Injustices, inequities, and inefficiencies are what Gregory Ewing is working to combat in his research field. An engineer, Gregory Ewing received his bachelor’s and master’s in civil and environmental engineering from the University of Michigan and is currently working towards his Ph.D. at the University of Iowa through the Sustainable Water Development program. Prior to starting his program at Iowa, Ewing investigated the benefits of continuous water monitoring and active control in water conveyance networks, specifically metropolitan sewers. This opened his eyes to the potential of the technology to impact, both positively and negatively, a broad scope of society and its water issues. Ewing’s project aims to be an initial attempt to explicitly recognize social objectives as a component of decision-making for water conveyance networks and other environmental areas.
In 2021, Ewing received funds from the Iowa Water Center Supplemental Student Grant Competition Program. This competition selects research projects that seek to address emergent water resource problems in Iowa. Preference is given to new scholars who will significantly impact their fields. Ewing was awarded $5,000 from this grant program for his project “Coupling Physical and Societal Objectives for Decision Making in Water Resources Management.” He matched the grant with $10,000 from partners and university support.
Gregory Ewing’s project sets out to reform the decision-making methodology made possible by the growing deployment of sensor networks in urban sewers— which monitor flooding and water quality of rivers— as well as controlling technologies that enable manipulation of these systems. Ewing argues that the current methodology does not consider societal and political structures, focusing only on physical and technical characteristics. He proposes to explore a new methodology that incorporates ethics and justice into the decision support toolchains. “This is an exciting area because its application in engineering practice is growing,” says Gregory Ewing of this new combination of technologies.
The main programming software to be utilized throughout this project is Pystorms, a Python Programming Language package facilitating the application of active control in water conveyance networks. Pystorms produces real-world stormwater control scenarios that will be applied and tested by Ewing through a programming interface and stormwater simulator (EPA-SWMM via PySWMM). Pystorms is a benchmark tool for simulation that will bolster the reproducibility and usability of Ewing’s experiment and results.
The experiment is set to use different control approaches— with and without explicitly including societal objectives and ethical considerations in making a decision— to see how outcomes differ. Provided in the benchmarking toolbox are sample scenarios. These scenarios are pairs of conveyance networks (possessing different characteristics), and the objective is to perform on the networks (i.e., minimize flooding, maximize water to a treatment plant, normalize flows to the outlet, etc.). These scenarios will be used as the foundation on which the investigation will build. The importance of downstream communities will be weighed and voting-based preference models— wherein the public is given a voice to express their opinions and concerns— will be created. For example, one question to test as part of the experiment would be, “How does the overall outcome of the simulation change if there is a secondary objective to avoid flooding in the poorest (or richest) neighborhood?”
İbrahim Demir, an Associate Professor at the Department of Civil and Environmental Engineering and Electrical and Computer Engineering at the University of Iowa, has helped oversee Ewing’s project. When asked how he believes this project will benefit the scientific community, Demir said it is an important step towards initiating “conversations on values and system expectations at every step of the decision process.” This framework will help engage people and lead them to consider ethical preferences regarding algorithmic decision-making.
“The project’s outcome will be an initial understanding of how explicitly incorporating social objectives into control decisions may affect the water environment outcomes,” says Gregory Ewing. This project is set to be a benchmark study for a methodology that seeks to combine physical and social objectives during decision-making processes and observe the outcomes.