Study Examines How Algorithms Can Help with Water Management
Most people take their tap water for granted, but it is actually delivered by a complex and important water-management infrastructure. In the face of a changing climate and growing population, Colorado water managers need new tools to continue providing a high-quality water delivery system.
Assistant Professor Joseph Kasprzyk, working alongside Associate Professor Lisa Dilling of CU鈥檚 Environmental Studies Program and Rebecca Smith of the Bureau of Reclamation, researched the potential for water agencies to use multi-objective evolutionary algorithms (MOEAs) to enhance their long-term planning. Their research was recently in the journal Environmental Modelling and Software.
MOEAs were developed to aid decision-making for complex systems. In a system with several competing objectives, MOEAs can analyze thousands of combinations to find a solution that maximizes positive outcomes for the entire system. In a water-management system, this means that MOEAs could balance tradeoffs between efficiency, reliability and environmental impact, lowering costs and lengthening the lifespan of water infrastructure.
The research team hosted a workshop with Colorado water managers to analyze the effectiveness of MOEA use. The water managers went through the decision-making process for several different water-management scenarios. They learned how they could utilize MOEAs, while researchers analyzed their performance.
The workshop allowed the managers to explore the multiple benefits of using MOEAs. While there are challenges to their widespread adoption, the benefits of using MOEAs may allow them to become a vital tool in the planning process, Kasprzyk said.
鈥淚t is an exciting time to be part of the CEAE department, since there are an increasing number of utility managers who are interested in putting the tools from our research into practice.鈥