DOE PAGES title logo U.S. Department of Energy
Office of Scientific and Technical Information

Title: A multistage distributionally robust optimization approach to water allocation under climate uncertainty

Abstract

This paper investigates a Multistage Distributionally Robust Optimization (MDRO) approach to water allocation under climate uncertainty. The MDRO is formed by creating sets of conditional distributions (called conditional ambiguity sets) on a finite scenario tree. The distributions in the conditional ambiguity sets remain close to a nominal conditional distribution according a ø-divergence (e.g., Kullback-Leibler divergence, Hellinger distance, Burg entropy, etc.). Here, the paper discusses a decomposition algorithm to solve the resulting MDRO with ø-divergences, which uses the dual formulation and solves only linear subproblems instead of convex ones. Some properties of the algorithm such as generating feasible policies and valid upper/lower bounds are established. The paper then applies the modeling and solution techniques to allocate water in 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 primary sources of uncertainty in the Tucson region include (1) unpredictable population growth, (2) the availability of water from the Colorado River, and (3) the effects of climate variability on water consumption. This paper integrates forecasts for all these sources of uncertainty into a single optimization model for robust and sustainablemore » water allocation. Then, it uses this model to analyze the value of constructing additional treatment facilities to reduce future water shortages. The results indicate that the MDRO approach can be very valuable for water managers by providing insights to minimize their risks and help them plan for the future.« less

Authors:
 [1]; ORCiD logo [2]
  1. Lawrence Berkeley National Laboratory (LBNL), Berkeley, CA (United States)
  2. The Ohio State Univ., Columbus, OH (United States)
Publication Date:
Research Org.:
Lawrence Berkeley National Laboratory (LBNL), Berkeley, CA (United States)
Sponsoring Org.:
USDOE Office of Science (SC), Advanced Scientific Computing Research (ASCR); National Science Foundation (NSF)
OSTI Identifier:
2282605
Alternate Identifier(s):
OSTI ID: 1907301
Grant/Contract Number:  
AC02-06CH11357; AC02-05CH11231; CMMI-1563504
Resource Type:
Accepted Manuscript
Journal Name:
European Journal of Operational Research
Additional Journal Information:
Journal Volume: 306; Journal Issue: 2; Journal ID: ISSN 0377-2217
Publisher:
Elsevier
Country of Publication:
United States
Language:
English
Subject:
54 ENVIRONMENTAL SCIENCES; OR in environment and climate change; Multistage distributionally robust optimization; Phi-divergences; Water resources; Nested benders decomposition

Citation Formats

Park, Jangho, and Bayraksan, Güzin. A multistage distributionally robust optimization approach to water allocation under climate uncertainty. United States: N. p., 2022. Web. doi:10.1016/j.ejor.2022.06.049.
Park, Jangho, & Bayraksan, Güzin. A multistage distributionally robust optimization approach to water allocation under climate uncertainty. United States. https://doi.org/10.1016/j.ejor.2022.06.049
Park, Jangho, and Bayraksan, Güzin. Mon . "A multistage distributionally robust optimization approach to water allocation under climate uncertainty". United States. https://doi.org/10.1016/j.ejor.2022.06.049. https://www.osti.gov/servlets/purl/2282605.
@article{osti_2282605,
title = {A multistage distributionally robust optimization approach to water allocation under climate uncertainty},
author = {Park, Jangho and Bayraksan, Güzin},
abstractNote = {This paper investigates a Multistage Distributionally Robust Optimization (MDRO) approach to water allocation under climate uncertainty. The MDRO is formed by creating sets of conditional distributions (called conditional ambiguity sets) on a finite scenario tree. The distributions in the conditional ambiguity sets remain close to a nominal conditional distribution according a ø-divergence (e.g., Kullback-Leibler divergence, Hellinger distance, Burg entropy, etc.). Here, the paper discusses a decomposition algorithm to solve the resulting MDRO with ø-divergences, which uses the dual formulation and solves only linear subproblems instead of convex ones. Some properties of the algorithm such as generating feasible policies and valid upper/lower bounds are established. The paper then applies the modeling and solution techniques to allocate water in 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 primary sources of uncertainty in the Tucson region include (1) unpredictable population growth, (2) the availability of water from the Colorado River, and (3) the effects of climate variability on water consumption. This paper integrates forecasts for all these sources of uncertainty into a single optimization model for robust and sustainable water allocation. Then, it uses this model to analyze the value of constructing additional treatment facilities to reduce future water shortages. The results indicate that the MDRO approach can be very valuable for water managers by providing insights to minimize their risks and help them plan for the future.},
doi = {10.1016/j.ejor.2022.06.049},
journal = {European Journal of Operational Research},
number = 2,
volume = 306,
place = {United States},
year = {Mon Jun 27 00:00:00 EDT 2022},
month = {Mon Jun 27 00:00:00 EDT 2022}
}

Works referenced in this record:

Tutorial on risk neutral, distributionally robust and risk averse multistage stochastic programming
journal, January 2021


A stochastic dynamic programming approach to analyze adaptation to climate change – Application to groundwater irrigation in India
journal, March 2018

  • Robert, Marion; Bergez, Jacques-Eric; Thomas, Alban
  • European Journal of Operational Research, Vol. 265, Issue 3
  • DOI: 10.1016/j.ejor.2017.08.029

A time-consistent Benders decomposition method for multistage distributionally robust stochastic optimization with a scenario tree structure
journal, March 2021

  • Yu, Haodong; Sun, Jie; Wang, Yanjun
  • Computational Optimization and Applications, Vol. 79, Issue 1
  • DOI: 10.1007/s10589-021-00266-7

On distributionally robust multiperiod stochastic optimization
journal, July 2014


Mathematical Foundations of Distributionally Robust Multistage Optimization
journal, January 2021

  • Pichler, Alois; Shapiro, Alexander
  • SIAM Journal on Optimization, Vol. 31, Issue 4
  • DOI: 10.1137/21M1390517

Quantifying the Urban Water Supply Impacts of Climate Change
journal, January 2008

  • O’Hara, Jeffrey K.; Georgakakos, Konstantine P.
  • Water Resources Management, Vol. 22, Issue 10
  • DOI: 10.1007/s11269-008-9238-8

The impact of global climate change on water quantity and quality: A system dynamics approach to the US–Mexican transborder region
journal, January 2017

  • Duran-Encalada, J. A.; Paucar-Caceres, A.; Bandala, E. R.
  • European Journal of Operational Research, Vol. 256, Issue 2
  • DOI: 10.1016/j.ejor.2016.06.016

Systematic stress tests with entropic plausibility constraints
journal, May 2013


Decomposition Algorithms for Risk-Averse Multistage Stochastic Programs with Application to Water Allocation under Uncertainty
journal, July 2016

  • Zhang, Weini; Rahimian, Hamed; Bayraksan, Güzin
  • INFORMS Journal on Computing, Vol. 28, Issue 3
  • DOI: 10.1287/ijoc.2015.0684

Data-driven chance constrained stochastic program
journal, July 2015


Spot water markets and risk in water supply
journal, September 2005


Spot water markets and risk in water supply
journal, September 2005


Risk-averse two-stage stochastic programming with an application to disaster management
journal, March 2012


On Solving Multistage Stochastic Programs with Coherent Risk Measures
journal, August 2013

  • Philpott, Andy; de Matos, Vitor; Finardi, Erlon
  • Operations Research, Vol. 61, Issue 4
  • DOI: 10.1287/opre.2013.1175

Entropy based risk measures
journal, August 2020


Maxmin expected utility with non-unique prior
journal, January 1989


Reformulation linearization technique based branch-and-reduce approach applied to regional water supply system planning
journal, March 2015


Distributionally robust SDDP
journal, May 2018

  • Philpott, A. B.; de Matos, V. L.; Kapelevich, L.
  • Computational Management Science, Vol. 15, Issue 3-4
  • DOI: 10.1007/s10287-018-0314-0

Expected Value Minimization in Information Theoretic Multiple Priors Models
journal, June 2018


Evaluating policies in risk-averse multi-stage stochastic programming
journal, May 2014


Distributionally Robust Stochastic Dual Dynamic Programming
journal, January 2020

  • Duque, Daniel; Morton, David P.
  • SIAM Journal on Optimization, Vol. 30, Issue 4
  • DOI: 10.1137/19M1309602

Application of Coherent Risk Measures to Capital Requirements in Insurance
journal, April 1999


Exact Converging Bounds for Stochastic Dual Dynamic Programming via Fenchel Duality
journal, January 2020

  • Leclère, Vincent; Carpentier, Pierre; Chancelier, Jean-Philippe
  • SIAM Journal on Optimization, Vol. 30, Issue 2
  • DOI: 10.1137/19M1258876

Robust Control and Model Uncertainty
journal, May 2001

  • Hansen, Lars Peter; Sargent, Thomas J.
  • American Economic Review, Vol. 91, Issue 2
  • DOI: 10.1257/aer.91.2.60

Near-Optimal Bayesian Ambiguity Sets for Distributionally Robust Optimization
journal, September 2019


Analysis of stochastic dual dynamic programming method
journal, February 2011


Calibration of Distributionally Robust Empirical Optimization Models
journal, September 2021

  • Gotoh, Jun‐ya; Kim, Michael Jong; Lim, Andrew E. B.
  • Operations Research, Vol. 69, Issue 5
  • DOI: 10.1287/opre.2020.2041

Modeling time-dependent randomness in stochastic dual dynamic programming
journal, March 2019


Controlling risk and demand ambiguity in newsvendor models
journal, December 2019

  • Rahimian, Hamed; Bayraksan, Güzin; Homem-de-Mello, Tito
  • European Journal of Operational Research, Vol. 279, Issue 3
  • DOI: 10.1016/j.ejor.2019.06.036

Risk-averse two-stage stochastic programming with an application to disaster management
journal, March 2012


Convexity and concavity properties of the optimal value function in parametric nonlinear programming
journal, January 1986

  • Fiacco, A. V.; Kyparisis, J.
  • Journal of Optimization Theory and Applications, Vol. 48, Issue 1
  • DOI: 10.1007/BF00938592

Robust Dual Dynamic Programming
journal, May 2019

  • Georghiou, Angelos; Tsoukalas, Angelos; Wiesemann, Wolfram
  • Operations Research, Vol. 67, Issue 3
  • DOI: 10.1287/opre.2018.1835

Municipal Groundwater Management: Optimal Allocation and Control of a Renewable Natural Resource
journal, September 2015

  • Murali, Karthik; Lim, Michael K.; Petruzzi, Nicholas C.
  • Production and Operations Management, Vol. 24, Issue 9
  • DOI: 10.1111/poms.12389

Decomposition and Partitioning Methods for Multistage Stochastic Linear Programs
journal, October 1985


Convergence Analysis of Sampling-Based Decomposition Methods for Risk-Averse Multistage Stochastic Convex Programs
journal, January 2016

  • Guigues, Vincent
  • SIAM Journal on Optimization, Vol. 26, Issue 4
  • DOI: 10.1137/140983136

Economic consequences of optimized water management for a prolonged, severe drought in California: ECONOMIC CONSEQUENCES OF PROLONGED SEVERE DROUGHT
journal, May 2010

  • Harou, Julien J.; Medellín-Azuara, Josué; Zhu, Tingju
  • Water Resources Research, Vol. 46, Issue 5
  • DOI: 10.1029/2008WR007681

Multi-stage stochastic optimization applied to energy planning
journal, May 1991

  • Pereira, M. V. F.; Pinto, L. M. V. G.
  • Mathematical Programming, Vol. 52, Issue 1-3
  • DOI: 10.1007/BF01582895

A Data-Driven Model of Virtual Power Plants in Day-Ahead Unit Commitment
journal, November 2019


Measuring Distribution Model risk
journal, October 2013

  • Breuer, Thomas; Csiszár, Imre
  • Mathematical Finance, Vol. 26, Issue 2
  • DOI: 10.1111/mafi.12050

The twenty-first century Colorado River hot drought and implications for the future: COLORADO RIVER FLOW LOSS
journal, March 2017

  • Udall, Bradley; Overpeck, Jonathan
  • Water Resources Research, Vol. 53, Issue 3
  • DOI: 10.1002/2016wr019638

On Distributionally Robust Multiperiod Stochastic Optimization
text, January 2014

  • Analui, Bita; Pflug, Georg Ch.
  • Humboldt-Universität zu Berlin, Mathematisch-Naturwissenschaftliche Fakultät II, Institut für Mathematik
  • DOI: 10.18452/8441

Evaluating the forecast accuracy and bias of alternative population projections for states
journal, November 1992


Adaptive Distributionally Robust Optimization
journal, February 2019

  • Bertsimas, Dimitris; Sim, Melvyn; Zhang, Meilin
  • Management Science, Vol. 65, Issue 2
  • DOI: 10.1287/mnsc.2017.2952

The nested Sinkhorn divergence to learn the nested distance
journal, September 2021


Rectangular Sets of Probability Measures
journal, April 2016


Robust Solutions of Optimization Problems Affected by Uncertain Probabilities
journal, February 2013

  • Ben-Tal, Aharon; den Hertog, Dick; De Waegenaere, Anja
  • Management Science, Vol. 59, Issue 2
  • DOI: 10.1287/mnsc.1120.1641

On the Convergence of Decomposition Methods for Multistage Stochastic Convex Programs
journal, February 2015

  • Girardeau, P.; Leclere, V.; Philpott, A. B.
  • Mathematics of Operations Research, Vol. 40, Issue 1
  • DOI: 10.1287/moor.2014.0664

Decomposition and Partitioning Methods for Multistage Stochastic Linear Programs
journal, October 1985


Maximum-loss, minimum-win and the Esscher pricing principle
journal, October 2011


Drinking Wastewater
journal, April 2012

  • Ormerod, Kerri Jean; Scott, Christopher A.
  • Science, Technology, & Human Values, Vol. 38, Issue 3
  • DOI: 10.1177/0162243912444736

Water Markets and Water Quality
journal, May 1993

  • Weinberg, Marca; Kling, Catherine L.; Wilen, James E.
  • American Journal of Agricultural Economics, Vol. 75, Issue 2
  • DOI: 10.2307/1242912

Centralized versus Decentralized Wastewater Reclamation in the Houghton Area of Tucson, Arizona
journal, May 2013


Quantifying the Urban Water Supply Impacts of Climate Change
journal, January 2008

  • O’Hara, Jeffrey K.; Georgakakos, Konstantine P.
  • Water Resources Management, Vol. 22, Issue 10
  • DOI: 10.1007/s11269-008-9238-8

Data-Driven Stochastic Programming Using Phi-Divergences
book, September 2015