As part of this presentation, we will explore Hybrid Approaches that combine the capabilities of Machine Learning (ML), Metaheuristics, and state-of-the-art optimization solvers to provide the desired solution quality and computational efficiency for solving several problems of interest.

To demonstrate the potential breakthrough of our proposed approach, we use a variant of the Team Orienteering Problem (TOP) (Chao et al., 1996) and the Reconnaissance Mission Planning Problem (RMPP). In the RMPP, given a map containing multiple target locations with reward values and threat zones, a set of Reconnaissance (Recon) with a starting location, and a travel length constraint, the goal is to generate round trip routes for each Recon asset within their allotted travel length, maximizing the total rewards collected. The presence of the threat zones in the Reconnaissance Mission Planning Problem differentiates it from the traditional TOP instance. Further, the threat zones denote the presence of an early warning radar coupled with surface-to-air missiles (SAM), surface-to-surface missiles (SSM), or a combination of both.

We will then discuss the possible addition of uncertainty in various aspects of the decision-making process. This will provide some potential interesting future work in ISR and weaponeering for domains such as Maritime Domain Awareness and Space Domain Awareness.

BIO: Dr. Moises Sudit holds the position of Professor of Industrial and Systems Engineering at the University at Buffalo (UB) and serves as the Executive Director of the Center for Multisource Information Fusion. Additionally, he fulfills the role of Associate Dean for Research for the School of Engineering and Applied Sciences. Dr. Sudit also assumes the position of Chief Scientist at CUBRC, a not-for-profit Research Center in Buffalo. His primary research interests lie in the theory and applications of Discrete Optimization and Information Fusion. Specifically, he focuses on the design and analysis of methods to address challenges in Integer Programming and Combinatorial Optimization. His research endeavors aim to develop efficient exact and approximate (heuristic) procedures for solving large-scale engineering and management problems. Recently, Dr. Sudit has directed his efforts toward exploring the potential benefits of leveraging Artificial Intelligence and Machine Learning in scenarios where high-fidelity mathematical models are unavailable. He has successfully integrated Discrete Mathematics with Information Fusion to tackle Big Data NP-Hard problems. Dr. Sudit is recognized as an NRC Fellow through the Information Directorate at the Air Force Research Laboratory and has been honored with numerous scholarly and teaching awards, including the prestigious IBM Faculty Scholarship Award. During his tenure as Associate Vice President for Sponsored Research at UB, the university witnessed a 20% increase in sponsored research funding and the adoption of state-of-the-art administrative tools under his leadership. Dr. Sudit boasts an impressive portfolio of publications in distinguished journals and has served as the Principal Investigator in numerous research projects. He earned his Bachelor of Science in Industrial and Systems Engineering from the Georgia Institute of Technology, his Master of Science in Operations Research from Stanford University, and his Doctorate in Operations Research from Purdue University.

Host: Kari Sentz - CCS-3