Skip to content
Patricia Randall
Patricia Randall

Patricia Randall

Director
Princeton Consultants
Bio

Patricia Randall is a Director at Princeton Consultants, where she leads teams designing and deploying large-scale optimization, simulation, and AI-driven decision support systems. Her work focuses on bridging rigorous mathematical models with the realities of complex operational environments, helping organizations turn advanced analytics into systems that can be trusted, adopted, and scaled.

Patricia specializes in architecting analytics platforms that coordinate multiple models, rules, and data sources to support strategic, tactical, and operational decision-making. She has led the development and deployment of optimization and simulation solutions across transportation, manufacturing, biotech, finance, and consumer products, working closely with client executives, domain experts, and cross-functional technical teams. Her recent work emphasizes modern system design patterns that integrate optimization with AI-enabled orchestration and conversational decision support.

An experienced speaker, Patricia has presented at numerous industry and academic conferences, including INFORMS Analytics+, the INFORMS Business Analytics Conference, Gurobi Days, and the AnyLogic Conference. She is also the author of the cover article of INFORMS Analytics Magazine (“How to Avoid Chaos in the Field by Combining Simulation and Optimization”).

Patricia joined Princeton Consultants in 2007 and has progressed to Director, where she is responsible for technical leadership and client success across complex analytics engagements. She holds a PhD in Industrial Engineering from Clemson University.


From Monolithic Models to Decision Ecosystems: Orchestrating Optimization with AI Precision Agents

Real-world operational decisions rarely align with a single optimization model or dataset. Business user questions often span multiple domains and require coordinated, cross-functional analysis. This presentation introduces AI Precision Agents as a practical architectural pattern for integrating GenAI language models with optimization models and analytical tools to create a cohesive decision ecosystem.

Moving beyond the “GenAI hype”, we will focus on how Precision Agents orchestrate the use of mathematical models and analytical tools to produce concrete and actionable responses to multi-faceted operational questions. Rather than replacing optimization or simulation models, Precision Agents provide a structured way to expose and coordinate them—particularly in conversational or question-driven interfaces. Using real world case studies, we examine design patterns that support transparency, safety, and human oversight in AI-forward decision making. Attendees will leave with a concrete framework for integrating optimization into modern, conversational decision support systems while preserving control and trust.

Essential / Professional / Leadership