From Data to Decision: Transforming T&E with Digital Engineering, Automation, and Artificial Intelligence

Modern defense systems operate in increasingly complex, multi-domain, and Joint environments, often leveraging advanced artificial intelligence (AI). Traditional test and evaluation (T&E) methods struggle to match this complexity or the rapid pace of technological advancement. To ensure rigorous, mission-relevant evaluations, the T&E enterprise must adopt digital transformation, automation, and AI-enabled tools.

This presentation outlines a vision for revolutionizing T&E through digital engineering, automated test design, and advanced statistical methods, such as Bayesian inference, to integrate data from modeling, simulation, developmental testing, operational testing, and training exercises. These tools serve as a universal translator across acquisition phases, enabling portfolio-level insights and aligning evaluations with realistic mission contexts.

The talk will also explore how AI-enabled tools can further enhance analytical rigor. For example, reinforcement learning–based “test planners” can autonomously stress systems within high-fidelity simulations, uncovering edge cases and hidden vulnerabilities that might elude traditional test designs. AI algorithms deployed at test ranges or onboard platforms can detect anomalies in real time and rapidly process large streaming datasets, enabling faster triage and focused follow-up analysis. Additionally, large language models can help automate the drafting of test plans, scenarios, and reports, freeing skilled analysts to concentrate on high-value tasks such as interpreting results and providing mission-focused recommendations.

This talk will provide a vision for how these digital and AI-enabled capabilities provide a new paradigm for the T&E community to “test as we fight,” efficiently integrating diverse data streams, expanding test coverage, and preserving the independence and credibility of operational assessments—even amid growing system complexity and constrained resources.