Multi-Objective Analysis of Stormwater Infrastructure Retrofits using Genetic Algorithms: Two Case Studies
Hydrologic and hydraulic models have become fundamental engineering tools for planning and design of stormwater infrastructure, including projects involving flood control, water quality improvement, hydromodification management, and stream rehabilitation. The traditional modeling approach involves identifying a suite of potential design alternatives, assessing the performance/benefits of each with respect to specific objectives, and then selecting the most cost-effective alternative. While this approach is often useful and appropriate, the number of alternatives to evaluate can quickly become extremely time consuming for complex stormwater networks and projects with multiple objectives. With the advancement of optimization software and cloud based computing, the potential solution space can be much larger and allow for a greater number of alternatives that can be effectively evaluated.
This presentation will introduce the use of OptimizerTM, a highly flexible and customizable platform for stormwater infrastructure planning. Operating on the widely used, open source EPA SWMM model, this platform utilizes genetic algorithms and cloud computing to evaluate the costs and benefits of thousands of feasible alternatives to identify near optimal solutions. Since it can operate on a complete hydrologic/hydraulic network, the platform is ideal for evaluating combinations of gray and green infrastructure and/or combinations of upland (i.e., flow control) and in-stream (i.e., stream rehabilitation) approaches. The flexibility to define optimization objectives and criteria within a truly networked system, the transparency of the workflow, and highly trained customer support staff sets this approach apart from less flexible or less transparent optimization platforms.
Two case study examples will be presented that optimized stormwater infrastructure retrofits. These cases present opportunities to transfer lessons learned from progressive communities in other parts of North America. The first example is for a 27 square mile urban sewershed in Chicago’s southside that is subject to frequent flooding including widespread basement backups and impassable streets. This study evaluated whether green infrastructure could be used to replace or reduce the size downstream pipe improvements while also providing additional benefits such as improved water quality and neighborhood beautification. This is the first stormwater retrofit planning study in Chicago that demonstrates that green infrastructure can be effectively implemented to improve the Level of Service (LOS). The study received this year’s Innovation Award at the Illinois Association for Floodplain and Stormwater Management Conference.
The second example is for a 0.5 square mile neighborhood in the City of Calgary also subject to frequent basement backups and street flooding during intense storm events. The neighborhood discharges to Nose Creek, an urbanized stream with sediment loading targets analogous to TMDL waste load allocations in the United States. The neighborhood is experiencing infill growth and the City is concerned with increased flooding and increased pollutant loading. OptimizerTM was used to evaluate an array of potential alternatives including pipe upgrades, downspout disconnection programs, street planters, and regional storage to identify the least cost solution that provides the highest benefit.
For both of the examples, collaboration was key to the success of the projects. For the Chicago project, an extensive public outreach effort was completed to better understand the issues and concerns the residents have with urban flooding and potential solutions. For Calgary, city planners were directly involved in formulating the project objectives and discussing how their triple bottom line (TBL) approach for assessing capital projects based on the social, economic, and environmental costs and benefits could be integrated into the optimization process.
Mr. Leisenring has 15 years of water resources and watershed management experience working on projects for a variety of private and public sector clients. He specializes in environmental data analysis, regulatory compliance, pollutant source control, stormwater treatment and management, and watershed based approaches. His professional experience has included conducting water quality assessments, developing decision support tools, preparing and implementing monitoring plans, and working with academics, regulators, and program managers to identify mutually agreeable solutions to complex water resources problems.
Aaron Poresky supports clients in the municipal, private, and applied research sectors with challenging issues related to watershed management. He has more than 11 years of professional experience, including high profile and complex projects related to stormwater management, green infrastructure/low impact development, integrated planning, stream protection/restoration, flood risk mitigation, groundwater augmentation, and environmental monitoring.