Computer modeling is a powerful tool for analyzing complex urban systems. Indeed, for decades metropolitan-scale transportation planning has been informed by increasingly sophisticated computer models. In addition, models are commonly used to study all types of infrastructure systems, the urban environment, even possible location of future of urban growth. In fact, I’m building an attractiveness model for future residential development in South Florida in a class this semester.

However, models can have insidious effects. They excel when applied to deterministic systems, where the rules are static and known, but often fail when applied to systems with arbitrary or random characteristics. Even more troubling, models can impede decision-making by hiding their assumptions, introducing bias into the simulation.

In this light, let’s consider a simple model developed by the oil company Chevron. Their “Energyville” game is located the Will You Join Us website, now being promoted through magazine and TV ads that position the company as an energy company interested in finding energy “solutions” and using it “wisely.” Energyville is presented as a neutral challenge: “What energy sources will power your city?” A disclaimer reminds the user that the assumptions are “based on The Economist Intelligence Unit’s assessment of global facts and trends obtained form numerous credible sources.” The warning observes the game makes many simplifications, acknowledging “global forces and technological developments may change current and future assumptions.” The game aesthetics shows a clear influence of the popular SimCity.

Launching the simulation, I begin by placing some wind turbines in the city. After installing three turbines, the limit is reached: “Geographical and other constraints prevent Wind power from providing any more power to Energyville.” Next, I turned to solar panels. After just two placed, I get this error: “Unavailable! Solar panels are still too cost prohibitive and inefficient to provide any more power to Energyville.” The only remaining renewable energy source is a massive conventional-looking hydroelectric dam. After installation on the river, most of my city’s electrical needs are met.

All except for the ever-important petroleum. There’s no Better Place-type electric car networks possible here. “Warning! low on fuels,” a message quickly appears, saying I need petroleum for airplanes, vehicles, and mass transit. Only once I put a huge petroleum platform in the ocean could I proceed to the next level.

Before level 2 begins, the simulation presents a policy choice: should I adopt energy efficiency measures that will improve environmental quality and “security,” while placing a tax on economic output? Round two is similar, with a couple surprises. First, my wind farms are in trouble:

Ironically, my attempt at developing renewable sources was thwarted by the very global warming I am concerned with! Next, my solar program is in trouble:

I’m a bit confused by this one. After all, all energy costs money. Solar panels can only be “too costly” if cheaper alternatives are available. What these options are – continued petroleum, nuclear, or some other source, is not explained. In this round my fossil fuels are unaffected by catastrophe.

Repeated playing revealed other game paths have other possible events. In one case, solar panels become more attractive to homeowners due to net metering policies, and two actually make wind power even more attractive due to vaguely specified improved technology or other benefits. If you invest in nuclear it warns you uranium may increase in price due to global demand. Once, a terrorist attack in the middle east tightens oil supply. But petroleum price and supply rarely plays a role in the problem – despite the historical evidence as recently as the summer as 2008 that it can be subject to major price volatility. (Incidentally, I think the game was created in 2007.)

In the end, is this a fair simulation? Despite the capricious nature of some of the factors, most of the assumptions are probably reasonable. Presumably Chevron is too savvy to deliberately plant obvious biases in some of the assumptions, nevertheless I’m sure a serious energy wonk could find plenty to quibble about. However, like too many models, Energyville doesn’t clearly reveal its underlying assumptions, or allow the user to question or manipulate them. Although the limits placed on the speed alternative fuels can be rolled out are probably derived from mainstream sources, history shows change — whether beneficial or catastrophic — can be surprisingly rapid.

This means Energyville misses a major educational opportunity. (Despite it appearing on an educational blog – the only Technorati link to the site.) The flash interface makes it impossible to copy text and contains no links to external sources, and the “about” page lists dozens of unlinked articles, reports, and websites, and no assumption is presented as contested. Although thousands of players may learn a few facts embedded in the game, or gain a vague sense of the benefits and limitations of various energy sources, it doesn’t support serious examination and debate about energy technology or policy. But maybe that’s the point.

> WillYouJoinUs.com