Building conflict uncertainty into electricity planning: A South Sudan case study

https://doi.org/10.1016/j.esd.2019.01.003Get rights and content

Highlights

  • We explore South Sudan electricity planning strategies under conflict uncertainty.

  • Stochastic optimization is used to produce a near-term hedging strategy.

  • Analysis considers key uncertainties that could affect planning outcomes.

  • Conflict hedging has the least economic value under the most extreme scenarios.

  • Solar photovoltaics may be an attractive hedge against future conflict.

Abstract

This paper explores electricity planning strategies in South Sudan under future conflict uncertainty. A stochastic energy system optimization model that explicitly considers the possibility of armed conflict leading to electric power generator damage is presented. Strategies that hedge against future conflict have the greatest economic value in moderate conflict-related damage scenarios by avoiding expensive near-term investments in infrastructure that may be subsequently damaged. Model results show that solar photovoltaics can play a critical role in South Sudan's future electric power system. In addition to mitigating greenhouse gas emissions and increasing access to electricity, this analysis suggests that solar can be used to hedge against economic losses incurred by conflict. While this analysis focuses on South Sudan, the analytical framework can be applied to other conflict-prone countries.

Introduction

Electricity supply security is critically important, especially in fragile and conflict-affected states where resumption of electricity services can restore confidence in the government and society, strengthen security, and revive the economy (World Bank, 2013). Addressing fragility, conflict, and violence (FCV) is required to end poverty and promote shared prosperity (World Bank, 2015). While the provision of affordable and reliable electricity supply can promote economic development and help countries exit the conflict trap (Collier, 2003), electric power systems are also vulnerable to conflict conditions. Attackers in many conflict environments have targeted electricity transmission lines and power generation plants, which can lead to long outages and the need for system restoration (Zerriffi, Dowlatabadi, & Strachan, 2002).

Acknowledging that each conflict has its own unique dynamics (Goldstone, 2008), recommendations should be based on a thorough examination of specific conflict situations. In this paper, we explore potential electricity development pathways in South Sudan. South Sudan has been ranked as the most fragile country in the world for the last several years (Fund for Peace, 2017), and it is also one of the least developed countries in the world. There are approximately 250 km of paved roads and less than 30 MW of installed electric generating capacity serving 13 million people in a landlocked area slightly smaller than the US state of Texas (CIA, 2018). Soon after South Sudan gained its independence in 2011, the government started to attract investment funding for hydropower installations (IEA, 2014). Two years later, in 2013, a civil war erupted and it is still ongoing despite a peace agreement signed in 2015 (The Guardian, 2016). To the best of our understanding, most of the investments in the electricity infrastructure expansion have been suspended. Despite having an abundance of natural resources, conflict in South Sudan makes the country prone to economic collapse (World Bank, 2016).

Electrification strategies under FCV conditions should explicitly consider the risk of conflict in the decision making process. However, this is often not the case. For example, EAPP (2011) examined future electricity development by employing a conventional least-cost capacity planning model and concluded that South Sudan should focus on developing a series of large-scale hydroelectric dams along the White Nile. Political issues were considered, but only exogenously to the optimization model. Such a focus on large scale infrastructure projects with long construction times produces inefficient outcomes. These hydroelectric projects never broke ground, and instead hundreds of millions of dollars have been spent on generators and diesel fuel (Mozersky & Kammen, 2018). While incorporating conflict risk in energy system planning is challenging and subject to considerable uncertainty, it should not be ignored (Bazilian & Chattopadhyay, 2016).

This paper focuses on developing planning strategies for the South Sudan electric power system that explicitly consider conflict uncertainty. We model the South Sudan system using an open source energy system optimization model, and incorporate conflict by performing multi-stage stochastic optimization (Birge & Louveaux, 2011; Pereira & Pinto, 1991; de Queiroz, 2016). Optimization is performed over a scenario tree that represents different conflict-related outcomes in the future, and the resultant stochastic solution suggests a near-term planning strategy. Given the paucity of data and large future uncertainties, we perform sensitivity analysis to identify critical assumptions and develop insights that explicitly consider conflict-related uncertainty.

While the application of stochastic optimization yields a planning strategy, this analysis should nonetheless be viewed as an exercise to explore the decision space when conflict is explicitly considered. We are not able to capture all of the real-world conflict dynamics and potential power system failure modes. In addition, we emphasize that models alone cannot provide a solution in such complex decision landscapes, but can yield insight that informs decision making. This paper is intended to further the discussion between modelers and the decision makers, planners, and consultants who develop electrification strategies in FCV countries.

The results presented here suggest promise for further application. Much of the analysis focused on energy development in Africa has been focused on universal access and climate change mitigation through renewables deployment (Lucas, Dagnachew, & Hof, 2017; Africa Progress Report, 2015; AREI, 2017; Wu et al., 2017; Deichmann, Meisner, Murray, & Wheeler, 2011). Considering conflict-related uncertainty can add another dimension to future analysis, ensuring that energy supply is also resilient in the face of conflict, fragility, and violence.

Section snippets

Methods

Key aspects of the modeling effort are described in this section. We begin by describing Tools for Energy Model Optimization and Analysis (Temoa), the open source energy system optimization model and the South Sudan input dataset used to conduct this work. Next, we describe Method of Morris, a sensitivity analysis technique that allows us to identify the input parameters with the largest effect on total system cost. Then we describe the stochastic model formulation, the method by which

Results and discussion

We begin by describing the Method of Morris results, which inform the analysis performed with the stochastic version of the model. Next, results from the stochastic optimization are presented under different assumed conflict scenarios. We conclude by discussing the stochastic output metrics – EVPI, VSS and ECIC – and use them to draw insights about future electric power development in South Sudan.

Conclusions

Fourteen of the top 20 most fragile countries are in sub-Saharan Africa (Fund for Peace, 2017). This modeling exercise demonstrates the need for such countries to explicitly consider the risk of conflict as they build out their electric power systems. We construct an analytical framework that employs an energy system optimization model along with sensitivity analysis and stochastic optimization to examine how potential future conflict can affect near-term electricity planning. We apply this

Caveats and future work

As mentioned in the introduction, this analysis explores a simplified decision landscape that includes the explicit consideration of conflict in an electricity expansion planning exercise. Given the complexity of real world conflict dynamics, we made a number of simplifying assumptions.

First, we assume an exogenously specified demand increase over time, with demand curtailment occurring at a prescribed cost. The use of a curtailment cost – above which electricity is no longer demanded – is a

References (51)

  • AfDB

    South Sudan: An infrastructure action plan. A program for sustained strong economic growth

  • Africa Progress Report

    Power, people and planet: Seizing Africa's energy and climate opportunities

  • AREI

    Africa renewable energy initiative: Progress report

  • J.R. Birge

    The value of the stochastic solution in stochastic linear programs with fixed recourse

    Mathematical Programming

    (1982)
  • J.R. Birge et al.

    Introduction to stochastic programming

    (2011)
  • D.G. Cacuci et al.

    A comparative review of sensitivity and uncertainty analysis of large-scale systems-II: statistical methods

    Nuclear Science and Engineering

    (2004)
  • CIA

    The World Factbook: South Sudan

  • P. Collier

    Breaking the conflict trap: Civil war and development policy

    (2003)
  • G.B. Dantzig

    Linear programming under uncertainty

  • EAPP

    Final master plan report. Volume II: Regional power system master plan and grid code study

    (2011)
  • EIA

    Updated capital cost estimates for utility scale electricity generating plants

    (2013)
  • L.F. Escudero et al.

    The value of the stochastic solution in multistage problems

    TOP

    (2007)
  • L.G. Fishbone et al.

    Markal, a linear-programming model for energy systems analysis: Technical description of the bnl version

    International Journal of Energy Research

    (1981)
  • Fund for Peace

    Fragile States Index

  • J.A. Goldstone

    Pathways to state failure

    Conflict Management and Peace Science

    (2008)
  • Cited by (14)

    • Optimal design of low-carbon energy systems towards sustainable cities under climate change scenarios

      2022, Journal of Cleaner Production
      Citation Excerpt :

      By using a set of potential scenarios, SO is able to capture various uncertainties pertinent to energy systems, including but not limited to climate change. McCollum et al. assessed how uncertainties in fuel prices influenced carbon emissions (McCollum et al., 2016), while Spyrou et al. and Pantankar et al. considered conflict-related risks and consequences on power system planning in polity-fragile regions (Patankar et al., 2019; Spyrou et al., 2019). The impacts of climate uncertainties on urban energy systems has been widely studied in recent years (Ioannou et al., 2017; Mavromatidis et al., 2018b; Firouzmakan et al., 2019).

    • Using robust optimization to inform US deep decarbonization planning

      2022, Energy Strategy Reviews
      Citation Excerpt :

      Inputs to ESOMs include existing technology capacity, capital costs, fixed and variable operational costs, conversion efficiencies, and emissions coefficients; outputs include installed technology capacity, energy consumed and produced by each technology, energy prices, and emissions. The insights obtained from ESOMs are often used to inform policy decisions, and thus energy models should account for uncertainty that can affect policy outcomes [2–6]. While a variety of methods can be used to quantify model sensitivity and the effects of future uncertainty [7], most of these methods require the selection of uncertain values prior to the model run, and thus represent a learn-then-act approach [8].

    • The role of regulatory, market and governance risk for electricity access investment in sub-Saharan Africa

      2021, Energy for Sustainable Development
      Citation Excerpt :

      The latter research has focused largely on techno-economic dimensions (Dagnachew et al., 2017; Ellman, 2015; Mentis et al., 2017; Moner-Girona et al., 2019), without explicitly encapsulating the heterogeneity in the investment environment quality. The only notable exercises that have begun taking this direction are the recent work in Korkovelos et al. (2020), Spyrou, Hobbs, Bazilian, and Chattopadhyay (2019), Patankar, de Queiroz, DeCarolis, Bazilian, and Chattopadhyay (2019), who modelled the role of conflict in electrification and power systems planning. The remainder of the paper is structured as follows.

    View all citing articles on Scopus
    View full text