Keynote Lecture
Using ERP Data to Support Operational Planning and Risk Analysis: An Expeditionary Case Study
Brandon McConnell
NC State University
United States
Brief Bio
Brandon McConnell is a research assistant professor in Industrial & Systems Engineering at North Carolina State University and leads the Military Operations Research Group. He is a former U.S. Army Infantry officer and has held various key leadership positions both in combat and garrison. His interests include expeditionary logistics planning, risk analysis, and stochastic models. His work emphasizes using OR models and analysis to identify tradeoffs, illuminate decision-spaces, and support the operational user in near-real time. He earned his PhD and masters degree in Operations Research from NC State and his B.S. in OR from the United States Military Academy at West Point.
Abstract
The U.S. Army’s adoption of an enterprise resource planning (ERP) system known as the Global Combat Support System – Army (GCSS-A) creates a new opportunity to link operational data to modeling and analytical efforts. Recent advances in queuing theory have opened new opportunities for analytic modeling of complex systems. This permits modeling both time-varying (nonstationary) and non-Markovian (nonexponential) properties across complicated systems and networks. This presentation offers a data-driven approach to forecast logistical requisitions for an expeditionary military operation using GCSS-Army data. Model inputs include task organization, mission set, and operational timeline. The result is a sample-path based approach that can feed multiple modeling techniques; examples will include logistics capacity planning and risk analysis application all in an expeditionary environment. We discuss the benefits and challenges associated with integrating some of these modern advancements with previous deterministic approaches to obtain near-real time stochastic performance predictions while staying faithful to detailed problem nuances. The case study presents a solution that required integrating recent advances in transient queue analysis with a deterministic logistics model. These challenges are presented in the context of logistical planning and risk analysis for a notional contingency scenario.