- Split View
-
Views
-
Cite
Cite
Guru Madhavan, John L Anderson, Alton D Romig, Engineering for inevitable surprises, PNAS Nexus, Volume 1, Issue 1, March 2022, pgac014, https://doi.org/10.1093/pnasnexus/pgac014
- Share Icon Share
A decade ago, Japan's peculiar electric grid stalled its recovery from an earthquake, a tsunami, and a nuclear accident at Fukushima. The afflicted northeastern parts of the country operated on German generators at 50 Hz; the unaffected southwestern regions powered by American units ran at 60 Hz. This incompatibility gave rise to a profusion of disruptions felt worldwide. Power transfer became impossible when needed, exacerbating the aftereffects of the triple-shock. Global automotive supply chain operations ground to a halt in many countries with Japan's sudden inability to manufacture and export prime engineering ingredients for assembling motor vehicles, such as airbags, fuel injectors, specialty paint pigments, key molders, and microchips.
As software, sensors, and electronics gain more functionality, a concentrated failure in one region can swiftly spread into a distributed calamity. Such point-to-pervasive incidents have become recurrent and ruinous (1) Escherichia coli can spread from a back porch lunch gathering to contaminate a country; bats can topple Wall Street from a wooded pond. The coronavirus pandemic induced many supply chain setbacks that have made global recovery harder. The stakes of such repercussions can range from delaying critical deliveries to climate, commercial, and civic survival (2).
Our vision often links to what we have experienced in the past. Consequently, when facing complex problems, we tend to revert to the tactics used to fight a previous emergency rather than comprehending a current one. For instance, the modus operandi deployed in response to the Chernobyl nuclear disaster was too often used to try to mitigate the effects of the Fukushima disaster, even though the circumstances were materially different. Similarly, COVID-19 had its fair share of comparisons to the influenza pandemic a century earlier. Such nexus scenarios create a confluence of disparate time horizons, competencies, cultures, and perspectives. They require a critical capability to simultaneously look to the past for reliable evidence on what has succeeded and failed, while weighing the exigencies of the present and, importantly, equipping ourselves for future surprises.
As a typology, though, strict surprises are not entirely new. They may even be predictable, like a surprise birthday party (3). But inevitable surprises—in health, infrastructure, climate, and democracy—depart from norms and expectations, requiring consideration of substantive possibilities over numerical probabilities (4). As processes of phase transition illustrate, something with one set of properties can transform into a different state with a disparate set of impacts. So, how then do we guide the evolution of complex systems? How do we adeptly engineer for their adaptations and maladaptations?
Systems—and their surprises—can become too complex even for the more informed and influential entities to be planned top-down (5). Insights from cultural evolution suggest that engineering approaches must become experimental and more collaborative when civic welfare is the unit of selection. By default, such methods will involve variation, selection, and finetuning of tradeoffs—margins and uncertainties in technical understanding, competing market motivations, and opposing policy pressures—to iteratively reduce detrimental consequences, unwanted, unintended, and unattended. This leads to a prime problem in engineering education and practice resulting from the fallacious separation of the “technical” from the “social.” Likening engineering only to technology and not sociology is akin to saying evolution is only gene-based and not related to culture, and economics is about rationality and has nothing to do with behavior. After all, engineering is a communal endeavor. But by not effectively engaging with the effects of technology, behavior, culture, and policy, engineering can also disappoint in its civic responsibilities.
Systems engineering provides a spectrum of tools—scenario planning, multicriteria simulations, maintenance regimes, and failure mode evaluations—to engage the shifting connections and conflicts that characterize our world. To engineer resilient communities requires anticipation of and diligent planning for inevitable surprises, and a fundamental appreciation of people, systems, and culture—the guiding themes of the National Academy of Engineering's programs. Because engineers are dual citizens in the worlds of disaster and achievement, blending wonder and worst-case thinking is a linchpin of engineering practice. Indeed, no other form of professional practice reaches the scale of creation and consequences inherent in engineering design. Therefore, in the mixed company of technology, society, and policy, engineering for people, systems, and culture is no longer sufficient. The emphasis needs to expand to encompass the people, systems, and culture of engineering. One vital step in this progression is making the profession more inclusive and welcoming to the broadest range of participants, perspectives, and philosophies. More diverse viewpoints will inevitably lead to greater insight into what is possible, whether that is the height of accomplishment or the depth of failure. Engineering's past was grounded in the industrial revolution; engineering's future must be about inclusive evolution.
In a 1975 article, “The Survival of the Wisest,” the vaccine pioneer Jonas Salk suggested a visual metaphor for societal transitions (6). He posited an S-shaped curve to denote two key phases, much like the beginning of an alternating-current sine wave. He called the first, rising part of the curve Epoch A, where humans are more individualistic and independent with opportunistic and short-term rewards. In Epoch B, the flattening, upper part of the curve, humans are more collaborative and engaged, recognizing the ever more interdependent need for prosocial design and long-range systems thinking. Epoch A is a period with a zeal for efficiency, while in Epoch B, resilience is favored. If Epoch A is “just-in-time,” then Epoch B is “just-in-case.” If governments and businesses in Epoch A dodged disasters, then in Epoch B, they will need to circumvent catastrophes.
In our time, thus far with COVID-19, we can see a phase change. With its patchwork of policies, the Epoch A pandemic has evolved into an Epoch B syndemic: a confluence of health, economic, natural, and political calamities. As countries go into further debt to rescue families and communities, resilience is the multivalent vaccine against the strains and stresses of inevitable surprises. Engineering resilient supply networks of both industrial commodities and intellectual concepts is a starting point for survival in this newfound nexus.
An insight of historian Henry Brooks Adams is germane to engineering: if chaos is the law of nature, then order is the dream of humans (7). If we aspire to achieve order, whether it's power grid resilience or pandemic response, we cannot afford to be operating on incompatible frequencies.
Authors' Contributions
GM, JLA, and ADR developed the concept and edited the piece. GM wrote the first draft.
Notes
Competing Interest: Guru Madhavan is Norman R. Augustine Senior Scholar and Senior Director of Programs, John L. Anderson is President, and Alton D. Romig Jr. is Executive Officer of the National Academy of Engineering.