The complexity of building energy consumption
It is estimated that buildings contribute 20-30% of energy use in the United States at an annual cost of over $100B. Buildings also contribute an estimated 35-40% of all US CO2 emissions resulting from building energy consumption. Any effort to decrease building energy consumption can thus have a substantial economic and environmental impact.
Much of the effort invested in building energy efficiency and conservation is focused on analyzing or simulating individual physical systems within a building, to help designers understand, e.g., what savings could result by replacing standard lights with high-efficiency fluorescents, or by using light-colored paint on a building’s roof. Typical approaches combine simulating the actual physical properties of building systems, and statistical data based on historical usage. However, the complex interactions between building systems and the environment make accurate estimations difficult.
The complexity of this problem increases dramatically when occupant behavior is included. Consider a simple hypothetical example: a building consumes $1M per year in electricity for lighting. Analysis might show that, given current use patterns, installing high-efficiency lighting would cost $1M and result in 50% electricity savings – $500,000 per year – which would lead to break-even in two years. However, suppose that the building owner invests in a campaign to increase awareness, leading to a 25% reduction in how much lighting is used by occupants, or an annual electricity cost of $750,000. The same high-efficiency lighting would now only save $375,000 per year, and would thus take nearly three years to reach break-even.
Even more complex interactions take place when one starts to consider all building systems – such as heating and cooling, appliances and data networks – and other aspects of occupant behavior which impact demand and usage patterns. For instance, improving climate control might encourage occupants to spend more time inside the building, leading to an increase in energy consumption (this is sometimes referred to as the rebound effect).
This type of emergent behavior is a hallmark of complex systems, systems whose overall behavior is determined in sometimes unpredictable ways by the elements of the system interacting with one another and with the environment. We can see examples of human-made complex systems all around us: traffic jams, stock market fluctuations and even sports team performance cannot be predicted even when we know well how each individual is behaving within the system. Similar principles are at work in natural systems, including for examples the flocking behavior of birds, the schooling behavior of fish, of the ability of social insects such as termites to build incredibly complex structures – without the benefit of blueprints and architects.
Traditional analytical techniques are ill-equipped to manage complex systems, whose behavior often exhibits sharp nonlinearities such as tipping points. In recent years, researchers at academic centers such as the Santa Fe Institute, as well as commercial entities such as Icosystem, have successfully studied and managed complex systems using Agent-Based Simulation (ABS), a simulation technique that captures the behavior of systems from the bottom-up. While ABS was initially studied primarily in academic settings, recently ABS has been used to solve a variety of complex business and technology problems in many industry sectors and problem areas.
ABS replicates in software the behavior of individuals, as well as their interactions with the environment and with other individuals. ABS then shows how overall system behavior emerges from these interactions, replicating complex system behaviors that cannot be captured with other analytical techniques and that are often unexpected or counter-intuitive.
Traffic is a classical example of a problem that is best captured with ABS. In particular, traffic jams are an example of an emergent behavior that seems almost paradoxical: each driver is trying to reach his or her destination, and yet traffic jams form even when there are no external factors to cause them. An ABS developed with the NetLogo software can be used to show how simple driver behaviors can lead to traffic jams. In the simulation, drivers accelerate when there is nobody in front of them, and they decelerate when they approach a car ahead. Using these simple rules it is possible to replicate traffic jams that looks surprisingly like traffic jams observed under real conditions.
In the context of building energy efficiency, ABS could be used to understand how occupant behavior impacts energy consumption, by simulating occupants going about their normal daily activities and reacting to their environment, e.g., turning on lights when it gets dark (and sometimes forgetting to turn them off), opening windows or turning on air conditioning when it gets warm, using different types of appliances, and so on. More importantly, an ABS could be used to estimate the impact of various initiatives, including changes to the building itself as well as communications campaigns to encourage energy conservation. This type of quantitative approach could be beneficial in understanding how best to allocate resources to improve energy efficiency.
Leave your comment below, or reply to others.
Read more from the Meeting of the Minds Blog
Spotlighting innovations in urban sustainability and connected technology
Over the last three months, the City of Tomorrow Challenge has convened communities in Pittsburgh, Miami-Dade, and Grand Rapids to share transportation experiences and build understanding around people’s personal mobility struggles. Join the conversation and submit a mobility idea for a chance to win $100K in pilot funding at challenges.cityoftomorrow.com.
Akron Civic Commons launched in 2016 as a demonstration project of Reimagining the Civic Commons. After selecting Summit Lake as one of the sites for reinvestment, we immediately recognized that one of the greatest challenges to the work was overcoming decades of broken promises. There was a legacy of things being done to the community, not with them, and a healthy skepticism and mistrust of government and community organizations. If we wanted to do this work, it was imperative that we restore trust as part of the process.
The data we have gathered about trees in this region are powerful, but are mostly meaningful because they are in fine enough in detail to be applicable at a local scale. We spent our first few years gathering data so we could identify solutions based on need and not speculation.