Revolutionizing Industrial Scale Materials Processing
DARPA is seeking information from industrial-scale raw materials producers and likely future performers for a program to use physics-informed modeling and material informatics to reduce variability and improve yield in large-scale metals, ceramics, and composites processing.
⚑ RFI / special notice only; no award is being made in this notice · Future performers are expected to be industrial producers and may include companies and universities · Likely requires access to proprietary multi-year industrial processing data · Federal contract context rather than a grant announcement
This is an industrial materials-processing and physics-informed engineering RFI centered on manufacturing data, process control, and materials science. It does not match IPPRA’s social-science portfolio in risk perception, warning communication, policy, or public response research.
JUDGED AGAINST THE ROSTER'S PUBLICATION DOSSIER · GPT-5.4-MINI · 2026-07-07
Unit fits — one characterization, each unit's own rules
| Physical Sciences & Engineering (demo) | 90 strong | technical depth: central; funds applied research |
| IPPRA | 40 partial | outside portfolio topics; social/behavioral work is none; funds applied research |
| Tom Love Innovation Hub | 30 weak | funds applied research; deep-tech content |
Description
BACKGROUND DARPA is seeking engagement from industrial producers of raw materials including cast, rolled, and forged metals, bulk ceramics, and composites to aid in developing a future program. This RFI is meant to provide the reasoning behind our need for engagement and to present the potential value proposition. Included is a simple survey to gather information on willingness to participate in a future effort. DARPA is considering a program to bring together performer teams composed of the best material scientists, process engineers, and data scientists to revolutionize industrial scale materials processing. Currently, process uncertainty forces system designers to rely on minimum expected property values (e.g., Metallic Materials Properties Development and Standardization (MMPDS)) to guarantee reliability. As a result, even though most processed materials exceed these data sheet minimums, designers must either sacrifice potential system performance margins or use unnecessarily expensive materials to account for statistical variability. Furthermore, this same process uncertainty inevitably yields some products that fall below minimum specifications, reducing overall yield and forcing producers into costly re-processing or scrapping of materials. DARPA believes recent advances in material informatics and physical modeling can demystify the "art" of large-scale processing. While modern industry excels at minimizing variability to meet current baselines, there is significant room for growth; for example, many 7000 series aluminum alloys have a potential performance margin exceeding 15% at fixed plate thicknesses and heat treatments. The challenge lies in the complex, interconnected nature of production variables—such as slight fluctuations in composition, heat treatment, rolling or forging temperatures, and environmental conditions. Because these factors are highly interdependent, optimizing one step does not guarantee a better final product if a subsequent step introduces new variations. Currently, navigating this complexity relies heavily on the intuitive, tacit knowledge that operators develop over decades—an approach that is virtually impossible to replicate or scale. DARPA aims to develop quantitative, physics-informed techniques to uncover the exact causes of product variability. This approach allows producers to fix inefficiencies using their current processes, avoiding the need for expensive capital investments or major facility overhauls. As a result, producers can increase yield and offer superior products at standard market prices, easily outcompeting lower-quality alternatives. To make this work, the program needs access to multi-year processing data from large-scale producers. DARPA completely understands that this data is sensitive and vital to your competitive advantage. Therefore, while DARPA is a government agency, the teams executing this program will likely include private companies and universities equipped to manage and protect proprietary information.
Eligibility
Applicant restrictions (federal contract). Set-aside: . Notice type: Special Notice. Organization: DEPT OF DEFENSE / DEFENSE ADVANCED RESEARCH PROJECTS AGENCY (DARPA) / DEF ADVANCED RESEARCH PROJECTS AGCY.
Apply
Proposal brief SEE AN EXAMPLE →
A one-page internal memo: fit assessment, submission requirements, document scaffold, and next steps dated back from the deadline — tailored to your project idea if you add one.
Proposal shell · Department of Defense (BAA-style) conventions SEE A DOD EXAMPLE →
Funder-faithful document skeletons — Department of Defense (BAA-style)'s document set with section headings, page limits, reviewer guidance, and writing prompts; add a project idea to get [DRAFT] starter bullets. Download as .md for Word or Overleaf.