Summary of dissertation structure

accordingly.(Publication IV)

1.6

With a total of six chapters, the first chapter provides an overview about the background of energy security and the importance of its analysis, together with the motivation and objectives. The scope and limitations together with contributions of this research are all presented in the first chapter. Following, the second chapter addresses the current approaches to study energy security from theoretical frameworks through partial and spatial attempts to the limited applications. At the end of the second chapter the needed solutions are presented. Moving forward to the third chapter, methods for designing own energy security framework, its application on the energy sub-system of storage, a soft analysis of energy security for a future scenario for the case of Jordan, building an energy security index, numerical evaluations of energy security on the global level and visualising the results on global maps are all addressed. Once the reader has gone through these three chapters, the fourth chapter presents the four publications that are used to achieve the goals of this dissertation. Moving to the end, discussion about the obtained results are detailed in the fifth chapter whereas conclusions are presented in the sixth chapter.

27

2 Current approaches for energy security studies

Studies that addressed energy security have a huge variation in their intakes on the topic.

As mentioned earlier in the introduction chapter, energy security was analysed from many different angles, including: policymaking (Turton & Barreto, 2006; Winzer, 2012), politics (Jonsson et al., 2015), national energy policies (Franki & Viškovi , 2015), national security issues (Dyer & Trombetta, 2013), international relations (Kirchner &

Berk, 2010) and social sciences (Löschel et al., 2010). Such different starting points in addressing the topic of energy security resulted in different approaches to the topic. Some scholars approached the topic on the framework level, some do the analysis for a certain location, while others tackle parts of the energy system. In this chapter, current approaches for energy security are presented with the state of the art found in literature.

2.1

Since energy security is a concept rather than a strategy or policy (Chester, 2010), it needs to be addressed as such. It was concluded by many researchers that the concept is defined narrowly and disparately (Bohi & Toman, 1993; Kucharski & Unesaki, 2015; Narula &

Reddy, 2015), is not defined clearly (Löschel et al., 2010; Winzer, 2012) or is with no common consensus (Checchi et al., 2009; Kruyt et al., 2009). Therefore it was the first step for researchers when designing a theoretical framework for energy security to decide what the definition of energy security is, although the term of energy security was described before by terms such as abstract, elusive, vague, inherently difficult and blurred (Checchi et al., 2009; Chester, 2010; Löschel et al., 2010; Narula & Reddy, 2016;

Sovacool et al., 2011). Although, defining energy security is the first step in building a framework, definitions have been context-dependent and polysemic in nature (Chester, 2010; Jonsson et al., 2015; Kruyt et al., 2009; Vivoda, 2010) due to various assumptions (Ciut , 2010). To overcome this dilemma, energy security definitions moved to be more generic, seePublication I (pages 3-4).

There were many attempts to identify energy security through research history (Bohi &

Toman, 1993; Cherp & Jewell, 2013; Ciut , 2010; Deese, 1979; Dreyer, 2013; Hossain et al., 2016; Hughes, 2009; IEA, 2001; 2007; Jan & Goldwyn, 2005; Jansen, 2009; Jewell et al., 2014; Johansson & Naki enovi , 2012; Kononov, 2014; Laki , 2013; Lovins &

Lovins, 1981; Miller et al., 1977; Müller-Kraenner, 2007; Narula & Reddy, 2016; Ojeaga, 2014; UNDP, 2000; Willrich, 1976; Winrow, 2009; Yergin, 2006). Most studies provided the definition of energy security by some of its elements, for example, energy security is defined as “The continuous availability of energy in varied forms, in sufficient quantities, and at reasonable prices” by UNDP (2000). However, limiting the definition to certain elements deteriorates any further possible framework design. Therefore, many researchers started to add more elements into their definitions of energy security, for example, energy security was defined as “How to equitably provide available, affordable, reliable, efficient, environmentally benign, proactively governed and socially acceptable energy services to end-users” by Sovacool (2011) and Sovacool et al. (2013a).

With time, the previous approach of including more elements into the definition proved to be incapable of providing the reality of the energy security concept. Therefore, researchers started to limit the definition by providing a reference to these elements, and then, in a next phase of building the framework, these elements were illustrated. An example of this development was noted in recent years by Kanchana & Unesaki (2014), where they defined energy security as “Access to modern energy services”, with further explanations to the elements of access and what services are. Another example was provided by Jewell et al. (2014), where energy security was defined as “Low vulnerability of vital energy systems.” Afterwards, it was their task to determine what is vital for the energy system and what vulnerabilities are expanded on. However, both attempts did not manage to capture the nature of energy security. As can be seen inPublication I (page 5), energy security was defined as “the feature (measure, situation or a status) in which a related system functions optimally and sustainably in all its dimensions, freely from any threats”. This definition was generic enough to account for all elements (dimensions, parts of the system and threats). Afterwards, continuation to identify all these elements can be seen inPublication I.

The next step in building the theoretical framework for energy security analysis, after the definition is proposed, is the formulation of what elements need to be included. Although some researchers would call these elements as boundaries and vulnerabilities (Jewell et al., 2014), the vast majority of all researchers prefer the use of dimensions and parameters as the elements of energy security analysis (Ang et al., 2015; Azzuni & Breyer, 2018;

Chester, 2010; Sovacool & Brown, 2010; Sovacool & Mukherjee, 2011; Winzer, 2012).

A drawback of the approach by Jewell et al. (2014) is the negligence of many important elements of energy security such as environment and military, as well as many others.

Their approach lags behind as Yergin (2006) states that energy security discussion should be extended to all possible dimensions that have a relationship to energy security. Further discussion about what dimensions are included in literature are provided in section 2.3.

Furthermore, all detailed relationships between parameters and dimensions with energy security are well-established and presented inPublication I.

At this point, conceptualisation of the energy security framework is done and some researchers stopped at this depth (Chester, 2010). However, most researchers continued building an analytical framework by proposing numerical indicators to measure, calculate and/or evaluate each of the parameters and dimensions (Ang et al., 2015; Azzuni &

Breyer, 2018). The number of proposed indicators varied, some researchers proposed very few indicators (Badea et al., 2011; Radovanovi et al., 2017), where only eight and six indicators were used, respectively. The most intensive proposal of numerical indicators was done by Sovacool & Mukherjee (2011) in which hundreds of indicators were purposed. Although it is theoretically possible to propose such a high number of indicators, in order to build an applicable and systematic analysis to evaluate energy security, limiting criteria is needed.Publication IV (pages 2-3) provides detailed criteria, which is required when choosing suitable numerical indicators.

2.1 Theoretical framework 29 Once the set of indicators are ready, researchers start the collection of data for such indicators. When all values are prepared, researchers are faced with three main challenges; and their approaches to overcome these challenges vary. The first challenge is the inequal ranges of differing indicators, therefore normalization techniques are required. Through all previous research, normalization was done by varying techniques:

min–max (Gnansounou, 2008; Kamsamrong & Sorapipatana, 2014; Lefèvre, 2010), distance to a reference (USChC, 2012) or standardization (Martchamadol & Kumar, 2012; Sovacool & Brown, 2010). Nevertheless the most common approach is max-min as summarized by Ang et al. (2015).

After normalizations, scholars faced the second challenge of how important each indicator is, how important each parameter is and how important each dimension is. To answer this question, weighting techniques are needed. As found by Ang et al. (2015), the most common technique in literature is equal weighting technique (Onamics, 2005;

Sovacool & Brown, 2010; Sovacool et al., 2011). Other researchers tried other weighting techniques: import/fuel share (Sharifuddin, 2014; WEC, 2014), Principle Component Analysis (PCA) (Gnansounou, 2008; Martchamadol & Kumar, 2012), Analytical Hierarchy Process (AHP) (Wu et al., 2012) and Data Envelopment Analysis (DEA) (Zhang et al., 2013); while some researchers did not use any of these analytical techniques but rather relied on subjective evaluations.

Once weighting for all indicators, parameters and dimensions is done, the third challenge of how to aggregate an index is faced by researchers. Mostly this is done by simple addition. Figure 1 summarizes all three steps of building numerical indexes for energy security frameworks that lead researchers to varying approaches of how to present their results, for example by numbers, clusters, or coloured maps.

Figure 1: Summary techniques to build an energy security analysis framework, adopted and modified from Ang et al. (2015). Abbreviations: Principle Component Analysis (PCA),

Analytical Hierarchy Process (AHP) and Data Envelopment Analysis (DEA).