at the conceptualization and ideation stages and lasts through the design evaluation-update cycle. UQ techniques leverage data, statistical techniques, numerical algorithms, domain knowledge, and machine learning (ML) to characterize the effects of a wide range of uncertainty sources, such as lack of data, noise, and model-form errors, on the design process. Building on UQ, design under uncertainty aims to optimize the design to meet specific probabilistic constraints, including those related to reliability and robustness.
While both the inner-loop UQ and the outer-loop design under uncertainty have been successfully used in a wide range of socioeconomically important applications, new algorithmic and theoretical developments are needed to address (1) the rapid pace of developments in the broad field of ML, such as emerging deep learning and generative AI techniques, (2) engineers’ increasing access to various datasets with different levels of fidelity and modality, (3) the need for understanding a system’s state and performance requirements under dynamically evolving external conditions, and (4) the industrial demands to design products with unparalleled properties that reliably perform under extreme operating conditions. Motivated by these observations, this special issue aims to bring together fundamental contributions to UQ and design optimization techniques that support the design of complex engineering systems under uncertainty.
In the context of this special issue, complexity refers to features such as high dimensionality of the design space or performance metrics, multi-scale or multi-physics nature of the system, multi-modality and/or multi-fidelity characteristics of datasets, evolving states and external conditions, or co-existence of multiple uncertainty sources. We expect the primary novelty of articles in this special issue to be either on the application side where domain-specific uncertainties challenge technological advancements or on the creation of new UQ and design under uncertainty methods founded on novel ML or data-driven techniques. We also welcome contributions that critically (1) compare state-of-the-art methods on emerging design applications or (2) examine the knowledge gaps in the literature and envision future research directions.
Special topics of particular interest include but are not limited to:
- Interpretable uncertainty quantification for decision making and design under co-existing uncertainties;
- Scalable uncertainty quantification for large scale engineering systems;
- Development of physics-based uncertainty modeling methods;
- Uncertainty quantification and/or propagation in multiscale and/or multi-physics engineering systems;
- Uncertainty reduction and probing via contemporary AI- and data-driven tools such as large language models;
- Uncertainty quantification in coupled multidisciplinary systems using novel ML or data-driven techniques;
- Reliability-based and/or robust design optimization of large-scale engineering systems under uncertainty;
- Applications of novel UQ methods in the design of new materials, advanced manufacturing processes, autonomous and intelligent systems, biomedical devices, and similar complex systems; and
- Uncertainty quantification of evolving cyber-physical-social systems and digital twins.
Papers should be submitted electronically to the journal at journaltool.asme.org. If you already have an account, log in as author to your ASME account. If you do not have an account, sign up for an account. In either case, at the Paper Submittal page, select the ASME Journal of Mechanical Design and then select the special issue Data-Driven Design under Uncertainty
Special Issue Timeline:
Paper Submission Deadline: Jan 15, 2025
Initial Reviews Completed: March 15, 2025
Publication: September 2025 (in print); immediate on-line publication upon acceptance
Guest Editors
Ramin Bostanabad, University of California, Irvine, Raminb@uci.edu
Kathryn A. Maupin, Sandia National Laboratory, kmaupin@sandia.gov
Subhayan De, Northern Arizona University, Subhayan.de@nau.edu
Audrey Olivier, University of Southern California, audreyol@usc.edu
Anton van-Beek, University College Dublin, anton.vanbeek@ucd.ie
Zhen Hu, University of Michigan-Dearborn, zhennhu@umich.edu
Leifur Thor Leifsson, Purdue University, leifur@purdue.edu
Wei Chen, Northwestern University, weichen@northwestern.edu
