Data-DrivenWater Allocation under Climate Uncertainty: A Distributionally Robust Approach
Abstract: This paper investigates the application of techniques from distributionally robust optimization (DRO) to water allocation under future uncertainty. Specifically, we look at a rapidly-developing area of Tucson, Arizona. Tucson, like many arid and semi-arid regions around the world, faces considerable uncertainty in its ability to provide water for its citizens in the future. The main sources of uncertainty in the Tucson region include (1) the unpredictable future population growth, (2) the availability of water from the Colorado River in light of competing claims from other states and municipalities, and (3) the effects of climate variability and how this relates to water consumption. This paper presents a new data-driven approach for integrating forecasts for all these sources of uncertainty in a single optimization model for robust and sustainable water allocation. We use this model to analyze the value of constructing additional treatment facilities to reduce future water shortages. The results indicate that DRO can provide water resource managers important insights to minimize their risks and, in revealing critical uncertainties in their systems, plan for the future.
Keywords: Distributionally Robust Optimization, Phi-Divergences, Water Allocation, Decentralized Water Infrastructures, Benefit-Cost Analysis
Category 1: Applications -- Science and Engineering (Civil and Environmental Engineering )
Category 2: Stochastic Programming
Category 3: Robust Optimization
Citation: Manuscript, submitted for publication.
Entry Submitted: 03/16/2018
Modify/Update this entry
|Visitors||Authors||More about us||Links|
Search, Browse the Repository
Give us feedback
|Optimization Journals, Sites, Societies|