Welcome to ResearchWorks at the University of Washington

First time adding to ResearchWorks? Please contact us.

ResearchWorks is the University of Washington's digital repository (also known as "institutional repository") for disseminating scholarly work.

Looking to deposit a dataset? UW has a membership with the generalist data repository Dryad. Use our guide to see if Dryad is the right place for you to deposit your data.

About ResearchWorks UW Bothell UW Tacoma UW Press

Communities in ResearchWorks

Select a community to browse its collections.

Now showing 1 - 7 of 7

Recently Added

Tools for Coordination: A multi-state chronic wasting disease human dimensions data collection guide
(2026-08-25) Callahan, Megan; McInturff, Alex; DeVivo, Melia
An Integrated Process–Failure Simulation Framework for Predicting Composite Allowables via Multi-Fidelity Stochastic Simulation and Machine Learning
(2026-09-01) Eskandariyun, Amirali; Fu, HuilongNavid Zobeiry; Zobeiry, Navid
For material screening, qualification, and certification of aerospace composites, it is essential to establish statistical design properties that account for both inherent material variability and process-induced uncertainty. Regulatory standards require the determination of A-basis and B-basis statistical values as conservative lower bound design allowables. At present, allowables are determined through extensive experimental campaigns, which are costly and time-consuming, thereby limiting the exploration of new material systems. In this work, we present a multiscale, multi-fidelity process-failure modeling framework that enables the digital prediction of allowables. The framework integrates thermo-chemical-mechanical process simulations with stochastic multi-fidelity failure simulations to capture the effects of curing-induced properties, residual stresses, and microstructural variability on mechanical performance. Machine learning surrogate models, trained via transfer learning, accelerate uncertainty propagation across scales, while Bayesian inference leverages limited experimental data to quantify uncertain parameters. High-fidelity macroscale failure simulations then predict tensile strength distributions from which A- and B-basis allowables are estimated. Application to Hexcel AS4/8552 cross-ply laminates demonstrates close agreement with test data, with conservative predictions of statistical allowables. By combining stochastic process-failure simulations with machine learning, this framework establishes a predictive pathway for estimating allowables and reducing the extent of costly experiments required during composite qualification and certification.
Sharing Power? The Landscape of Participatory Practices & Grantmaking Among Large U.S. Foundations.
(2021-06-06) Husted, Kelly; Finchum-Mason, Emily; Suarez, David
This report explores the landscape of participatory practices and grantmaking among the largest foundations in the United States (by total assets), guided by four key research questions. What is the scope of direct engagement in participatory practices and grantmaking among large philanthropic foundations in the United States? What is the scope of indirect engagement (through funding) in participatory practices and grantmaking among large philanthropic foundations in the United States? What are the benefits and barriers foundations face in adopting and implementing participatory practices and grantmaking? How do foundations define and measure the outcomes associated with participatory practices and grantmaking?
Understanding Current and Future Skagit Flood Risk Through New Modeling Capabilities
(2026-09-02) Swinomish Indian Tribal Community; Skagit River System Cooperative; Skagit Climate Science Consortium
On December 10, 2025, a Level 3 “Go” evacuation order was issued for 75,000 residents. While the forecast predicting record flood levels did not fully materialize, the event still caused significant harm and revealed a critical need for better predictive tools and a unified community response.
Machine Learning for Reduced-Order Modeling of Composites Processing
(6/29/2021) Kim, Min; Zobeiry, Navid
Machine learning concepts have made their way into industry practice and can offer quick and accurate solutions for various design and in-service applications including process simulation of composites. High-fidelity finite element simulation tools are currently used for thermo-chemical analysis of parts during processing. This research investigates reduced-order modeling of complex composite parts using machine learning methods to speed-up the simulation. At first, process simulations of representative 2-D stringer geometries were conducted using finite element analysis to extract leading and lagging responses. Equivalency of exothermic responses of 2-D simulations to 1-D finite element analysis was established using an in-house developed machine learning framework at the University of Washington (CompML). It is demonstrated that for a given 2-D geometry, a trained NN can identify an equivalent 1-D thermal stack to yield similar exothermic responses. The results and methods developed in this study can significantly reduce the computational cost of finite element simulation by successfully establishing reduced-order models of complex geometries.
Machine Learning-Based Process Simulation Approach for Real-Time Optimization and Active Control of Composites Autoclave Processing
(6/29/2021) Humfeld, Keith D.; Zobeiry, Navid
For manufacturing of composites, several parts may be processed simultaneously in an autoclave or oven. Depending on the equipment design, tool/part geometries, and tool nesting, convective heat transfer Boundary Conditions (BCs) may vary around parts in each load. As a result, temperature histories in some of the parts may not conform to specifications such as limits on maximum part temperature, or part temperature rate. To mitigate risk, in addition to conducting finite element simulations prior to fabrication based on assumed BCs, leading and lagging thermocouples, embedded in parts or placed in proxy locations, are used to monitor temperature history during processing. In this study, a recently developed machine learning framework, CompML (Composites Machine Learning) is used for active control of the autoclave. CompML uses TC data at the start of the autoclave processing for real-time inverse modeling of the thermo-chemical problem, and to identify BCs for all parts in each load. The results are then used for real-time optimization of autoclave cure recipe to the shortest cycle that satisfies specifications in all parts. A successful virtual demonstration of the approach for HEXCEL AS4/8552 parts processed on Invar tools is discussed in the paper.
Investigation of Degradation Effects on Crystallization of Thermoplastic Composites
(4/17/2023) Wynn, Mathew; Zobeiry, Navid
Thermoplastic composites such as PEKK or PEEK reinforced with carbon fibers go through heating and consolidation steps during processing. Upon heating and subsequent cooldown, a semi-crystalline structure nucleates and grows in the molten polymer. However, thermal degradation or partial oxidation of thermoplastics may severely affect this process and impact their mechanical properties as well as chemical resistance to common solvents. This also affects the recyclability of the material, as well as available repair-time or time to bring large-scale parts to melt. This paper presents a novel approach to investigate and quantify degradation effects in thermoplastic composites using a combination of polarizing light microscopy (PLM), Fourier transform infrared (FTIR) spectroscopy, and machine learning (ML) analysis. While PLM is used for in-situ investigation of the effect of degradation on crystallization, FTIR and ML are used for in-vitro analysis of degradation effects on chemical signature of the material. The results can be used to potentially develop robust manufacturing processes to optimize performance while minimizing degradation.
A Factory-Centric Workforce Development Approach for Aerospace Industry
(8/24/2026) Zobeiry, Navid; Seaton, Charles; Salviato, Marco; Chen, Xu; Banerjee, Ashis; Devasia, Santosh; Yang, Jihui; Blom-Schieber, Agnes; Buttrick, James; Pedigo, Samuel
Given the rapid transformation of the aerospace sector in the last decade, teaching practices should prepare engineers to face fast-paced industries that are dealing with exceedingly complex problems. Now more than ever we need engineers who are capable of working and communicating effectively within large and multi-disciplinary groups, considering the introduction of new material systems such as advanced composites in primary structural elements, development of automated processing methods such as Automated Fiber Placement (AFP), and the transition to interconnectivity among production systems, workers, products and customers. Exceedingly we need to train well-rounded and practical-minded engineers and scientists. At the University of Washington, we are dedicated to support the aerospace industry by training the next generation of engineers. A new 16,000 sq. ft. facility, Advanced Composites Center (ACC), will be dedicated for manufacturing of aerospace composite parts. Taking advantage of automation and AFP processing, sensor technologies, autoclaves and other manufacturing equipment, we aim to replicate a small factory within the university environment at the ACC. Partnering with industry, this facility will enable students to gain practical experience working on industry relevant problems within a factory setting. In addition, while working on industrial projects, students will gain experience on business, intellectual property (IP) and project management aspects.
A Machine Learning-Based Portable Inspection Method for Evaluation of Tool Surface Condition and Release Coating in Composites Manufacturing
(5/23/2022) Schoenholz, Caleb; Li, Shuangshan; Bainbridge, Kyle; Huynh, Vy; Gray, Alex; Chen, Xu; Zobeiry, Navid
During composites manufacturing, release coatings are applied on production tools to minimize tool-part friction and adhesive bonding. In addition to facilitating the removal of cured parts, applying release coatings reduces process-induced deformations (PIDs). The aerospace industry typically uses semi-permanent release coatings that undergo physical and chemical changes (i.e., aging) with each processing cycle. Fresh layers are frequently reapplied on top of aged coats to mitigate aging effects. Due to a lack of knowledge on the relationship between processing and aging, reapplications of release coating are often untimely and consequently lead to cost-deficient tool preparation schedules, or excessive PIDs. This paper presents a novel machine-learning (ML) framework to evaluate the condition of release coating using non-destructive and portable Fourier-transform infrared spectroscopy (FTIR) and contact angle goniometry (CAG). ML methods and other numerical tools are used to connect measurements gained from low-precision portable equipment to results obtained from high-precision laboratory instruments. The accuracy and portability of the technique demonstrates potential for scale-up and implementation into an automated industrial process. The research results contributes to understand the aging mechanisms of release coating, improve the efficiency of tool cleaning and preparation, and potentially mitigate PIDs in composites manufacturing.
Investigating the Effects of Release Coating on Tool-Part Interaction and Process-Induced Deformations in Composites Manufacturing
(4/17/2023) Schoenholz, Caleb; Zobeiry, Navid
Although modern-era composites manufacturers possess advanced processing capabilities, several production challenges remain prevalent. One such challenge is mitigating residual stresses and process-induced deformations (PIDs) in composite parts while maintaining a cost-efficient manufacturing workflow. For example, applications and touch-ups of release coatings are labor-intensive process steps and generate high recurring production costs, yet are critical to minimize tool-part interaction and PIDs. One intuitive approach to reduce the frequency of disruptive tool coatings or cleanings may be to apply greater quantities of fresh release coats to a tool surface before completing successive cure cycles. However, the consequential effects of such an approach on tool-part interaction and PIDs are currently undetermined and neglected. This paper first investigates the relationship between release coating quantity and tool surface physicochemical properties using laser microscopy and contact angle goniometry. Then, a novel test fixture installed in a Dynamic Mechanical Analyzer (DMA) is presented and used to quantify tool-part stress developments as a function of fresh release coating quantity applied on a tool surface. Lastly, findings from tool surface characterization and DMA testing were validated by curing long symmetric laminates on tools treated with different release coating quantities in an autoclave and measuring warpages. The results in this paper can be used to expand the current understanding of tool-part interaction and improve the efficiency of tool preparation in composites manufacturing.