0018-9162/00/$10.00 © 2000 IEEE 44 Computer Toward Self-Healing Infrastructure Systems V irtually every crucial economic and social function depends on the secure, reliable oper- ation of energy, telecommunications, trans- portation, financial, and other infrastructures. Indeed, these infrastructures have provided much of the good life that the more developed countries enjoy. However, with increased benefit has come increased risk. As they have grown more complex to handle a variety of demands, these infrastructures have become more interdependent. The Internet, computer networks, and our digital economy have increased the demand for reliable and disturbance-free electricity; banking and finance depend on the robustness of elec- tric power, cable, and wireless telecommunications. Transportation systems, including military and com- mercial aircraft, land vehicles, and sea vessels, depend on communication and energy networks. Links between the power grid and telecommunications and between electrical power and oil, water, and gas pipelines con- tinue to be a linchpin of energy supply networks. This strong interdependence means that an action in one part of one infrastructure network can rapidly create global effects by cascading throughout the same network and even into other networks. The potential for widespread disturbances is much higher, as the “The Cost of Cascading Failure” sidebar describes. Moreover, interdependence is only one of several char- acteristics that challenge the control and reliable oper- ation of these networks. These characteristics, in turn, present unique challenges in modeling, prediction, simulation, cause-and-effect relationships, analysis, optimization, and control. What set of theories can capture a mix of dynamic, interactive, and often non- linear entities with unscheduled discontinuities? Deregulation and economic factors and policies and human performance also affect these networks. The Complex Interactive Networks/Systems Initiative (CIN/SI), a joint Electric Power Research Institute-US Department of Defense program, is addressing many of these issues. The goal of the five-year, $30 million effort, which is part of the Government-Industry Collaborative University Research program, is to develop new tools and techniques that will enable large national infrastructures to self-heal in response to threats, material failures, and other destabilizers. 1,2 Of particular interest is how to model enterprises at the appropriate level of complexity in critical infrastructure systems. Part of CIN/SI’s work, which began in spring 1999 and involves 28 universities, draws from ideas in sta- tistical physics, complex adaptive systems (CAS), dis- crete-event dynamical systems, and hybrid, layered networks. CAS researchers view the complex system as a collection of individual intelligent agents that adapt to events and surroundings, acting both com- petitively and cooperatively for the good of the entire system. By simulating agent-based models, stake- holders can better grasp the true dynamics of complex intercomponent and intersystem actions. As models become progressively more realistic, designers can map each system component to an adaptive agent. The adaptive agents would then manage the system using multilevel distributed control. 3-5 Through its environ- mental sensor, each agent would receive continuous messages from other agents. If agents sense any anom- alies in their surroundings, they can work together, essentially reconfiguring the system, to keep the prob- lem local. Thus, the agents would prevent the cascad- A joint industry-government initiative is developing a mathematical basis and practical tools for improving the security, performance, reliability, and robustness of energy, financial, telecommunications, and transportation networks. The first challenges are to develop appropriate models for this degree of complexity and create tools that let components adaptively reconfigure the network as needed. Massoud Amin Electric Power Research Institute COVER FEATURE