Virginia Tech researchers develop system-of-systems framework to unify knowledge across disciplines
April 22, 2019
With the goal of reconnecting fragmented knowledge, John Little and his colleagues have developed a tiered, system-of-systems framework to help solve complex socio-environmental problems by incorporating models from a variety of disciplines.
A few years ago, Little taught a class on sustainability. “I thought I would look into the sustainability knowledgebase and pick out the quantitative procedures that can be used to evaluate sustainability, so I could teach those. I found that what I was looking for hardly exists and thought that there had to be better approaches to evaluating sustainability,” said Little, the Charles E. Via Jr. Professor of Civil and Environmental Engineering in the College of Engineering at Virginia Tech.
The idea behind this approach is to simplify each system, or upscale it, in order to capture the essential dynamics. Within each system, there are multiple interacting elements. Systems connect to other systems when their elements interact. For example, agricultural systems often interact with soil and land use systems; the business of farming interacts with economic systems, which interconnect with social systems.
“The idea is to upscale the models to the systems level and connect them together there. There may be some cases where we don’t have a process level model, so we go directly to the systems level,” said Little, an affiliated faculty member of the Global Change Center, housed within the Fralin Life Science Institute. “By connecting them there, we can see what the main drivers are of the problem that we are trying to solve. In a way, modeling is the common language of science.”
Little and colleagues recently published a study in the journal Environmental Modelling and Software. They proposed that a tiered, system-of-systems framework would be most useful in developing sustainable and resilient solutions for complex socio-environmental problems.
Sustainability incorporates social, economic, and environmental subsystems, and when all three subsystems are in balance, the overall system is thought to be sustainable, or able to be maintained through time. A system-of-systems approach groups smaller, individual systems into a larger one that functions at a higher level than the sum of each individual system.
Little’s approach trades detail for broader knowledge of the entire system; detailed process-level models inform more aggregated system-level models. By including a simplified representation of the entire system, more systems can be incorporated.
“We need to have a system-of-systems model to capture the essential dynamics of the individual systems. Then we can see what drives the problem and focus on those things. We have to change the way we think and that’s the biggest battle,” Little said. “I want people to take whole systems and couple them. Then we may be able to solve the big problems.”
By simplifying and upscaling these watershed and estuary process models, they can be coupled with the economic system at a similar spatial and temporal resolution. Indicators reflect the actual state of the system, whereas orientors represent the desired state of the system.
“We have all these different, big-picture problems. The idea is that you choose your problem, your supreme orientor, and the initial systems you want to include in your system of systems and plug them in,” Little said. “Once the initial systems are being successfully simulated, you can add more systems, building complexity.”
In this framework, sustainability is the supreme orientor, or the ultimate goal for the system of systems. When the indicators match the orientors, the system of systems is considered to be in a sustainable state. As more systems are added, such as social systems, the evaluation becomes more comprehensive.
For example, a system-of-systems model could be used to address dead zones – areas of reduced oxygen levels – in the Chesapeake Bay. Cities in the watershed have large populations who need to eat. Fertilizer runoff from agricultural systems flows into the estuary, creating dead zones. How can agriculture and urban development be managed to prevent dead zones from forming? A system-of-systems model could be used to answer this question.
To address complex socio-environmental issues, researchers need to communicate across a wide variety of disciplines. However, collaboration is difficult when experts with different backgrounds cannot easily communicate. Little proposes that specially-trained “systems” experts in each discipline could upscale information from the process to the systems level, and then communicate that information with specially trained experts in other disciplines. In this way, knowledge would propagate more quickly across disciplines.
“Right now, there’s a barrier between all these disciplines – we can hardly communicate with people in other fields,” said Little. “We need to train people to communicate across different disciplines – not every detail, but the big picture. Information at the systems level would propagate very quickly because of these systems experts. They could integrate knowledge and overcome this barrier,” said Little.
Computer scientists are designing the proposed software framework so that it can be used to integrate many system-level models. The software will handle the exchange of information and the visualization of results.
Ideally, communities of scientists would develop models and upload them to a cyberinfrastructure repository. This would allow researchers to share models at the regional scale and use whichever models are needed for their specific problem. However, it is difficult to organize collaboration among such large communities.
“We need specially trained people to connect these models, so they can communicate across disciplines, as well as within their own discipline. They could speed up the propagation of information across knowledge domains,” Little said.
To address this, Little and colleagues received funding from the National Socio-Environmental Synthesis Center (SESYNC) to hold workshops to develop design guidelines for the framework. The goal is to make the software as user-friendly as possible to guide researchers in sharing their models and using other models as part of their systems analyses. Four Ph.D. students – including one advised by Little – are implementing portions of the framework as part of their Ph.D. research.
“The Global Systems Science workshop involving Tony Jakeman and Sondoss Elsawah that was held in Blacksburg last March and funded by Fralin, and the seed funding from the Global Change Center, were crucial in getting this going,” Little said. The Global Systems Science Destination Area focuses on finding solutions to critical problems associated with human activity and environmental change, including disease, water quality, and food production.
-Written by Rasha Aridi