Installation Resilience - Center for Infrastructure Defense
Modeling Compound Threats to Interdependent Infrastructure Systems on Military Installations
Sponsor: Office of the Secretary of Defense (OSD) Strategic Environmental Research and Development Program (SERDP)
Project Abstract: The purpose of this project is to develop methods that measure worst-case disruptions across interdependent infrastructure systems on US Department of Defense (DoD) military installations and to create models that support DoD infrastructure planning and management. The extensive damages experienced in the wake of disasters like Hurricane Florence at Marine Corps Base Camp Lejeune revealed two important deficiencies in DoD infrastructure planning and recovery: (1) military installations lack models to measure losses resulting from interdependent failures across infrastructure (e.g., water and electricity), and (2) military leadership lacks methods to incorporate compound threats into infrastructure investment plans.
Technical Approach: This project addresses deficiencies in infrastructure planning and recovery by applying modeling methods used to identify worst-case losses in national infrastructure systems (e.g., the US power grid) to interdependent systems on military installations. Specifically, this project centers on advancing established operations research models for worst-case disruptions called “attacker-defender” models. Research activities include the development of modeling architectures for interdependent infrastructure systems and novel attack scenarios to identify worst-case disruptions to compound threats. Modeling architectures and methods will be used to design a new mission dependency index that embeds the interruptibility, relocateability, and replaceability of interdependent assets. Finally, we will develop case studies that identify vulnerabilities and optimal protection capabilities for military installations and inform long-term investment and planning.
Naval Postgraduate School
Dr. Daniel Eisenberg, Operations Research, Principal Investigator
Dr. David Alderson, Operations Research
Tahmina Karimova, Program Manager, NPS Energy Academic Group
Jacob Wigal, GIS Specialist, NPS Energy Academic Group
US Department of Energy
Converge Strategies, LLC
Towards a New Mission Dependency Index using Attacker-Defender and Defender-Attacker-Defender Models
LCDR Aaron Fish | M.S. Thesis in Operations Research (March 2021, expected)
The Commander, Navy Installations Command (CNIC) established the Mission Dependency Index (MDI) as the foundation for assessing the relative importance of infrastructure via risk and mission criticality. MDI is described in CNIC requirements as, “a survey… …to identify the intra and inter dependencies between functional/mission areas and the risks associated with interrupting or relocating the functions that take place in [installation] facilities,” (CNIC 2018). Since its inception, MDI has been adopted by all major DoD services for infrastructure planning, protection, and investment. Despite its popularity, MDI relies on expert elicitation and an unsound calculations (Kujawski & Miller 2009) and currently fails to deliver concise information for protecting infrastructure. We propose to reimagine MDI calculation by using vulnerability assessment methods based on network flow models. The goal is to develop a framework that links MDI and its assessment to reproducible methods and a rigorous analysis of facility criticality with respect their operational needs. This thesis answers the following questions:
- How can we prioritize protecting critical infrastructure systems (e.g., fuel, electricity) serving high MDI facilities?
- How do these decisions change for different MDI scores across an installation?
- How can we guide MDI scoring and prioritization to ensure mission-critical facilities receive necessary protection?
Simulation Framework for Analyzing the Resilience of Forward Expeditionary Port Refueling Infrastructure
LT Daniel Pulliam | M.S. Thesis in Operations Research (March 2021, expected)
Sustained conflict with a near-peer adversary such as China will require a sizeable combat-logistics fleet and forward basing to maintain maritime superiority in the Pacific. The massive resource requirements of such a conflict will require military and commercial ports to serve as locations for refueling, rearming, and resupply efforts. While these ports and their associated critical infrastructure will present strategic targets for enemy attacks that threaten military operations, there are few models guiding the procurement and use of technologies that can protect and recover failed port operations. We are developing a discrete event simulation framework that models a forward commercial port (e.g., the Port of Saipan) and assesses its capacity to provide combat-logistics support for U.S. naval forces. The framework is designed to analyze the effects of different threat scenarios on ship refueling operations and measure the benefits of technologies that improve operational resilience. This work answers the following technical questions:
- How to develop a simulation framework that supports the analysis of refueling operations at numerous commercial and non-commercial ports in the Pacific region?
- What technologies are available to improve the robustness, extensibility, rebound, and adaptability (i.e., resilience) of port operations to worst-case disruptions?
- How should the US Navy Expeditionary Warfare Center (EXWC) prioritize efforts to improve port resilience?
Recovery Actions for Microgrid Infrastructure
LT Marci Hester-Dudley | M.S. Thesis in Systems Engineering (March 2021, expected)
A gap that exists for the DoD is the lack of a systems engineering focus in building microgrid resilience within DoD installations. Up until recently, microgrid research and improvements focus on system operation and benefits but have not used a holistic approach to improve resilience. One important gap is a lack of rapid, efficient, and reliable recovery actions for microgrid systems that have failed. In this work, we define metrics to evaluate the rebound of microgrid systems and create task networks for assessing the speed of recovery given uncertain and difficult decisions. The goal of this thesis is to establish a systems engineering understanding of how microgrid design and associated recovery actions influence installation resilience. The two research questions that will guide the systems engineering process of determining these recover actions are:
- What are the recovery action(s) to restore operation of a microgrid system given intentional attack and natural disaster threat scenarios?
- What is the potential improvement to microgrid system resilience when choosing from a set of different possible recovery actions?
Mission-Informed Interdependent Infrastructure Vulnerability Analysis
LCDR Amanda Jones | M.S. Thesis in Operations Research (September 2021, expected)
Military missions depend on critical infrastructure systems to provide electricity, fuel, water, mobility, and telecommunications among other services. These systems are also dependent on each other, such that worst-case failures for infrastructure operations can cascade across systems and affect missions. In this work, we will link past methods for interdependent system modeling and assessing the relationship between infrastructure systems and mission essential facilities to assess a mission-informed interdependent system vulnerability. In particular, we will focus on supply chain infrastructure via surface roads for accessing and relocating missions and critical facilities and their vulnerability to interdependent disruptions from power, fuel, water, and/or telecommunications outages.
Python-based Framework for Interdependent Infrastructure Analysis and Vulnerability Assessment
Maj Mattias Kuc, German Army | M.S. Thesis in Operations Research (Completed December 2020)
Critical infrastructure systems that provide services like electricity, water, mobility, and communications are interdependent on each other — the function of one depends on another, and vice versa. These interdependencies also introduce shared vulnerabilities – failures in a one system (e.g., electricity) can cascade into others (e.g., transportation) exacerbating failure consequences. Although the existence of interdependencies and their potential for cascading failures is well-known, most infrastructure models and analyses still focus on a single system in isolation. Currently, there is no standard way to connect two infrastructure models together for interdependent analysis and entirely new models must be made to consider cascades. The goal of this thesis is to develop a framework to link infrastructure models together and assess their shared vulnerability. This thesis completes the following technical tasks:
- Harnesses the structure of network flow models for measuring infrastructure operations to define standard techniques to link models via interdependencies;
- Develops methods to combine Pyomo network flow models to automate the creation of interdependent infrastructure models;
- Demonstrates methods by assessing interdependent vulnerabilities across fuel, electric power, and transportation networks that were not originally created for interdependent analysis.