Artificial Intelligence (AI) for Manufacturing Workshop

View/Download Agenda (PDF)

Sessions Description

Agentic AI for Manufacturing:
As manufacturing shifts from traditional automation to autonomous intelligence, "Agentic AI" has emerged as a critical frontier for engineering and operations. Unlike static models, AI agents can reason, plan, and act independently to solve complex shop-floor disruptions. However, wide-scale adoption remains stalled by concerns over reliability, safety, and a lack of standardized performance metrics.

This 90-minute session brings together industry leaders to present real-world agentic use cases—ranging from autonomous process optimization to intelligent supply chain orchestration. Participants will engage in a collaborative deep dive to identify technical and cultural barriers to adoption, specifically focusing on the Measurement Science required to define agentic system performance and to validate agentic decisions.

By the end of this session, we aim to define a roadmap for the metrics, benchmarks, and verification protocols necessary to build trust in autonomous industrial agents.

Industrial Foundation Models: Data and Application Challenges
The emerging capabilities of large foundation models have sparked interest in leveraging them for transforming manufacturing systems. Trained on various data modalities, industrial foundation models (IFM) can be adapted to improve or refine the execution of a wide range of industrial operations. Exploration into the industrial use of foundation models requires identifying key enabling technologies, such as data pipeline integration and agent communication protocols; potential use cases in manufacturing product design, operations, and maintenance; and the data-level and model-level challenges of integrating the industrial foundation model into the manufacturing system.

This 90-minute session brings together experts across industry and academia to discuss the capabilities and technologies of IFMs, as well as the role these tools can play in improving industrial operations and manufacturing applications.

Participants will engage in a deep dive discussion of challenges to IFM integration, best practices for measuring the engineering impact of IFM integration, and real-world use cases.

By the end of this session, we aim to identify key technology transfer challenges among academic, industrial, and governmental organizations, leading to a roadmap from cutting-edge IFM research to industry integration.

Physical AI in Manufacturing
This tract invites submissions on the latest advancements and applications of Artificial Intelligence (AI) in manufacturing as it interacts with physical processes, equipment, and assets. During this time we will explore the integration of AI with machining equipment and related devices for planning, control, monitoring, digital twins, and the critical aspects of user trust and reliance on these systems. Topics of interest include, but are not limited to: the development and validation of AI-driven operations, maintenance, or control strategies; the assessment of trust, safety, or trustworthiness of AI systems in manufacturing environments; and the measurement and evaluation of the impact of AI on physical processes and assets. We welcome contributions that address the challenges and opportunities arising from the increasing deployment of AI in manufacturing, and that provide insights into the development of reliable, efficient, and trustworthy AI systems that can effectively interact with and augment physical assets.

Human-AI Teaming in Manufacturing
Human+AI Teaming is creating a good deal of excitement in the manufacturing world, with the promise of better workflows and productivity on the horizon. How this will look, in practice, is still taking shape. This session will bring together experts in team cognition, human-machine collaborative design, manufacturing operations, and the impacts of AI use on the workplace to bring us up-to-date on challenges and possibilities in the field.

Through case studies and thought experiments that touch process optimization, system design, error quantification, and organizational practices, panelists will guide our audience through the key frameworks and pain points of this field. Participants can expect a candid glimpse at what Human+AI teaming can get right, where things can easily go awry, and what things we don't know how to measure yet.

AI for Manufacturing Standards
Artificial intelligence is rapidly reshaping manufacturing — from generative design and foundation model-driven optimization to agentic and multi-agent systems that autonomously coordinate production workflows — yet the standards infrastructure supporting reliable and trustworthy deployment of these technologies remains fragmented and incomplete. This session brings together manufacturing practitioners, automation engineers, AI developers, and standards experts to systematically identify the most pressing measurement science and standards gaps for AI in manufacturing. Building on a landscape analysis of existing horizontal AI standards (ISO/IEC JTC1/SC42, ITU) and domain-specific standards under development in ISO TC184, IEC TC65, and IEEE, participants will assess where existing standards can be extended, where new standards are urgently needed, and where foundational measurement science must be established before standardization can proceed. Discussions are organized around six thematic clusters: generative AI and foundation models; agentic AI behavior and autonomy; multi-agent system coordination and safety; physical AI and digital twin integration; data infrastructure and interoperability; and cross-SDO roadmap prioritization. The session will produce a consolidated set of prioritized recommendations — organized by gap, proposed action, and suggested SDO home — to directly inform a forthcoming NIST Advanced Manufacturing Series report.

Human-Machine Teaming Standards
This session examines the future of Human-Machine Teaming (HMT) in manufacturing as a forward-looking conversation about what teaming between human manufacturing experts and machine agents could enable over the next five years. The session opens with a short presentation framing HMT in manufacturing as providing human-facing agentic AI able to collaborate in long-running scenarios. Manufacturing digital twins, organizational learning, and the use of domain-specific languages are among the areas where the relationship between such AI and production experts is beginning to take shape. A panel discussion follows, drawing on participants' experience to identify where the opportunities are greatest and what the standards community needs to do — and to avoid doing — to support rather than constrain the field as it matures.