Aerospace Systems Design Laboratory, Georgia Institute of Technology
August 2018 - April 2019
Project Manager: Sponsored by Lockheed Martin, Rolls Royce, Australian Department of Defense; requiring a flexible simulation environment for quantifying sustainment capabilities of strategic airlift missions, with respect to humanitarian crises.
Objective: Aerial sustainment operations – particularly in the context of Humanitarian Aid and Disaster Relief (HADR) missions – are complex and vital missions with a high degree of inherent uncertainty. To better understand and prepare for these operations; motivated by their increasing frequency, severity, and their use as an ideal case study for strategic airlift sustainment operations, the Agent-Based Tactical Logistical Assistance Simulation (ATLAS) environment has been developed to model a humanitarian response to a Category 5 cyclone such as Cyclone Winston that devastated Fiji in 2016.
Utilizing an agent-based modeling approach in Python, a decision support environment was developed and presented; enabling operational and fleet composition tradeoffs to gauge the impact on key mission metrics such as the amount of cargo delivered, sortie generation, and utilization rate. The capability of this environment was demonstrated as applied to the Fiji case study, and modified scenarios based on real-world HADR responses by the Australian Defence Force are employed to validate the model.
Contribution:
Led the 6-person team to develop the Agent-based Tactical Logistics Assistance Simulation (ATLAS); designed and developed as an agent-based modeling framework to quantify the impact of alternative fleet compositions on mission effectiveness metrics
Modeling and simulation environment implemented in Python, with results presented at the 27th ASDL External Advisory Board, Lockheed Martin executives, and to the Australian Department of Defence in Melbourne, Australia
Developed environment still presently utilized; for both analysis of humanitarian scenarios, and as a base for future iterative development across numerous projects
Produced documentation, and mentored and instructed succeeding grad students on the framework to facilitate legacy usage
Work published as lead author at AIAA Scitech 2021, with three subsequent ASDL publications leveraging the developed framework
Skills:
Software development (Python)
Agent-based modeling and discrete-event simulation
Project management, leadership, and teamwork
Industry sponsor relationships
Presentation and communication
Takeaways: The development of the ATLAS environment – which started as an industry-sponsored ASDL Grand Challenge in 2018 – has been one of my most instructive and educational projects to date. Along with the technical aspects of the project, leading a 6-person team of first-year graduate students, and additional undergraduate students was a novel challenge, in addition to taking broad customer requirements and translating them into actionable research questions, and technical deliverables. Continuing interaction, feedback, and updates from the customer were key aspects that gave me a higher-level view of such projects.
Further developing this project into a master’s thesis increased my technical knowledge of software development, particularly when applied to specific research questions. Continued development and integration of the platform with the efforts of successive Grand Challenges has allowed the platform to remain in use. As part of the Lockheed Martin Aerial Firefighting project, I adapted the framework to model tactical aerial firefighting operations to analyze the impact of improved aerial asset performance. Updating and applying the original software to a range of applications gave me a broader understanding of how modeling methodologies can be applied to a host of real-world problems.