Utilizing Computer Simulation to Ensure the Continuity of Critical infrastructure Services on a Dynamically Changing Planet
Posted: August 27, 2015 at 12:15 pm
Richard G. Little and William A. Wallace
Department of Industrial and Systems Engineering
Rensselaer Polytechnic Institute
Those tasked with ensuring the reliable delivery of infrastructure services in the face of climate change and its myriad effects on the built environment will need to confront many new challenges. Heretofore, the resilience of critical infrastructure has generally focused on designing robust systems to resist the loads imparted by some “design basis event” that connotes the worst that is likely to occur given what we know (or think we know) regarding a particular hazard. Designing for such a maximum probable event assumes that the earth of the future will behave much as it has in the past, but this will probably no longer be true as the rate of global change increases. How well these impacts are anticipated and addressed could have a profound effect on improving a community’s resilience.
As governments at all levels struggle to find the resources necessary to fund the most basic of public services, identifying and implementing the “best way” to increase community resilience to extreme events can stress human and financial resources to the breaking point. Not only must governments determine “What to do,” they also need to identify “How to do it.” Some examples of these questions and possible answers are shown in the following table.
What to Do
How to Do it
Avoid the Hazard
Withstand the Hazard’s Effects
Respond and Recover from the Hazard
All of these questions build on a more fundamental one. Namely, what is the nature of the hazard and what are its likely effects on the communities in question? For example, although many people associate high winds with tropical storms, in the U.S. most of the damage is caused by storm surge and related flooding. For example, “Superstorm Sandy” did not generate hurricane force winds when it made landfall along the New Jersey and New York coastlines in 2012. The extensive damage to property and infrastructure, as well as storm-related deaths, were caused for the most part by a storm surge superimposed on an unusually high tide. Approximately one year after Sandy, Tacloban in the Philippines was devastated by a massive storm surge caused by Typhoon Haiyan that killed thousands. Worldwide, deaths from historical storm-related surges are estimated to be in the millions. Obviously, both the effects of high winds and storm surge must be addressed to improve the performance of infrastructure during an extreme weather event and ensure the rapid restoration of critical services in the aftermath of one.
Traditional engineering design assumes that a “maximum probable event” can be specified and a design produced that will survive the expected stresses of that event. Implicit in this assumption is that we can predict, with some degree of certainty, what the future will look like and how things will behave in that future. However, the uncertainties of hazards, in terms of their timing and the magnitude of impacts as well as the impacts themselves, requires that planning take place closer to the realm of the “unknown – unknown.” This creates a situation where there is not a single event that can be addressed with a single design solution but rather a range of plausible scenarios that must be countered with robust strategies that will work when we have little or no idea of what the “right” answer actually is. The effects of climate change and sea level rise only compound this uncertainty. Dealing with so many possible events and outcomes requires tools that can manipulate huge datasets rapidly and display the results in a manner that is comprehensible to a broad range of stakeholders with vastly different backgrounds and technical expertise.
The Critical Role of Computer Modeling and Simulation
Computer-based simulation tools can rapidly construct multiple scenarios and do so based on the probability of an event occurring, the likelihood and magnitude of its consequences, and the cost and effectiveness of alternative ways to address them. Raising the awareness of public and corporate officials, business leaders, and the general public to the geographically specific impacts of potential hazards is a critical first step in increasing community resilience. These tools can also be used to test alternative strategies for response and recovery from these events; a role typically played by costly and time consuming “boots on the ground” drills and full-scale exercises.
MUNICIPAL is a decision support tool that was developed at Rensselaer Polytechnic Institute to assist emergency managers and managers of critical civil and social infrastructure (CCSI) in restoring these systems following a hurricane or other extreme weather event. The technology includes modeling capabilities and optimization and GIS software. Users can view CCSI systems and their interdependencies; visually assess damage to these systems; and make preparedness, restoration, and recovery decisions. MUNICIPAL consists of three modules: (1) the vulnerability module, which converts actual or assumed weather information such as wind gust, flooding, and storm surge into damage to CCSI systems and resultant service outages; (2) the optimization module, which computes the outages caused by the damage and the best solution for restoration and recovery decisions; and (3) the GIS software, which enables the users to visualize the CCSI systems, the interdependencies among them, damage and outages caused by the modeled hurricane, and the optimal restoration plan.
MUNICIPAL is the first attempt to design a decision support tool that is capable of supporting pre-disaster preparedness and post-disaster restoration decisions for CCSI systems, while specifically accounting for the interdependencies among these systems. The test bed for this project was New Hanover County, NC, a hurricane prone coastal county of approximately 250,000 residents. The technology was built while working closely with the emergency management team of New Hanover County and members of the local infrastructure agencies and the guidance provided by this stakeholder group from the beginning stages was instrumental in the development of this tool. Although initially designed to be a working tool deployed during an actual emergency, feedback obtained from stakeholders indicated that it could have more immediate value to the practitioner community as an educational and training aid.
The vast majority of natural disasters are experienced at the local level and much of the disaster research produced after the 9/11 terrorist attacks and Hurricane Katrina emphasizes the importance of local governments in the response and recovery process. This is not surprising. Despite resources that may eventually be provided by higher levels of government, “many communities can expect to be ‘on their own’ for the first seventy-two hours.” As a result, MUNICIPAL was developed with the needs of local practitioners in mind and tailored to be “user friendly.”
There are difficulties in developing an appreciation among public officials and citizen groups of natural hazards when the knowledge necessary to do so is not readily available either physically or intellectually. This does not necessarily call for the development of new knowledge but rather, the more effective dissemination and utilization of what is already known. MUNICIPAL makes use of readily available, open source data and provide users with a means to rapidly assess and display the impacts of extreme weather events and storm surge on the provision of infrastructure services and help to establish priorities for repair that will minimize the time necessary to restore the community to full functionality—a key step toward improved resilience.
One of the problems frequently encountered during the restoration phase of emergency management is the lack of cooperation among different organizations that may have vastly different cultures and organizational structures. This is especially true when dealing with utilities because each organization views its primary task as restoring service to its own customers. A principal reason for this is that people and organizations are key elements in effective solutions but at its core, disaster recovery is a problem for which neither our institutions nor the professionals who staff them are typically organized or trained to deal with holistically. By their nature, extreme events are rare and unpredictable and if not unique, nearly so. It is difficult to structure traditional training exercises so that they address the multiplicity of possible paths over which the event could unfold.
Decisions regarding the restoration of infrastructure damaged by a hurricane can be complicated and without a structured process and guidance, important considerations can be overlooked. Among the most important of these are the complexities introduced by the interdependencies that exist between infrastructures and the socio-technical nature of the systems themselves. These factors require broad awareness, collaborative decision-making, and a global perspective that the needs of the community be put ahead of the institutional objectives of the individual service providers.
One objective in the development of MUNICIPAL was to show that better collaboration during the restoration of critical infrastructure services can lead to significant improvements in overall community resilience. If the managers of the critical services, emergency managers and responders, and community representatives understand that joint, informed decisions can result in better outcomes, then the restoration process can be more efficient and effective. The value of a decision support tool such as MUNICIPAL is that public officials and senior managers who may not have direct familiarity with extreme events, and who rarely have time to participate in full scale exercises, can immediately see the results of alternative approaches to responding to, and recovering from, an extreme weather event.
Disruptive events are unavoidable in the modern world. Our basic systems are at risk from threats we may not yet foresee. We need to anticipate these threats to our physical infrastructures, design systems that are inherently safer and more robust, and be prepared to restore them when they fail. The overarching benefit of MUNICIPAL will be to enhance the resilience of critical civil infrastructure systems through proactive planning, capacity building, risk reduction, training, and education; all of which are keystones of community resilience. Although developed for a coastal community in North Carolina, the methodology and tool is scalable to both larger and smaller communities facing similar hazards.
Richard G. Little is a Visiting Research Scholar in the Department of Industrial and Systems Engineering at Rensselaer Polytechnic Institute working on issues of disaster preparedness and community resilience. He can be reached at email@example.com.
William A. Wallace is the Yamada Corporation Professor in the Department of Industrial and Systems Engineering at Rensselaer Polytechnic Institute. His research interests are in analytical approaches to emergency management and the computer simulation of complex public and private processes.
 United States Department of Commerce, Service Assessment: Hurricane/ Post-Tropical Cyclone Sandy, October 22-29, 2012, (Washington, D.C.: National Oceanic and Atmospheric Administration, 2013), available at http://www.nws.noaa.gov/os/assessments/pdfs/Sandy13.pdf.
 Kate Hodal, “Typhoon Haiyan: At Least 10,000 Reported Dead in Philippine Province,” The Guardian, November 10, 2013, http://www.theguardian.com/world/2013/nov/10/typhoon-haiyan-thousands-dead-philippines.
 “Rensselaer Researchers Unveil Prototype Disaster Management Simulator: Engineers and Scientists Collaborate To Develop Powerful New Decision-Making and Data Visualization Tools,” RPINews, June 7, 2012, http://news.rpi.edu/luwakkey/3053.
 Earl E. Lee, William A. Wallace, John E. Mitchell, and David M. Mendonca, “Decision Technologies for Protection of Critical Infrastructures,” (paper, proceedings of Working Together: R&D Partnerships in Homeland Security, Boston, May 2005), available at http://eaton.math.rpi.edu/faculty/Mitchell/papers/decisiontechnologies.pdf.
 Ryan A. Loggins and William A. Wallace, “Rapid Assessment of Hurricane Damage and Disruption to Interdependent Civil Infrastructure Systems,” Journal of Infrastructure Systems 10.1061 (2015), available at http://ascelibrary.org/doi/abs/10.1061/(ASCE)IS.1943-555X.0000249.
 Richard G. Little, Ryan A. Loggins, and William A. Wallace, “Building the Right Tool for the Job: Value of Stakeholder Involvement When Developing Decision-Support Technologies for Emergency Management,” National Hazards Review 10.1061 (2015), available at http://ascelibrary.org/doi/abs/10.1061/(ASCE)NH.1527-6996.0000182.
 Jeanne-Marie Col, “Managing Disasters: The Role of Local Government,” Public Administration Review 67, S1 (Dec. 2007), available at http://unpan1.un.org/intradoc/groups/public/documents/undp/unpan032134.pdf.
 Margaret R. O’Leary, ed., The First 72 Hours: A Community Approach to Disaster Preparedness, (Nebraska: iUniverse, 2004).
 David E. Alexander, “Making Research on Geological Hazards Relevant to Stakeholders’ Needs,” Quaternary International, 171-172 (Aug.-Sept. 2007): 186-192, available at http://www.sciencedirect.com/science/article/pii/S1040618207000274.
 Louise K. Comfort, “Rethinking Security: Organizational Fragility in Extreme Events,” Public Administration Review 62, Special Issue (Sept. 2002): 98-107, available at http://www.jstor.org/stable/3110179; William Leavitt and John J. Kiefer, “Infrastructure Interdependency and The Creation Of A Normal Disaster: The Case Of Hurricane Katrina And The City Of New Orleans,” Public Works Management and Policy 10, no. 4 (Apr. 2006): 306-314, available at http://pwm.sagepub.com/content/10/4/306.full.pdf; Michael McGuire and Debra Schneck, “What if Hurricane Katrina Hit in 2020? The Need for Strategic Management of Disasters,” Public Administration Review 70, Special Issue (Dec. 2010), available at http://onlinelibrary.wiley.com/doi/10.1111/j.1540-6210.2010.02273.x/pdf.
Write to the Editors at firstname.lastname@example.org