While consumers face tough choices that affect both their finances and the environment, decision making is even tougher for executives and policy makers. The ability to decide when and where to expand production capacity—or whether to replace, repair, recycle, or simply retire equipment—can determine if a company will be economically and environmentally sustainable.
Sarah Ryan, a professor of industrial and manufacturing systems engineering, helps industries achieve sustainability by developing integrated mathematical models that help decision makers deal with risk and uncertainty as they strive to balance our increased demand for energy, products, and services with preservation of the earth’s environment—all the while remaining profitable as businesses.
“A company may be operating optimally from an economic standpoint in the short term,” says Ryan. “But if it is using nonrenewable resources or polluting the environment, it will not be sustainable in the long term. On the other hand, an industry can be ‘green’ and take all kinds of environmentally friendly actions. But if it is not profitable, it is not going to survive.”
A rational view of systems
Ryan’s interest in decision-making tools began soon after she finished her PhD at the University of Michigan in 1988, where her studies focused on optimization over time. In her first appointment at the University of Pittsburgh, she studied electric power issues, observing how the inconsistent availability of different generating plants complicated planning processes.
Ryan gained perspective on that challenge as well as capacity expansion issues when, in the early 1990s, she joined the utilities division of a chemical manufacturing plant that generated its own electric power and steam for process heat. As the company grew and production increased, she notes, the plant’s boilers became increasingly strained. Managers, therefore, needed to determine the most cost-effective schedule for either replacing existing equipment or installing additional boilers to meet increased demand—a formidable challenge, given the uncertainties of markets and equipment needs over time.
“They didn’t know exactly what demand was going to be,” Ryan says. “So they drew on their day-to-day experience to forecast demand and developed methods over the years that would give them an approximate idea of when to install new boilers.”
After returning to academia in 1995, Ryan focused on developing models to quantify—and thus rationalize—responses to these kinds of problems. Using a combination of probability and optimization modeling, the models allow decision makers to see how various components of an issue interact with each other over time. By recognizing this interdependence, Ryan says, managers can make decisions that provide the best results for each contributing part, as well as for systems as a whole.
Sustaining ‘sustainable’ energy
A 2007 AT&T Faculty Fellow in Industrial Ecology, today Ryan has sharpened her focus on the interdependence of economic and ecological factors in demonstrating how these information modeling techniques can help companies make decisions that reduce their consumption of both materials and energy.
Nowhere are these capabilities more needed than in the energy industry itself, with the wind industry offering a case in point. Although a green technology, the fact that wind produces power from a seemingly “free” and inexhaustible feedstock does not alone make wind sustainable, as it is still subject to the laws of economics.
The challenges are many: How do we integrate wind with other forms of energy? How do we minimize capital costs? How do we forecast the availability of wind with any degree of reliability? Where do we build wind farms, and how do we transmit their intermittent power to urban load centers over a grid currently ill equipped to handle that intermittence?
Yet another set of questions involves the ecology of developing, manufacturing, transporting, erecting, maintaining, and disposing of wind blades and turbines once they’ve completed their service lives. Formed of composite materials, the blades can be up to 150 feet long, making disposal a costly endeavor—both in dollars and environmental impact.
“We want to know how all of these pieces—people, materials information, equipment, and energy—fit together to make the whole industry sustainable,” Ryan says.
The confluence of systems
Yet wind represents just a small piece of the total energy picture. Supported by the National Science Foundation and Iowa State’s Electric Power Research Center, Ryan is currently working on several multidisciplinary projects to improve the efficiency, reliability, and environmental impacts of the larger U.S. energy infrastructure.
Electricity transmission networks, for example, are heavily dependent on other systems: if a bottleneck in the transportation network prevents the delivery of low-cost fuel, utilities might be forced to substitute more expensive alternatives, resulting in higher prices to consumers. So Ryan is also modeling the transportation networks for fuels such as coal, oil, and natural gas.
To better understand these challenges, last year Ryan spent five months at the University of Auckland’s Electric Power Optimization Centre studying New Zealand’s electricity market, which is similar to but smaller than the U.S. market. There, her students updated and modified a fuel transportation network model developed by ECpE Professor Jim McCalley’s students that defines the network’s parameters, variables, and constraints.
“The parameters are the constants in the equations,” Ryan explains. “Coal, for example, costs so much per unit. The variables include how much coal you ship from one location to another, and the constraints are you can’t ship more coal than what is available—plus you don’t want to ship more coal than is needed by the consumer.”
A second model then provided the perspectives of utilities producing the electricity, as well as the system operator managing its transmission. The producer seeks to maximize profits, which are determined by the cost of fuel and the amount and market price of the electricity generated. Constraints include the fixed capacity of each generator to produce power. In addition, the utility must anticipate how its generation decisions might impact price—and at what point consumers might reduce consumption in order to lower their bills.
Ryan then put the equations from both models onto a single platform to solve everything at once. “We did this partly to get a better understanding of the whole system,” she explains, “but the real goal was to figure out where to expand the system in the most efficient and cost-effective way so that everybody—producers and consumers—is better off.”
A model for the developing world
Sustainability may be a global challenge, yet Ryan’s work focuses on the developed world because, she observes, that is where most of the world’s energy and resources are consumed.
“In the United States, we use far more resources and energy than we probably have the right to,” Ryan says. “So I think improving the processes in the developed world will have the biggest impact.”
That’s a sentiment shared by NSF program director Cerry Klein, who points out that, given the continued successful development of modeling technologies such as Ryan’s, the world will follow our lead.
“The two largest nations—China and India—are working hard to catch up with the rest of the world in jobs and income,” Klein acknowledges. “ Consequently, emissions are not a big concern to them.
“Their business leaders, however,” he continues, “are very technically oriented. If we can share mathematical models with them that explain why sustainable practices are cost effective—and they see our own successful enterprises that have used these models in decision making—they will be more willing to adopt these ideas.”