Data Scientist Senior Staff, Associate Fellow Lockheed Martin
Title: Deep research agentic AI framework for root cause corrective action
Summary: An Agentic AI framework automates deep‑research RCCA by using autonomous reasoning agents to query data, test hypotheses, synthesize evidence, and generate prioritized corrective action plans aligned with aerospace standards, dramatically shortening RCCA cycle times. The presentation covers the end‑to‑end workflow, continuous‑feedback improvement, and case‑study results demonstrating integration and measurable benefits in A&D manufacturing.
Full
Description: Effective RCCA (Root Cause Corrective Action) ensures that defects and failures are identified, understood, and eliminated quickly, preserving the stringent safety, reliability, and mission critical performance standards essential to aerospace and defense systems. However, development of effective RCCA plans can be time-intensive and complex, even for the most senior quality control personnel. This presentation introduces an Agentic Artificial Intelligence (Agentic AI) framework designed to automate deep research RCCA and generate data driven corrective action plans (CAPs) at scale. Agentic AI integrates large models with autonomous reasoning agents that can (1) query structured and unstructured data sources, (2) formulate and execute investigative hypotheses, (3) synthesize evidence across industry best practices and company lessons learned, and (4) propose prioritized CAPs aligned with industry standards (e.g., AS9100, MIL STD 1629). Key contributions include: 1. End to end workflow that transitions from raw manufacturing data to actionable CAPs without manual scripting. 2. Feedback-based continuous improvement. 3. Case study results from showing a significant reduction in RCCA cycle time. Attendees will gain insight into the architecture of Agentic AI for RCCA, practical considerations for integration into existing A&D manufacturing ecosystems, and the measurable benefits of augmenting human expertise with autonomous, data centric research agents. Keywords: Agentic AI, root cause analysis, corrective action plan, aerospace manufacturing, defense, autonomous agents, data driven quality improvement.
