Computational decision framework for enhancing resilience of the energy, water and food nexus in risky environments

Rajesh Govindan, Tareq Al-Ansari

Research output: Contribution to journalReview article

Abstract

The energy, water and food (EWF) nexus modelling and analysis frameworks proposed recently have demonstrated their effectiveness in the assessment and quantification of synergies and trade-offs in the interlinkages between the three sectors. They largely rely on static, deterministic or equilibrium-based models that facilitate in making decisions for well-behaved and predictable resource systems over time. These frameworks, however, are partly limited in their functionality due to the fact that they do not consider the exposure of systems to the dynamic nature of extrinsic uncertainties and the associated risks in the nexus. Hence, there is a need for a sequential learning, planning and optimal control modelling framework which could help achieve adaptive systems under volatile conditions with the objective to maximise economic output and enhance their operational resilience. In this paper, the authors discuss the development of a novel computational framework which incorporates “algorithmic resilience thinking” to achieve adaptive and robust inter-networked systems. Here, the question of adaptive systems for EWF nexus resilience is posed as a reinforcement learning problem based on sequential decision-making called the Markov decision process (MDP). The authors further discuss a case study, considering weather volatility, its spatial impact on vegetation, and the consequent risks on the water-food nexus for outdoor agricultural operations in the State of Qatar. The application of the developed framework particularly demonstrates promise in providing the functionality to track and mitigate emerging risks that have the potential to cause unprecedented disruption in the operations of integrated natural resource systems. The outcome of this study has positive implications for the advancement and effectiveness of EWF nexus planning and risk management to avert resource shortages and price risks, socio-economic disruption, and cascading failures of critical infrastructures, particularly when the global supply chains are subjected to stresses and shocks, such as extreme weather conditions.

Original languageEnglish
Pages (from-to)653-668
Number of pages16
JournalRenewable and Sustainable Energy Reviews
Volume112
DOIs
Publication statusPublished - 1 Sep 2019

Fingerprint

Adaptive systems
Water
Decision making
Planning
Critical infrastructures
Economics
Reinforcement learning
Natural resources
Risk management
Supply chains
Uncertainty

Keywords

  • Artificial intelligence
  • Energy, water, and food nexus
  • Regime switching
  • Reinforcement learning
  • Risk management

ASJC Scopus subject areas

  • Renewable Energy, Sustainability and the Environment

Cite this

Computational decision framework for enhancing resilience of the energy, water and food nexus in risky environments. / Govindan, Rajesh; Al-Ansari, Tareq.

In: Renewable and Sustainable Energy Reviews, Vol. 112, 01.09.2019, p. 653-668.

Research output: Contribution to journalReview article

@article{105189f64364412783631acc260492df,
title = "Computational decision framework for enhancing resilience of the energy, water and food nexus in risky environments",
abstract = "The energy, water and food (EWF) nexus modelling and analysis frameworks proposed recently have demonstrated their effectiveness in the assessment and quantification of synergies and trade-offs in the interlinkages between the three sectors. They largely rely on static, deterministic or equilibrium-based models that facilitate in making decisions for well-behaved and predictable resource systems over time. These frameworks, however, are partly limited in their functionality due to the fact that they do not consider the exposure of systems to the dynamic nature of extrinsic uncertainties and the associated risks in the nexus. Hence, there is a need for a sequential learning, planning and optimal control modelling framework which could help achieve adaptive systems under volatile conditions with the objective to maximise economic output and enhance their operational resilience. In this paper, the authors discuss the development of a novel computational framework which incorporates “algorithmic resilience thinking” to achieve adaptive and robust inter-networked systems. Here, the question of adaptive systems for EWF nexus resilience is posed as a reinforcement learning problem based on sequential decision-making called the Markov decision process (MDP). The authors further discuss a case study, considering weather volatility, its spatial impact on vegetation, and the consequent risks on the water-food nexus for outdoor agricultural operations in the State of Qatar. The application of the developed framework particularly demonstrates promise in providing the functionality to track and mitigate emerging risks that have the potential to cause unprecedented disruption in the operations of integrated natural resource systems. The outcome of this study has positive implications for the advancement and effectiveness of EWF nexus planning and risk management to avert resource shortages and price risks, socio-economic disruption, and cascading failures of critical infrastructures, particularly when the global supply chains are subjected to stresses and shocks, such as extreme weather conditions.",
keywords = "Artificial intelligence, Energy, water, and food nexus, Regime switching, Reinforcement learning, Risk management",
author = "Rajesh Govindan and Tareq Al-Ansari",
year = "2019",
month = "9",
day = "1",
doi = "10.1016/j.rser.2019.06.015",
language = "English",
volume = "112",
pages = "653--668",
journal = "Renewable and Sustainable Energy Reviews",
issn = "1364-0321",
publisher = "Elsevier Limited",

}

TY - JOUR

T1 - Computational decision framework for enhancing resilience of the energy, water and food nexus in risky environments

AU - Govindan, Rajesh

AU - Al-Ansari, Tareq

PY - 2019/9/1

Y1 - 2019/9/1

N2 - The energy, water and food (EWF) nexus modelling and analysis frameworks proposed recently have demonstrated their effectiveness in the assessment and quantification of synergies and trade-offs in the interlinkages between the three sectors. They largely rely on static, deterministic or equilibrium-based models that facilitate in making decisions for well-behaved and predictable resource systems over time. These frameworks, however, are partly limited in their functionality due to the fact that they do not consider the exposure of systems to the dynamic nature of extrinsic uncertainties and the associated risks in the nexus. Hence, there is a need for a sequential learning, planning and optimal control modelling framework which could help achieve adaptive systems under volatile conditions with the objective to maximise economic output and enhance their operational resilience. In this paper, the authors discuss the development of a novel computational framework which incorporates “algorithmic resilience thinking” to achieve adaptive and robust inter-networked systems. Here, the question of adaptive systems for EWF nexus resilience is posed as a reinforcement learning problem based on sequential decision-making called the Markov decision process (MDP). The authors further discuss a case study, considering weather volatility, its spatial impact on vegetation, and the consequent risks on the water-food nexus for outdoor agricultural operations in the State of Qatar. The application of the developed framework particularly demonstrates promise in providing the functionality to track and mitigate emerging risks that have the potential to cause unprecedented disruption in the operations of integrated natural resource systems. The outcome of this study has positive implications for the advancement and effectiveness of EWF nexus planning and risk management to avert resource shortages and price risks, socio-economic disruption, and cascading failures of critical infrastructures, particularly when the global supply chains are subjected to stresses and shocks, such as extreme weather conditions.

AB - The energy, water and food (EWF) nexus modelling and analysis frameworks proposed recently have demonstrated their effectiveness in the assessment and quantification of synergies and trade-offs in the interlinkages between the three sectors. They largely rely on static, deterministic or equilibrium-based models that facilitate in making decisions for well-behaved and predictable resource systems over time. These frameworks, however, are partly limited in their functionality due to the fact that they do not consider the exposure of systems to the dynamic nature of extrinsic uncertainties and the associated risks in the nexus. Hence, there is a need for a sequential learning, planning and optimal control modelling framework which could help achieve adaptive systems under volatile conditions with the objective to maximise economic output and enhance their operational resilience. In this paper, the authors discuss the development of a novel computational framework which incorporates “algorithmic resilience thinking” to achieve adaptive and robust inter-networked systems. Here, the question of adaptive systems for EWF nexus resilience is posed as a reinforcement learning problem based on sequential decision-making called the Markov decision process (MDP). The authors further discuss a case study, considering weather volatility, its spatial impact on vegetation, and the consequent risks on the water-food nexus for outdoor agricultural operations in the State of Qatar. The application of the developed framework particularly demonstrates promise in providing the functionality to track and mitigate emerging risks that have the potential to cause unprecedented disruption in the operations of integrated natural resource systems. The outcome of this study has positive implications for the advancement and effectiveness of EWF nexus planning and risk management to avert resource shortages and price risks, socio-economic disruption, and cascading failures of critical infrastructures, particularly when the global supply chains are subjected to stresses and shocks, such as extreme weather conditions.

KW - Artificial intelligence

KW - Energy, water, and food nexus

KW - Regime switching

KW - Reinforcement learning

KW - Risk management

UR - http://www.scopus.com/inward/record.url?scp=85067255590&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=85067255590&partnerID=8YFLogxK

U2 - 10.1016/j.rser.2019.06.015

DO - 10.1016/j.rser.2019.06.015

M3 - Review article

VL - 112

SP - 653

EP - 668

JO - Renewable and Sustainable Energy Reviews

JF - Renewable and Sustainable Energy Reviews

SN - 1364-0321

ER -