Techno-Socio-Economic Framework for Energy Storage System Selection in Jordan

Abstract

Renewable energy sources (RESs) are increasingly being recognized as sustainable and accessible alternatives for the energy future. However, their intermittent nature poses significant challenges to system reliability and stability, necessitating the integration of energy storage systems (ESSs) to ensure sustainability and dependability. This study examines various ESS alternatives, evaluating their suitability for different applications using a multi-criteria decision-making (MCDM) approach. The methodology accommodates diverse criteria types, including qualitative and quantitative factors, represented as linguistic terms, interval values, and crisp numerical data. A techno-socio-economic framework for ESS selection is proposed and applied to Jordan’s unique energy landscape. This framework integrates technical performance, economic feasibility, and social considerations to identify suitable ESS solutions aligned with the country’s renewable energy goals. The study ranks twelve energy storage systems (ESSs) based on key performance criteria. Pumped hydro storage (PHS), thermal energy storage (TES), supercapacitors (SCs), and lithium-ion batteries (Li-ion BESS) lead the ranking. These systems showed the best performance in terms of scalability, efficiency, and integration with grid-scale applications in Jordan. Key applications analyzed include renewable energy integration, grid stability, load shifting, peak load regulation, frequency regulation, and seasonal energy storage. Results indicate that Li-ion batteries are most suitable for renewable energy integration, while flywheels excel in grid stability and frequency regulation. PHS was found to be the preferred solution for load shifting, peak load regulation, and seasonal storage, with hydrogen storage emerging as a promising option for long-duration needs. These findings provide critical insights to guide policy and infrastructure planning, offering a robust model for comprehensive ESS assessment in energy transition planning for countries facing similar challenges.

1. Introduction

Countries all over the world started switching to energy generation from alternative energy resources such as photovoltaic and wind turbines due to the undesired climate change, increased price of fossil fuels, and the extensive demand for power. The viability concerns of renewable energy systems (RESs) increase due to their intermittent nature and reliance on uncertain weather conditions. These concerns necessitate looking for reliability measures to sustain the production of RESs. One of the potential solutions to this issue is employing energy storage systems (ESSs). Several ESS technologies are in use today, such as batteries, hydro pump storage, compressed air energy, hydrogen, supercapacitors, and thermal energy storage. These technologies differ from each other in terms of how they operate. For instance, mechanical storage, electrical storage, and electrochemical storage. The applications, along with specific characteristics and concerns for viability, are the major factors to consider when choosing the technology to implement.

Numerous papers have focused on the ESS topic, describing many studies. Reviews of the many kinds of uses and the state of the art in energy storage are conducted in [1,2,3]. Certain recent studies are devoted to discussing specific options for energy storage such as [4,5], where the battery option is considered and studied in terms of issues and measures. Similar studies were conducted for other types such as [6] for thermal energy storage [7], for hydrogen technology, and [8] for pumped hydro energy storage.

In recent studies, certain aspects of ESSs are included, considering the state of the art in each technology in terms of developments and application [9]. Li and Palazzolo [10] explored the recent improvements in flywheel energy storage (FWES), concentrating on the design and performance investigations, as well as the possible applications and the economic aspects. In [11], coupling a supercapacitor with batteries for operating electric vehicles is reviewed. The paper explored the types of power electronic devices suitable to handle the combination. The effectiveness of compressed air technology is presented in [12] with more discussion on the proposed construction of the air and heat storage. In [13] a framework is introduced to assess the techno-economic–environmental performance of integrated multi-vector energy networks incorporating geothermal energy storage. Findings indicate that configurations utilizing high temperature, high penetration of heat pumps, and low temperature are most effective in meeting heat demand efficiently and sustainably.

Hydrogen-based technologies are emerging as a promising solution for long-duration energy storage in power systems, especially to complement variable renewable sources like solar and wind. Hydrogen can store excess electricity generated during low-demand periods and later reconvert it into power using fuel cells or combustion-based systems. As of 2024, hydrogen storage is mainly concentrated in developed countries such as China, Germany, Japan, South Korea, and the United States. These countries have launched demonstration projects and market mechanisms, including auctions, to support hydrogen and ammonia use in electricity generation.

Despite its potential, hydrogen-based electricity generation currently accounts for less than 0.2% of the global power mix (IEA Global Hydrogen Review 2024). A key limitation is the low round-trip efficiency of existing systems. Despite these challenges, technical progress is advancing. In late 2023, South Korea’s Hanwha demonstrated 100% hydrogen firing in an 80 MW gas turbine. Similarly, Japan achieved a 30% hydrogen co-firing share in a 566 MW grid-connected combined-cycle gas turbine mix (IEA Global Hydrogen Review 2024). Hydrogen-based energy storage is gaining global attention as a solution for grid reliability and long-term energy security.

These technologies have undergone technical, social, political, and economical investigations in various studies. These studies focus on the characteristics and impacts of certain technologies, analyzing the feasibility of starting up the adequate ESS for different purposes. In [14], Li-ion batteries technology is reviewed regarding its modeling influence on the technical and economical consideration of the power system analysis. In [15], a model to predict battery degradation has been introduced, which is crucial for enhancing the reliability and efficiency of energy storage in power grids The economic aspects of producing hydrogen as well as utilizing Li-ion ESSs is evaluated by Mayyas et al. [16], concluding that these technologies have competitive costs for the USA’s future.

The evaluation of hydrogen as a fuel and energy carrier considers technical, economic, environmental, social, and political factors, as discussed in [17,18,19]. Other storage technologies are also analyzed individually under varying conditions and criteria in [20,21,22,23,24,25,26]. Given the diverse characteristics, benefits, and limitations of energy storage technologies, selecting the most suitable option is a complex task. To address this, researchers often apply algorithmic methods that weigh both advantages and disadvantages. Multi-criteria decision-making (MCDM) methods are widely adopted for this purpose. They support structured evaluations and have been used in multiple fields, including alternative renewable energy site selection [27], and energy storage system evaluation [28]. MCDM tools help decisionmakers navigate trade-offs and make informed choices. The process becomes even more challenging when accounting for system-specific needs, geographic variations, and contextual priorities. Some studies consider only one type of ESS and employed MCDM for certain issues such as the site selection process as studied in [29] where the optimal site of pumped hydro storage is investigated, and [30] studied the employment of MCDM for compressed air site selection. For different types of ESS, different studies are presented to select one type amongst others. For example, a detailed assessment of battery ESSs using fuzzy-MCDM is introduced in [31] from the views of technical, social, business, environmental, and functionality considering 15 criteria. The findings showed that Li-ion batteries are the best solution, followed by sodium–sulfur (NaS) batteries. For selecting the optimal electrochemical ESS while considering economic factors like the payback period, the study [32] employed a new hybrid MCDM technique combining the Bayesian best–worst method. Çolak et al. [33] took Turkey as a case study to investigate the validity of hesitant fuzzy information MCDM for ESS evaluation. Based on their results, a compressed air ESS is found to be the best alternative for Turkey. The study in [34] utilized the hesitant fuzzy analytic hierarchy process (MCDM-based HF-AHP) and hesitant fuzzy VIKOR methods to assess alternative ESS technologies based on technical, economic, environmental, and social criteria specifically for Oman’s context. The common focus of the listed literature is either assessing specific technology with the inclusion of different characteristics, challenges, and potentials or exploring different types of ESS including a limited number of factors. Further research is required to fill the gap of including further parameters, specifically for future contexts.

Jordan is one of the leading countries in the MENA region in integrating renewable energy resources (RESs). According to annual reports from the Ministry of Energy and Mineral Resources (MEMR), the installed capacity of renewable energy projects connected to the grid reached approximately 2840.2 MW by the end of 2024 (2194.8 MW solar and 622 MW wind) [35]. This includes 1.5 GW from commercial-scale projects and 1.3 GW from net-metering, wheeling, and value-based schemes.

Jordan’s commitment to further increasing renewable energy’s share, mainly solar and wind energy, in its energy mix to 50% by 2030 aligns with global climate and sustainability goals [35]. However, integrating higher levels of solar and wind energy introduces significant challenges to grid stability and reliability in Jordan’s national grid due to their intermittent nature. Moreover, during periods of low electricity demand, renewable energy generation is often curtailed to maintain grid stability and reliability. Without effective storage solutions, surplus energy cannot be utilized efficiently, leading to unnecessary waste and reduced overall system efficiency. To address these challenges, the deployment of advanced and appropriately scaled energy storage systems (ESSs) is essential. ESSs play a pivotal role in managing renewable energy intermittency, minimizing curtailment, and enhancing the resilience of the Jordanian national grid. Furthermore, policymakers are instrumental in developing effective energy strategies. Their decisions must be guided by comprehensive insights from stakeholders, industry experts, and researchers to identify the most optimal and sustainable solutions. This study aims to contribute to this effort by providing valuable data and analysis to support informed policy decisions and drive technological advancements.

This study provides a pioneering contribution by addressing the selection of ESSs through a multi-criteria decision-making (MCDM) approach, specifically tailored to Jordan’s unique geographical, technical, and policy context. To the best of the authors’ knowledge, no prior research has comprehensively analyzed ESS options within this framework for Jordan. By filling this gap, the study not only aids in optimizing energy storage solutions but also serves as a crucial resource for advancing Jordan’s renewable energy ambitions sustainably. Energy storage alternatives and evaluation criteria are identified through an in-depth analysis of relevant literature and expert consultations and inputs. This foundation enables a robust and comprehensive assessment using the technique for order of preference by similarity to ideal solution (TOPSIS) methodology. Energy storage alternatives and evaluation criteria are systematically identified through a thorough review of relevant literature and expert consultations. These inputs are then analyzed using the technique for order of preference by similarity to ideal solution (TOPSIS) method, enabling a structured and precise evaluation process.

The rest of this manuscript contains renewable energy status and storage systems applications and characteristics in Section 2 and the description of the MCDM framework in Section 3. The results and discussion are presented in Section 4.

2. Renewable Energy Status in Jordan and Energy Consumption

Despite Jordan’s minimal fossil fuel production, the country has natural resources that allow it to produce clean energy, increasing self-reliance in electricity generation and helping to cut carbon emissions and combat climate change. Jordan is among the countries facing significant energy challenges driven by rapid demand growth, heavy reliance on fossil fuels, and exposure to unstable international energy interconnections. These pressures have motivated the government to pursue a strategic shift toward renewable energy, with a focus on harnessing wind and solar resources.

Solar and wind energy are the most notable renewable energy sources in Jordan, which, due to its location in the direct solar belt, have tremendous expansion potential. Jordan has abundant renewable energy resources, particularly solar and wind. The country benefits from one of the highest annual average solar irradiance levels in the world, ranging from 4 to 7 kWh/m² per day, making it highly suitable for large-scale solar energy development [36].

Renewable energy regulations have been in effect since 2012, leading to the integration of numerous small MW PV projects into the distribution system. Until 2015, Jordan’s electricity sector was predominantly reliant on conventional sources, specifically natural gas and diesel, constituting over 95% of the energy mix [37]. However, the landscape shifted with the widespread implementation of various small and large PV projects, as illustrated in Figure 1, showcasing the development of renewable energy in Jordan. A significant milestone occurred in September 2015 with the commercial operation of Jordan’s first utility-scale renewable energy project, a 117 MW wind farm [38]. Subsequently, there has been a notable uptrend in the implementation of renewable energy projects across the Kingdom.

The renewable energy sector stands out globally as a success story for Jordan. Jordan’s energy transition represents a major shift toward renewable energy in the Middle East. The study in [39] examines the macroeconomic factors influencing renewable energy production in Jordan. It finds that economic growth, foreign direct investment, and financial development positively contribute, while CO₂ emissions negatively affect output. The results underscore the importance of targeted investment and emission control to advance Jordan’s energy transition.

Since the launch of Jordan’s renewable energy law in 2015, the integration of renewable sources—mainly solar and wind—has accelerated significantly. Figure 1 illustrates the growth in installed capacity for solar PV and wind power. In 2014, renewable energy accounted for less than 1% of the total electricity generation. By the end of 2024, this share increased to 26.9%, with approximately 78% from solar PV and 22% from wind [35]. Notably, a substantial portion of the installed solar PV capacity, reaching nearly 50%, is connected at the distribution system level across medium- and low-voltage networks.

Energy Consumption and Need for Storage

Over the past decade, there has been a 2% increase in demand for electricity. In 2021, the peak load of the electrical system reached 3770 MW in December, compared to 3630 MW in February 2020, reflecting a growth rate of about 3.9% [40]. The average annual growth for the period from 2010 to 2021 was approximately 3.3%. The peak load of the electrical system in 2022 reached 4010 MW in January, representing a growth rate of approximately 6.4%. The average annual growth for the period from 2010 to 2022 was about 3.5% [40].

The primary challenge in operating the national electrical grid is meeting load demand through the optimal and reliable utilization of available resources. Renewable energy sources (RESs) pose a challenge to this objective due to their intermittency, which limits their deep integration into the energy mix. With the increasing penetration of RESs, operations planning becomes more complex, requiring grid operators to redefine unit commitment scenarios. The optimal allocation of resources, considering factors such as system reliability, predictability, and load behavior, becomes more intricate when these resources are dynamic and intermittent [41,42]. Consequently, a significant amount of research has focused on the unit commitment problem, employing optimization algorithms to determine the optimal scheduling of available resources while adhering to network, load, and resource constraints.

Increasing the share of renewable energy in power grids requires keeping conventional generation units online at partial load to cover sudden drops in output. This approach is economically inefficient, as conventional plants operate with lower efficiency and higher fuel consumption when ramping to offset renewable fluctuations [42]. These dynamics complicate generation dispatch planning and raise system costs. Storage solutions offer a practical way to reduce these impacts by smoothing the variability of renewable energy. In Jordan, with a renewable capacity of around 2840 MW—over 78% of which is solar—generation often exceeds 70% of peak daytime demand. This creates a mismatch between supply and demand, leading to frequent curtailment of excess solar output. Energy storage can absorb surplus electricity during the day and release it during evening peaks, reducing curtailment, improving grid flexibility, and minimizing reliance on inefficient part-load operation of conventional plants. Without storage, system operators must derate traditional units to maintain balance, adding hidden operational costs. These challenges underscore the importance of selecting appropriate storage technologies to meet nation grid requirements.

3. Storage Systems Classification, Applications, and Characteristics

This section explores the types of energy storage systems (ESSs), provides various classifications for the available ESSs, and presents energy storage systems applications and characteristics.

3.1. Storage Systems’ Alterative Classifications

Various categorizations have been suggested for ESSs in the literature, and Figure 2 illustrates the most extensive classification. This classification categorizes ESSs based on technology, such as mechanical, electromechanical, thermal, and electrical [1,43]. Additionally, it includes categorizations based on application, such as grid application, renewable integration, microgrids and remote areas, transportation, and industrial and commercial applications [44,45,46,47,48,49,50,51,52,53,54,55,56]. Further divisions are made based on duration, encompassing short, medium, and long durations [57,58], as well as scale-based classifications, including micro, mini, medium, and large scales [43].

3.2. Storage Systems’ Alterative Applications

ESSs are used for different purposes and applications, and this parameter influences decision making and should be considered in the assessment process. Numerous articles have explored and compared these technologies depending on the applications of each type. The main uses of ESSs are frequency regulation (FR), load shifting (LS), peak load regulation (PLR), power quality (PQ), grid integration of large-scale renewable energy (GIRE), and frequency control (FC). The applications of each type based on the literature are summarized in

Table 1. The applications of each type of energy storage system.

Storage Alternative	Applications	Ref.
Pumped hydro storage (PHS)	FR, PLR, black start phase shift, reverse arbitrage, and LS.	[59,60,61]
Compressed air energy storage (CAES)	GIRE, PLR, LS, and intra-day trading.	[59,60]
Flywheel energy storage (FES)	PQ, FR, and non-polluting power supply.	[60,61]
Lead–acid (flooded LA, VRLA)	Peak LS, reserve power supply, transportation, communication, and aviation.	[60]
Lithium-ion (Li-ion)	FR, PLR, generation, transmission, distribution, and stabilizing the renewable energy production arbitrage, intra-day trading, and LS.	[59,60]
Nickel–cadmium	Small portable electrical appliances.	[62]
Flow (vanadium redox (VRFB) zinc bromine (ZnBr))	GIRE, PLR, emergency power supply, the user side.	[60]
High temperature	GIRE, peak LS, PQ.	[60]
Hydrogen	Long-term storage, seasonal fluctuation balancing.	[59]
Supercapacitors	FR, distributed generation, microgrid, and stability of the transient state.	[59,60,61]
Superconducting magnetic energy storage (SMES)	FC, low-frequency oscillation, PQ, system stability.	[59,60,61]
Thermal TES	Large buildings, small-scale heating applications, and energy generation applications.	[63]

.

3.3. Storage Systems’ Alternative Characteristics and Features

Each energy storage technology has unique qualities in the form of benefits and drawbacks. Different characteristics are essential to be known such as response time, energy density, power density, energy efficiency, self-discharge, lifetime, life cycle, social acceptance, power capital cost, energy capital cost, investment costs, and environmental impact. These characteristics vary amongst types. The technology with a short response time might have a longer lifetime and vice versa [64]. Based on available data in the literature, the main characteristics of ESSs have been summarized and classified and are shown in Figure 3. Detailed data and characteristics are listed and shown in

Table 2. Characteristics of different types of energy storage systems.

Storage Technology	Response Time	Energy Density (Wh/Kg)	Power Density (W/Kg)	Energy Efficiency %	Self-Discharge Losses (% Per Day)	Lifetime (Years)	Life Cycle	Social Acceptance	Power Capital Cost (USD/kW)	Energy Capital Cost (USD/kWh)	Investment Costs (€/kWh)	Environmental Impact
PHS	Medium [64], s–min [55], ~1–2 min [61]	0.5, 1.5 [65,66]	0.5–1–1.5 [59], f (h1~h2) kW/m3 [61]	65–75–85 [59], 75–85 [55], 80-82 [67], (65, 75) [68], (75, 80) [69], 70–80 [59], 75–80 [61],	0.0001, 0.0001 [70]	30–40–60 [59] 40–60 [55], 40–50 [59], ~50 Years [61]	10, 16, 50 × 103 [59], >13,000 [55], (>13,000) [67]	Very high [64]	600, 2000 [71], (700, 2000) [72], (500, 4600) [71]	(5, 100) [71], (5, 430) [73]	46–500 [59]	Large (–ve) [55]
CAES	Long [64], Minutes [46], 1–15 min [55], ~1–2 min [61]	30, 60 [65,66], 3.8–5–6 [6], More than PHESS [61]	More than PHESS [61]	41,75 [74], 54–70–88 [59], 70–89 [55], (41, 75) [58], Commercial 48–52 [59], ~70% [61]	0.0001, 0.0002 [72]	20–35–40 [59], 20, 40 [73], <50 Years [67], 30–50 [59]	6, 12, 20 × 103 [59], (>13,000) [67]	Very high [64]	400, 800 [72], 400–1000 [55], 6500–7000 YEN [72]	50–150 [71], 2–120 [55]	3–40–300 [59]	Large (–ve) [55]
FES	Medium [64], Seconds [46], <4 ms–s [55], ~1–2 min [61]	10, 30 [67], 5, 130 [75], 12.5–25–82 [59], >282.7–424 (kW-h/m) [61]	950–1500–6700 [59], >707–1767 [61]	85 [74], (80, 90) [76], 85–86–87 [59], 93–95 [55], (80, 90) [58], >95 [59], 90–95% [61], 85–87 [67]	20, 100 [72]	15–17.5–20 15 + [55], 15,20 [73], 20 [59], 10–20 years (HS), 20 years (LS) [61]	28–60–93 × 103 [59], (>100,000) [55], (>100,000)	Very high [64]	250–350 [71], 250–350 [55], 1700–2000 YEN [59]	1000–5000 [71], 1000–14,000 [55]	537–690–1543 [59]	Almost none [55]
Lead–Acid	Very short [64], <Seconds [46], 5–10 ms [55]	30, 50 [67], 23–33–37 [6], 10–30 [75]	3 to 27 to 53 [59]	(70, 80) [77], (75, 80) [78], 82 LA, 80 VRLA [76], 68–76–90 [59], 70–90 [55], 70–85 [59], 85–90 [67]	0.1, 0.3 [79]	10, 18, 20 [59], 3–15 [55], 3, 12 [73], 5 [59]	1500 [73], 0.3–1.6–1.8 × 103 [59], (2000) [55], 2200–4500 [67]	Medium [64]	300–600 [72], 300–600 [55], 500–1000 YEN [80]	150, 500 [72], 200–400 [55]	179–230–320 [59]	Moderate (–ve) [55]
Li-Ion	Very short [64], Seconds [46], 20 ms–s [55]	(75, 200) [81], (75, 250) [82], 84–115–145 [59]	253–640–1300 [59]	(65, 75) [76], 78 [82], (92, 96, 95, 95) [73], 81–91–98 [59], 85–90 [55], 87–92 [67]	0.1, 0.3 [67], (0.1, 0.05, 0.2, 0.1) [73]	7.5- [59], 5–15 [55], 5, 15 [76]	(2500, 10,000, 1000, 2000) [73], 0.73–2–8 × 103 [59], (1000–20,000) [55], 3200–5800 [59], (>100,000) [67]	High [64]	1200, 4000 [67], 900–4000 [55]	600, 2500 [71,76], (578, 1050, 352,420) [73], 600–3800 [55]	376–484–696 [72]	Moderate (–ve) [55]
Nickel–Cadmium	ms [55]	50, 75 [81]	50,75 [72]	60–65 [55]	0.9, 1.1 [71]	10–20 [55]	(2000–3500) [55]	High [64]	500–1500 [55]	400–2400 [55]	1200$/kWh [46]	Moderate (–ve) [55]
Vrfb	<1 ms [55]	10, 30 [55]	-	~85 [55], 75–80 [59], 65–75 [67]	0.13, 0.17 [73]	5–10 [55], 15 [59]	(12,000+) [55], (>10,000) [67]	High [64]	600–1500 [55], 17,500–19,500 [59]	150–1000 [55]	400$/kWh [72]	Moderate (–ve) [55]
Znbr	<1 ms [55]	30, 50 [55]	1–1.6–2.1 [59]	70 [4], 66–75–85 [59], ~75 [55], 60 [67]	0.13, 0.17 [73]	6.3–15–20 [59], 5–10 [55], 10 [59]	13,000, 1000 [73], 9–10–13.3 × 103 [59], (2000+) [55], (>10,000) [67]	High [64]	700–2500 [55], 12,500–15,000	347, 900 [73], 150–1000 [55]	161–458–860 [72]	Moderate (–ve) [55]
HT	1 ms [55]	120–148–158 [59]	113–160–196 [59]	(80, 84) [83], 75–86–90 [59], 80–90 [59], 87, 75 [67]	0.05, 5 [73]	10–14–17.5 [59], NaS = 10–15 [55], 15 [55,72]	3000, 5000 [73], 2.8–3.6–5.9 * 103 [59], (2500–4500) [55], NaS = (4500)	High [64]	1000–3000 [55]	(399, 368) [71], NaS = 300–500 [55]	172–615–440 [72]	Moderate (–ve) [55]
Hydrogen	Medium [64], Minutes [67], <1 s [55]	800, 1000 [72], 500 + [59]	500 [59]	35, 40 [83], 20–35 [82], 25–58 [73]	0.5, 2 [72]	5–15 [59], 5–20+ [55], 5, 15 [83]	1000 * 103 [54,66], (1000–20,000+) [67]	Medium [64]	500, 10000 [73], 500–10,000 [55]	2, 15 [73], 15 [55]	per kW 10k+ [72]	Small [73]
SC	8 ms [55], Milliseconds [61]	0.1, 15 [72], 5.2–8.7–21.7 [59], >53 (kW-h/m) [61]	1.450–3500–10,000 [59], >176,678 [61]	85, 98 [64], 90–95–97.5 [59], 90–95 [55], <95% [61]	20, 40 [83]	10–15–20 [59], 10, 20 [72], 10–20 Years [61]	21–50–100 * 103 [59]	Medium [64]	100, 300 [83], 100–450 [55]	300, 2000 [46], 300–2000 [55]	570–1468–6800 [72]	None [55]
SMES	<100 ms [55], Milliseconds [61]	0.5, 5 [71], 0.5–5 [59], >7.07 [61]	500–1000 [59], >530 [61]	90, 95 [67], 95–98 [71], 95–98 [55], ~95% [61]	10, 15 [83]	20+ [59], 20+ [55], 20, 30 [72], ~30 Years [61]	100+ * 103 [59], (>100,000) [55]	Medium [64]	200, 300 [83], 200–489 [55]	1000, 10,000 [55], 1000–72,000 [55]	1000–10.000 [72]	Moderate (–ve) [55]
TES	Long [64]	30, 60 [73]	-	14, 18 [72], 30–60 HT TES [59], 50–90 AL-TES [59]	0.05, 1 [83]	AL TES = 10–20 [55], HT TES = 5–15 [55], 5, 15 [73]	- HT TES = (>13,000) [55]	Medium [64]	100, 400 [73]	3, 130 [64]	30–40$/kWh [71]	Small [73]

.

Detailed data and characteristics are listed and shown in Table 2. This table is categorized based on the common characteristics and concerns that should be considered when making the decision of which technology is the best alternative. The table consists of technical characteristics such as response time, energy density, and lifetime as well as economical features like investment and capital cost. Social and environmental concerns are also presented under the categories of social acceptance and environmental impacts, respectively.

4. Methodology and MCDM Framework

The MCDM-TOPSIS algorithm is a powerful tool, which is widely used in decision making under a variety of features and characteristics. The decision is based on the assumption that the optimal option should be the closest to the ideal solution in terms of Euclidean distance (the line segment length connecting two points). This selection methodology has a decision matrix which is firstly created, and, after the matrix has been normalized using a certain normalizing procedure, its values are then multiplied by the weights of the criteria. This method is utilized to properly select one item among the variety of alternatives in different themes such as the energy field [64]. The normal TOPSIS takes crisp numbers and manages the selection based on these values and proposed weights. Several studies used this algorithm in its normal form such as [77].

Other studies use modified versions of TOPSIS. For instance, the extended TOPSIS to accommodate heterogeneous MCDM issues is introduced in [64]. This method is modified by [27] to suit problems with interval numeric values. Mathew et al. [84] developed a study that uses TOPSIS to accommodate interval-based values, in which this method is tested and compared with other algorithms and other previous studies, and the validations show the applicability of this modified algorithm. Other studies carried out a formulation for interval-based TOPSIS as introduced in [85].

The methodology of this study involves creating two precise numerical matrices from an initial matrix of interval numbers to simplify the evaluation process. The approach relies on a distance-based algorithm, where the optimal solution is determined by the shortest distance to the ideal solution. Input data for the algorithm are expressed in both qualitative and quantitative terms, including linguistic descriptions, interval values, and crisp numeric values. Therefore, the first critical step is to unify the input data by converting all forms—crisp numbers and linguistic terms—into intervals, as most criteria are initially presented in an uncertain interval format.

5. Data Preprocessing and Integration

The collected data consist of three types: interval numbers, crisp values, and linguistic terms. Each type requires transformation to ensure consistency in further analysis. Before performing distance measurements, decisionmakers must first represent these values as uncertain data. Using the method proposed by Zhang et al., all three data types can be systematically converted into fuzzy numbers. This transformation ensures a unified representation, allowing for accurate comparison and integration in multi-criteria decision making or fuzzy analysis processes.

5.1. Transformation of Interval Numbers

Each interval number $a_{ij}=\left(a_{ij}^L,a_{ij}^U\right)$ is defined by its lower and upper bounds. Since data units may vary, normalization is required:

$$a_{ij}^L={\displaystyle \frac{a_{ij}^{Ls}}{{\sum }_{i=1}^m\left({\left(a_{ij}^L\right)}^2+{\left(a_{ij}^U\right)}^2\right)}},\hspace{1em}{a}_{ij}^U={\displaystyle \frac{a_{ij}^{Us}}{{\sum }_{i=1}^m\left({\left(a_{ij}^L\right)}^2+{\left(a_{ij}^U\right)}^2\right)}}$$

(1)

Using the Guos method, the normalized interval is converted to a fuzzy number:

$$a_{ij}=\left(h_{ij},g_{ij}\right)\mathrm{where} h_{ij}=a_{ij}^L, \mathrm{and} g_{ij}=1-a_{ij}^U$$

(2)

Membership reflects the average; non-membership reflects the spread.

5.2. Transformation of Crisp Numbers

When attribute values are expressed as crisp (exact) numbers, they need to be converted into a fuzzy form for unified analysis and data standardization. This is performed using normalization and fuzzy mapping. Normalizing the crisp value and mapping the normalized value to a fuzzy number can be performed by the following equation. Through this process, crisp values are consistently transformed into fuzzy numbers, enabling compatibility with other data types in the decision-making framework.

$$a_{ij}^X={\displaystyle \frac{a_{ij}}{\sqrt{{\sum }_{i=1}^m\left({\left(a_{ij}\right)}^2\right)}}}, y_{ij}=\left(h_{ij},g_{ij}\right)\mathrm{where} h_{ij}=a_{ij}^X, \mathrm{and} g_{ij}=1-a_{ij}^X$$

(3)

5.3. Transformation of Linguistic Terms

Linguistic terms (qualitative descriptions), such as “large/high” or “short/low”, are transformed into interval numbers based on predefined mappings provided in

Table 3. Linguistic terms’ transformation into interval numbers.

Linguistic Term	Fuzzy Representation
Extremely High/Extremely Long/Excellent	(1.00, 0.00)
Very High/Very Long/Very Large/Better	(0.90, 0.10)
High/Long/Large/Good	(0.75, 0.20)
Medium/Moderate	(0.50, 0.45)
Low/Short/Small/Bad	(0.35, 0.60)
Very Low/Very Short/Very Small/Worse	(0.10, 0.90)
Extremely Low/None/Worst	(0.00, 1.00)

[59]. To normalize the interval values and standardize values across different scales, vector normalization methods are used. Converting linguistic terms into fuzzy representations transforms qualitative inputs like “low”, “medium”, or “high” into interval-valued fuzzy numbers suitable for processing in fuzzy multi-criteria decision-making (FMCDM) methods such as Fuzzy TOPSIS. Each fuzzy number is represented as a pair (a, b): where a = degree of membership in the positive sense and b = degree of membership in the negative sense. This paper adopts a seven-level linguistic scale, with each term represented by associated fuzzy numbers. Membership and non-membership values are defined based on the average and spread of the corresponding initial interval numbers.

This ensures consistency and comparability across all criteria. Once the input data are unified into interval numbers, the normalized matrix and the weighted normalized matrix are computed. Subsequently, the ideal best and ideal worst values are calculated.

6. Calculate the Normalized Matrix

The second step of the algorithm involves calculating the normalized decision matrix.

In the case of this modified algorithm the decision matrix is interval-based [$N_{ij}^l,N_{ij}^u$]. The interval upper and lower values of the normalized matrix are estimated using Equations (3) and (4) [60,61].

The flow chart of the MCDM-TOPSIS framework is shown in Figure 3.

$$N_{ij}^l={\displaystyle \frac{X_{ij}^l}{\sqrt{{\sum }_{i=1}^m{\left(X_{ij}^l\right)}^2+{\left(X_{ij}^u\right)}^2}}}$$

(4)

$$N_{ij}^u={\displaystyle \frac{X_{ij}^u}{\sqrt{{\sum }_{i=1}^m{\left(X_{ij}^l\right)}^2+{\left(X_{ij}^u\right)}^2 }}}$$

(5)

The second step in the algorithm is to calculate the weighted normalized matrix,

$$u_{ij}^l=N_{ij}^l\times W_j, u_{ij}^u=N_{ij}^u\times W_j;$$

(6)

where $W_j$ is the weight of the criteria.

Subsequently, the positive ideal solution and negative ideal solution are calculated, as presented in Equations (6) and (7).

$$A^{+}=\left[u_1^{+l}+u_1^{+u}\right], \left[u_2^{+l}+u_2^{+u}\right],\dots , \left[u_n^{+l}+u_n^{+u}\right]=\{\underset{i}{(\mathit{max}}{\{u}_{ij}^l, u_{ij}^u\},j\in B) ,\underset{i}{(\mathit{min}}{\{u}_{ij}^l, u_{ij}^u\},j\in N)\}$$

(7)

$$A^{+}=\left[u_1^{-l}+u_1^{+u}\right],\left[u_2^{+l}+u_2^{+u}\right],\dots ,\left[u_n^{+l}+u_n^{+u}\right]=\{\underset{i}{(\mathit{min}}{\{u}_{ij}^l,u_{ij}^u\},j\in B),\underset{i}{(\mathit{max}}{\{u}_{ij}^l,u_{ij}^u\},j\in N)\}$$

(8)

In this context, A denotes the ideal solution, where B represents the set of beneficial criteria and N corresponds to the set of non-beneficial criteria.

The next step involves calculating the distance of each alternative from the positive and negative ideal solutions, and the calculation is as follows [60,61]:

$$S^{+}={\displaystyle \frac{1}{2}}{\sum }_{i=1}^n\left(\left(u_j^{+l}+u_j^{+u}\right)-\left(u_{ij}^l{+u}_{ij}^u\right)\right)+{\displaystyle \frac{1}{2}}{\sum }_{i=1}^n\left(\left(u_{ij}^l+u_{ij}^u\right)-\left(u_j^{+l}{+u}_j^{+u}\right)\right)$$

(9)

$$S^{-}={\displaystyle \frac{1}{2}}{\sum }_{i=1}^n\left(\left(u_{ij}^l{+u}_{ij}^u\right)-\left(u_j^{-l}+u_j^{-u}\right)\right)+{\displaystyle \frac{1}{2}}{\sum }_{i=1}^n\left(\left(u_j^{-l}{+u}_j^{-u}\right)-\left(u_{ij}^l+u_{ij}^u\right)\right)$$

(10)

Finally, determine the relative closeness of the answer to the ideal solution, as shown in Equation (10), where the higher the number the better the alternative.

$$R_i={\displaystyle \frac{S^{-}}{S^{-}+S^{+}}}\hspace{1em}\mathit{where}\hspace{1em}0\le R\ge 1, and i\in \left\{\mathrm{1,2},3,\dots n\right\}$$

(11)

7. Energy Storage Criteria

Selecting the optimal energy storage system (ESS) requires a structured and comprehensive evaluation across multiple dimensions. These are typically categorized into four main criteria: technical, economic, environmental, and social. Based on the existing literature [22,23,24,25,26,27], relevant sub-criteria under each category are identified and illustrated in Figure 4. Technical criteria include response time, energy efficiency, energy density, storage capacity, self-discharge rate, and associated technical risks. Economic criteria encompass capital investment, operating and maintenance costs, and lifecycle cost-effectiveness. Environmental considerations focus on CO₂ emissions intensity, air and water pollution, and land use impacts. Social criteria cover job creation potential, government policy support, public acceptance, and health and safety implications. This classification enables a systematic and balanced approach to ESS evaluation, helping decisionmakers align technology selection with both system requirements and broader sustainability objectives.

As previously mentioned, certain characteristics are considered before making decisions about which is the proper alternative amongst the variety of energy storage options. Depending on the requirements, priorities, and primary concerns and of the authorities and stakeholders, different essential characteristics and weights would be selected.

Table 4. Selected criteria that are considered in this study for Jordan’s context.

	Technical Aspects								Economical Aspects			Environmental Aspects	Social Acceptance
Type	Response Time	Energy Density (Wh/Kg)	Power Density (W/Kg)	Round Efficiency	Self-Discharge Losses (% Per Day)	Lifetime (Years)	Life Cycle	Maturity	Energy Capital Cost (USD/kWh)	Power Capital Cost (USD/kW)	O&M Cost	Overall Environmental Aspects	Social Acceptance	Job Creation
PHS	Long	0.5, 1.5	0.5, 1.5	65, 75	0.0001, 0.0001	30, 50	10,000–16,000	Very High	5100	600–2000	Very Low	Moderate	High	Very High
CAES	Long	30, 60	PHS+	41, 75	0.0001, 0.0001	20, 40	6000–20,000	Moderate	5150	400–1000	Very Low	Moderate	Moderate	High
FES	Extremely Short	10, 30	500–5000	93, 95	20, 100	15, 20	28,000–93,000	Low	1000–5000	250–350	Very High	Low	Low	Moderate
Lead–Acid	Moderate	30, 50	150–400	70, 90	0.1, 0.3	10, 20	2200–4500	High	150–500	500–1000	Medium	High	Moderate	Moderate
Li-Ion	Extremely Short	75, 250	200–2000	92, 98	0.1, 0.3	7.5, 20	3200–5800	Very High	115–800	900–4000	Medium	High	High	High
Nickel–Cadmium	Short	50, 75	80–300	60, 65	0.9, 1.1	10, 20	2000–3500	Very Low	100–300	500–1500	Medium	Very High	Low	Moderate
VRFB	Moderate	10, 30	50–150	75, 80	0.13, 0.17	5, 10	10,000–12,000	Moderate	150–500	600–1500	Low	Moderate	Moderate	High
ZnBr	Moderate	30, 50	50–150	66, 85	0.13, 0.17	6.3, 20	9000–13,000	Moderate	100, 400	700–2500	Low	Moderate	Low	High
HT	Moderate	120, 158	90–230	75, 90	0.05, 5	10, 17.5	2800, 5900	Moderate	200, 500	1000–3000	High	High	Low	High
Hydrogen	Very Long	500, 1000	500+	20, 35	0.5, 2	5, 15	1000, 20,000	Moderate	2, 15	500–10,000	Medium	Moderate	Moderate	Very High
SC	Extremely Short	5.2, 21.7	1000–10,000	90, 97.5	20, 40	20, 30	21,000–100,000	Low	300–4000	100–300	Very Low	Low	Low	Moderate
SMES	Extremely Short	0.5–7.07	1000–10,000	90, 98	10, 15	20, 30	100,000–15,000	Low	1000–10,000	200–300	Extremely High	Low	Low	Moderate
TES	Very Long	30, 60	-	50, 90	0.05, 1	5, 15	13,000–20,000	High	3130	100–400	High	Low	High	High

presents the selected characteristics and their weights as applied in this study, based on Jordan’s needs and context. A new important criterion is proposed in this paper which is job creation and resource accessibility. This criterion is advised since Jordan is a developing country where job possibilities remain a crucial issue that has to be taken into consideration throughout any future planning. There are uncertain values of the characteristics as each paper in the literature introduces certain interval values or descriptions to each criterion. The selected values in this study were decided based on the most repeated value in the literature as well as the most recent publication.

For this study, a panel of 15 experts with extensive and diverse experience in Jordan’s energy and electrical sectors was surveyed and interviewed. The panel included senior engineers, department heads, and technical managers from key stakeholders such as NEPCO, EDCO, SEPCO, and MEMR, covering roles in generation, transmission (grid operation), and distribution. It also featured academics specializing in power and energy systems, as well as experts and professionals from the private sector. Importantly, the panel included energy storage system (ESS) owners and operators, offering practical insights into real-world operations. All participants had over 10 years of professional experience and were carefully selected for their comprehensive knowledge of Jordan’s energy landscape, storage technologies, and sector-specific challenges. The weighting of evaluation criteria was based on input from this diverse expert group. Their broad representation ensured that the assigned weights reflected both national strategic priorities and practical operational requirements.

8. Expert’s Main Criteria Survey Weights Results

Figure 5 illustrates the distribution of weights assigned to four main evaluation criteria—technical, economic, environmental, and social—by 15 experts and the sub-criteria, respectively, where each stacked bar represents an individual expert’s percentage allocation across these criteria.

Table 5. Distribution of experts’ weights.

	Technical Aspects								Economical Aspects			Environmental Aspects	Social Acceptance
Type	Response Time	Energy Density (Wh/Kg)	Power Density (W/Kg)	Round Efficiency	Self-Discharge Losses (% per Day)	Lifetime (Years)	Life Cycle	Maturity	Energy Capital Cost (USD/kWh)	Power Capital Cost (USD/kW)	O&M Cost	Overall Environmental Aspects	Social Acceptance	Job Creation
Expert 1	6	7.5	7	7.5	4.5	3.5	5	9	10	6.7	3.3	25	1	4
Expert 2	3.9	4.2	3.6	6	3	3	3	3.3	15	10	5	20	5	15
Expert 3	3.5	4.2	3.15	8.75	3.5	4.2	4.2	3.5	17.5	11.7	5.8	15	5	10
Expert 4	2.5	3.75	3.5	3.75	2	3	3	3.5	12.5	8.3	4.2	30	8	12
Expert 5	5.85	7.15	7.15	9.75	5.2	11.7	11.7	6.5	10	6.7	3.3	10	2	3
Expert 6	3.5	7	4.2	7	3.5	3.15	3.15	3.5	17.5	11.7	5.8	20	2	8
Expert 7	6.5	7.5	9.5	6.5	5	4	4	7	20	13.3	6.7	5	2.5	2.5
Expert 8	9	6	9	10.2	6.6	6.6	6.6	6	15	10	5	5	2	3
Expert 9	1	5	2	4.4	1.6	2	2	2	10	6.7	3.3	40	7	13
Expert 10	3.9	3.6	4.5	6	3.6	2.7	2.7	3	15	10	5	20	10	10
Expert 11	4	6	4	10	4	4	4	4	15	10	5	15	8	7
Expert 12	6.75	9	9	9	2.25	2.25	2.25	4.5	20	13.3	6.7	10	2	3
Expert 13	8.25	8.8	9.9	9.9	2.75	5.5	5.5	4.4	15	10	5	10	2	3
Expert 14	3	4.5	4.5	4.8	3	3	3	4.2	15	10	5	30	2	8
Expert 15	3.5	5.25	5.25	7	2.8	3.5	3.5	4.2	15	10	5	20	5	10

shows the numerical distribution of expert’s weights and Figure 6 visualizes the experts’ weights across sub-criteria. According to experts, technical criteria receive the highest weight, particularly from one expert (expert no. 8), who assigns them nearly 60%. Economic considerations generally rank second but vary across individuals. Environmental and social aspects show more fluctuation, with some experts assigning them minimal weight. Social criteria remain relatively consistent and lower in importance across the board. The chart highlights how expert judgments differ and converge in multi-criteria evaluation. The average values of these main technical, economic, environmental, and social criteria are 40%, 30%, 18%, and 12%, respectively.

9. Results and Discussion

9.1. Selection Among Different ESS Alteratives

Twelve energy storage alternatives were analyzed and ranked using the selected algorithm. As discussed earlier, linguistic terms were converted into intervals, along with crisp numerical values, to ensure consistency in the evaluation process. A MATLAB version R2021b script was utilized to perform the algorithmic calculations, employing normalization equations outlined in Section 3. The normalized matrix and weighted decision matrix were generated as part of this process, with detailed results presented in the corresponding

Table 6. Normalized decision matrix.

0.1963, 0.17667	0.00043, 0.00128	0.17317, 0.19982	0.000001, 0.000001	0.29629, 0.49381	0.06716, 0.08731	0.19224, 0.17301	0.04963, 0.16544	0.29361, 0.0783	0.13189, 0.22609
0.29446, 0.07852	0.0255, 0.051	0.10923, 0.19982	0.000001, 0.000001	0.19752, 0.39505	0.0403, 0.13432	0.19224, 0.17301	0.03309, 0.08272	0.29361, 0.0783	0.13189, 0.22609
0.1963, 0.17667	0.0085, 0.0255	0.24777, 0.2531	0.17937, 0.89685	0.14814, 0.19752	0.18805, 0.62459	0.19224, 0.17301	0.02068, 0.02895	0.03915, 0.35233	0.03768, 0.33914
0.03926, 0.35335	0.0255, 0.0425	0.18649, 0.23978	0.0009, 0.00269	0.09876, 0.19752	0.01478, 0.03022	0.34603, 0.03845	0.04136, 0.08272	0.19574, 0.17617	0.33914, 0.03768
0.03926, 0.35335	0.06375, 0.21251	0.24511, 0.26109	0.0009, 0.00269	0.07407, 0.19752	0.02149, 0.03895	0.34603, 0.03845	0.07445, 0.33089	0.19574, 0.17617	0.28262, 0.07536
0.13741, 0.23557	0.0425, 0.06375	0.15985, 0.17317	0.00807, 0.00987	0.09876, 0.19752	0.01343, 0.02351	0.28835, 0.07689	0.04136, 0.12408	0.19574, 0.17617	0.28262, 0.07536
0.13741, 0.23557	0.0085, 0.0255	0.19982, 0.21314	0.00117, 0.00152	0.04938, 0.09876	0.06716, 0.08059	0.28835, 0.07689	0.04963, 0.12408	0.19574, 0.17617	0.18841, 0.16957
0.13741, 0.23557	0.0255, 0.0425	0.17584, 0.22646	0.00117, 0.00152	0.06222, 0.19752	0.06044, 0.08731	0.28835, 0.07689	0.05791, 0.20681	0.19574, 0.17617	0.18841, 0.16957
0.13741, 0.23557	0.10201, 0.13431	0.19982, 0.23978	0.00045, 0.04484	0.09876, 0.17283	0.0188, 0.03962	0.28835, 0.07689	0.08272, 0.24817	0.19574, 0.17617	0.18841, 0.16957
0.13741, 0.17667	0.42503, 0.85005	0.05328, 0.09325	0.00448, 0.01794	0.04938, 0.14814	0.00672, 0.13432	0.19224, 0.17301	0.04136, 0.82722	0.13702, 0.23489	0.28262, 0.07536
0.13741, 0.23557	0.00442, 0.01845	0.23978, 0.25976	0.17937, 0.35874	0.19752, 0.29629	0.14104, 0.6716	0.19224, 0.17301	0.00827, 0.02482	0.03915, 0.35233	0.03768, 0.33914
0.29446, 0.07852	0.0255, 0.051	0.13321, 0.23978	0.00045, 0.00897	0.04938, 0.14814	0.08731, 0.13432	0.19224, 0.17301	0.00827, 0.03309	0.13702, 0.23489	0.18841, 0.16957

and

Table 7. Weighted Normalized matrix.

0.00883, 0.00982	0.00013, 0.00004	0.02997, 0.02598	0, 0	0.02469, 0.01481	0.00218, 0.00168	0.02595, 0.02884	0.01654, 0.00496	0.00587, 0.02202	0.04522, 0.02638
0.00393, 0.01472	0.0051, 0.00255	0.02997, 0.01638	0, 0	0.01975, 0.00988	0.00336, 0.00101	0.02595, 0.02884	0.00827, 0.00331	0.00587, 0.02202	0.04522, 0.02638
0.00883, 0.00982	0.00255, 0.00085	0.03796, 0.03717	0.08968, 0.01794	0.00988, 0.00741	0.01561, 0.0047	0.02595, 0.02884	0.0029, 0.00207	0.02642, 0.00294	0.06783, 0.00754
0.01767, 0.00196	0.00425, 0.00255	0.03597, 0.02797	0.00027, 0.00009	0.00988, 0.00494	0.00076, 0.00037	0.00577, 0.0519	0.00827, 0.00414	0.01321, 0.01468	0.00754, 0.06783
0.01767, 0.00196	0.02125, 0.00638	0.03916, 0.03677	0.00027, 0.00009	0.00988, 0.0037	0.00097, 0.00054	0.00577, 0.0519	0.03309, 0.00745	0.01321, 0.01468	0.01507, 0.05652
0.01178, 0.00687	0.00638, 0.00425	0.02598, 0.02398	0.00099, 0.00081	0.00988, 0.00494	0.00059, 0.00034	0.01153, 0.04325	0.01241, 0.00414	0.01321, 0.01468	0.01507, 0.05652
0.01178, 0.00687	0.00255, 0.00085	0.03197, 0.02997	0.00015, 0.00012	0.00494, 0.00247	0.00201, 0.00168	0.01153, 0.04325	0.01241, 0.00496	0.01321, 0.01468	0.03391, 0.03768
0.01178, 0.00687	0.00425, 0.00255	0.03397, 0.02638	0.00015, 0.00012	0.00988, 0.00311	0.00218, 0.00151	0.01153, 0.04325	0.02068, 0.00579	0.01321, 0.01468	0.03391, 0.03768
0.01178, 0.00687	0.01343, 0.0102	0.03597, 0.02997	0.00448, 0.00004	0.00864, 0.00494	0.00099, 0.00047	0.01153, 0.04325	0.02482, 0.00827	0.01321, 0.01468	0.03391, 0.03768
0.00883, 0.00687	0.08501, 0.0425	0.01399, 0.00799	0.00179, 0.00045	0.00741, 0.00247	0.00336, 0.00017	0.02595, 0.02884	0.08272, 0.00414	0.01762, 0.01028	0.01507, 0.05652
0.01178, 0.00687	0.00184, 0.00044	0.03896, 0.03597	0.03587, 0.01794	0.01481, 0.00988	0.01679, 0.00353	0.02595, 0.02884	0.00248, 0.00083	0.02642, 0.00294	0.06783, 0.00754
0.00393, 0.01472	0.0051, 0.00255	0.03597, 0.01998	0.0009, 0.00004	0.00741, 0.00247	0.00336, 0.00218	0.02595, 0.02884	0.00331, 0.00083	0.01762, 0.01028	0.03391, 0.03768

.

Table 8. Ranking of types of ESSs for Jordan scenario.

ESS Type	Si+	Si-	Ri (Score)	Rank
PHS	0.224652298	0.430474256	0.657085647	1
TES	0.233827828	0.421298727	0.643079912	2
SC	0.240520503	0.414606052	0.63286406	3
Li-Ion	0.255242612	0.399883943	0.6103919	4
CAES	0.280771504	0.374355051	0.571424022	5
VRFB	0.291821804	0.36330475	0.554556593	6
ZnBr	0.304998697	0.350127858	0.534443086	7
Hydrogen	0.308812751	0.346313804	0.528621228	8
Lead–Acid	0.316584142	0.338542413	0.516758801	9
FES	0.32263807	0.332488485	0.507517948	10
SMES	0.323441159	0.331685395	0.506292094	11
HT	0.334288968	0.320837587	0.489733754	12
Nickel–Cadmium	0.345504783	0.309621772	0.47261368	13

summarizes the final ranking of the energy storage systems (ESSs). The evaluation results show that pumped hydro storage (PHS) ranks highest among all energy storage systems, with a relative closeness score (Ri) of 0.657. This indicates its strong suitability for large-scale grid applications, particularly where geographic and hydrological conditions are favorable. Thermal energy storage (TES) follows closely with an Ri of 0.643, making it effective for applications that integrate heating and cooling loads. Supercapacitors (SCs) come in third, scoring 0.633, due to their excellent performance in rapid charge–discharge cycles and high power density, which suits short-term support and frequency regulation.

Lithium-ion (Li-ion) batteries occupy the middle ground with an Ri of 0.610. Their versatility, maturity, and decreasing costs support wide adoption in both small- and medium-scale applications. Compressed air energy storage (CAES) and vanadium redox flow batteries (VRFBs) rank fifth and sixth, respectively, offering medium- to long-duration storage capabilities, though both face practical constraints such as infrastructure needs and material costs. Technologies like zinc bromine (ZnBr), hydrogen storage, and lead–acid batteries fall into the lower-middle tier, with Ri values between 0.516 and 0.534. These options may still be viable in specific contexts but present limitations in efficiency, environmental impact, or lifecycle.

The lowest-ranked systems include flywheel energy storage (FES), superconducting magnetic energy storage (SMES), high-temperature (HT) batteries, and nickel–cadmium batteries, all scoring below 0.51. These systems are generally less favored due to high costs, limited scalability, or environmental concerns. Overall, PHS, TES, and SCs stand out as the most favorable options for Jordanian systems based on current expert-weighted evaluation criteria, while systems like SMES and nickel–cadmium lag behind due to economic and technological limitations.

9.2. Sensitivity Analysis and Selection Based on Applications

Figure 7 shows the ranking of selected experts. The rankings provided by six experts show strong consensus on the top-performing energy storage technologies. Pumped hydro storage (PHS), thermal energy storage (TES), and supercapacitors (SCs) consistently received top rankings, reflecting their technical maturity, reliability, and suitability for either large-scale or high-power applications. PHS was ranked in the top three by every expert, highlighting its established role in grid-scale storage. SCs also received multiple first-place rankings due to their exceptional power density and fast response time. TES was similarly valued for its cost-effectiveness and long-duration potential. Lithium-ion batteries were moderately well-rated, with experts recognizing their widespread use and performance benefits. However, they were marked slightly lower by some, possibly due to concerns about cost, degradation, or safety in large deployments. CAES held a consistent middle ranking across experts, signaling cautious optimism about its scalability and cost profile. Technologies like SMES, FES, and hydrogen storage showed more divergent opinions. For example, SMES received both high and low scores, reflecting differing views on its practicality and cost. Hydrogen’s variable ranking indicates differing perspectives on its current maturity versus long-term promise. Flow batteries (VRFB and ZnBr), high-temperature (HT) systems, and legacy chemistries like nickel–cadmium and lead–acid consistently ranked near the bottom. Experts appear to view them as less viable due to cost, complexity, environmental concerns, or limited commercial deployment.

Energy storage systems (ESSs) have a wide range of applications, as discussed previously. Based on the ESS classification, an algorithm was applied to each category individually to rank ESS options for each application. The TOPSIS methodology was employed to calculate ESS scores, considering expert input supported by a thorough literature review. The key applications evaluated include renewable energy integration, grid stability, load shifting and peak load regulation, frequency regulation, and seasonal energy storage.

Figure 8 illustrates the ranking scores of ESSs for each application. The findings highlight the primary uses of ESSs, categorizing each application with its most suitable ESS options. For renewable energy integration, Li-ion batteries ranked highest, followed by pumped hydro storage (PHS). In terms of grid stability, flywheels scored highest, followed by Li-ion batteries. For load shifting and peak load regulation, PHS was the top choice, followed by compressed air energy storage (CAES). In frequency regulation, flywheels were ranked first, followed by Li-ion batteries. Finally, for seasonal energy storage, hydrogen achieved the highest rank, followed by PHS. These insights provide a comprehensive overview of the optimal ESS solutions for various applications.

10. Conclusions

With the planned adoption of clean energy and the pivotal role of energy storage in fostering this transition, energy storage systems (ESSs) play a critical role in supporting the roadmap toward green energy adoption. In this study, a techno-socio-economic framework for energy storage system selection in Jordan has been developed and analyzed. This framework integrates technical performance, economic feasibility, and social considerations to identify the most suitable ESS for various applications in Jordan’s unique energy landscape. By aligning storage solutions with the country’s renewable energy goals, this approach ensures a sustainable and reliable pathway for green energy integration.

The study’s results provide a comprehensive ranking of twelve energy storage systems (ESSs), offering critical insights into their suitability for various applications. Pumped hydro energy storage, thermal energy storage (TES), supercapacitors (SCs), and Li-ion battery energy storage systems (BESSs) emerged as the top-ranked technologies due to their scalability, efficiency, and compatibility with modern energy systems. Compressed air energy storage (CAES) and vanadium redox flow batteries (VRFBs) followed in third and fourth places, demonstrating their value in specific use cases. This ranking reflects the strengths and limitations of each technology, highlighting their potential roles in supporting renewable energy adoption and grid modernization. In addition to the general ranking, the study also evaluated ESS options based on specific applications using the TOPSIS methodology. Key applications considered included renewable energy integration, grid stability, load shifting and peak load regulation, frequency regulation, and seasonal energy storage. The analysis revealed that Li-ion batteries were the most suitable choice for renewable energy integration, while flywheels performed best for grid stability and frequency regulation. PHS was the preferred solution for load shifting, peak load regulation, and seasonal energy storage, with hydrogen emerging as a promising option for long-duration seasonal storage. These findings underscore the importance of aligning ESS technologies with their intended applications to achieve optimal performance and support the transition to a sustainable energy future.