HHS · no agency ID · record R0837

CDER Regulatory Science Research (RSR) Projects AI for Process Control in Advanced Manufacturing

Department of Health and Human Services · HHS/FDA/CDER

Overview

Development stage: DeployedSource: c) Deployed – The use case is being actively authorized or utilized to support the functions or mission of an agency.
High-impact designation: Presumed high-impact, determined notSource: b) Presumed high-impact, but determined not high impact
Impact justification: This project explores the use of AI in advanced pharmaceutical manufacturing as part of an exploratory R&D effort focused on model predictive control strategies. The AI components are used solely in a research context to improve understanding of AI-enabled control systems and inform future regulatory readiness. The project does not involve operational use, decision-making, or direct impact on the public or regulated entities. Therefore, it does not meet the definition of a high-impact AI use case under OMB M-25-21.
Start date: 2024-07 (month precision)Source: Jul-24
Withheld from public reporting: NoSource: a) No

Mission

Topic area: Science
Operational functions: No function tag matched Derived by keyword rules; see methodology

Problem the AI is intended to solve

Need for better process control in continuous manufacturing and development of soft sensors for real-time release testing strategies.

Expected benefits

The outcomes of this work can be used to gain a better understanding of AI in advanced pharmaceutical manufacturing control, identify the associated risks, and help review future submissions involving this technology, ultimately supporting more efficient and reliable drug manufacturing.

System outputs

The AI model demonstrated remarkable performance in setpoint tracking and disturbance rejection for a digital continuous manufacturing line, underscoring the potential of AI-based control strategies in enhancing product quality and regulatory assessment.

Technology

AI classification: Classical / predictive MLSource: Classical/Predictive Machine Learning: Models trained on data to make predictions or classifications based on identified patterns or relatio…
System name(s): Not reported
Custom-developed code: Yes
Public source code: Not reported

Sourcing

How it was built: Developed in-houseSource: b) Developed in-house
Vendor (as reported): Not reported
Vendors (standardized): None on the standard list

A vendor is the supplier named by the agency. It does not identify the underlying model or AI technology.

Data and privacy

Involves PII: No
Privacy Impact Assessment: Not reported
Authorization to Operate: No
Demographic features: k) None of the above
Training and evaluation data: Data was generated using a digital twin of a manufacturing plant developed in-house
Federal Data Catalog: Not reported

Governance

0 of 8 minimum-practice questions answered. A blank answer means the agency reported nothing; it does not mean the practice is absent.

Pre-deployment testingNo answer reported

AI impact assessmentNo answer reported

Independent reviewNo answer reported

Ongoing monitoringNo answer reported

Operator trainingNo answer reported

Fail-safeNo answer reported

Appeal processNo answer reported

User and public consultationNo answer reported

Potential impacts and how they were identified

Not reported