These days, utility managers find themselves juggling a daunting array of concerns. Their list of challenges includes the increasing frequency of severe weather events, the looming danger of cyberattacks, shifting government policies that vacillate between renewables and fossil fuels, and the escalating demand for energy to support burgeoning data centers.
Compounding these issues is the looming threat of emergent complex systems to the power grid—an existential menace that played a pivotal role in the infamous Northeast blackout of 2003. This event left over 50 million people in the dark across eight U.S. states and the Canadian province of Ontario.
The massive outage began on August 14, 2003, when a high-voltage power line in Ohio, burdened by intense heat and demand, sagged into an overgrown tree. This incident triggered the line’s deactivation. Unfortunately, a software glitch in the FirstEnergy Corporation’s alarm system failed to notify operators of the problem, setting off a chain reaction. While some customers saw power restored within hours, others endured darkness for days.
For experts in fields ranging from computer science and physics to biology and social sciences, this blackout stands as a textbook example of emergent complex systems failure. Similar phenomena can be seen in the collapse of Lehman Brothers in 2008 and the Deepwater Horizon oil spill in 2010.
John Savage, the former head of computer science at Brown University, offers insight into these systems: “An emergent system is one where complex organized behavior or properties arise from the interactions among simpler components, even though none of the individual components exhibit those complex properties on their own.”
Often-Cited Examples
To computer scientists, physicists, biologists, even social scientists, this was a classic example of emergent complex systems failure. Other often-cited examples are the collapse of Lehman Brothers Inc. in September 2008, and the Deepwater Horizon oil spill in the Gulf of Mexico in April 2010.
John Savage, the retired head of computer science at Brown University, describes emergent complex systems this way: “An emergent system is a system in which complex organized behavior or properties arise from interactions among simpler components, even though none of the individual components exhibit those complex properties on their own.”
In other words, Savage told me, “Emergence occurs when the whole is more than the sum of its parts — the collective behavior cannot be fully predicted or explained just by analyzing the individual elements in isolation.”
Robert Gardner, who has been a leading researcher and instructor on the behavior of emergent complex systems at the National Security Agency, the National Institute of Standards and Technology and leading universities, told me that any endeavor which relies on complex systems is liable to face an emergent failure, but electric utilities and the financial system are most vulnerable to this danger.
For example, Gardner said, the Lehman Brothers crash in 2008 was instigated by bad actors but exacerbated by the complexity of the counterparty networks that made it a gigantic crisis.
“These so-called Complex Adaptive Systems had no nefarious, adversarial actors to defend against; the complex, heterogeneous nature of the systems themselves led to emergent behaviors — behaviors that cannot be predicted by examining individual components, but are rather produced by the system as a whole, facilitating perfect storms that conspire to enable these catastrophes,” he said.
Other major catastrophes that begin with a perturbation deep in the system but have widespread consequences include the Boeing 737 Max 8 autopilot system problems and the Titanic disaster.
Science of Chaos
Some compare the development of a complex emergent failure to the science of chaos, where a minor action on one continent can affect the weather on another.
Software is always of concern because it has often evolved, and the component codes aren’t necessarily known today. They can be a mystery to those dependent on them.
Axiomatically, the bigger the system, the greater the vulnerability and the more stupendous the consequences. For utilities, the big vulnerability is the grid — all or part of it.
The solutions run to ever-greater awareness, building in resilience and having a system that can shut down in parts defensively, and instantly when an unexplained perturbation in normal operations occurs.
As electricity supply gets more complex, more diversified, the chance of emergent disasters grows. No villains, just cantankerous complexity.
EPRI, the research organization for the electric utility industry, has studied emergent complex systems and leans toward thoroughness and robustness in design with in-depth resilience.
Advocates for microgrids see the relative simplicity of these and the ability to insulate them from the grid as a major advantage. Meanwhile, stuff happens.
