Analysis of Barriers and Strategies to the Integration of Renewable Energy in South Africa: A Hybrid Multi-Criteria Decision-Making Framework

Abstract

Renewable energy sources are fast becoming the most cost-effective option for adding new power generation capacity globally. In South Africa (SA), the transition from fossil fuels to renewable energy has steadily gained momentum over the years. However, this transition is beset by complex and multidimensional barriers. This research study analyses and prioritises renewable energy barriers and mitigation strategies in South Africa. The DEMATEL multi-criteria decision-making technique was employed to rank the barriers and assess their cause-and-effect relationships. The findings reveal the top three barrier categories as Agreement, Market, and Knowledge. The study further employed an integrated hybrid CRITIC-TOPSIS technique to prioritise the proposed mitigation strategies for each barrier in a defined category. The results indicate that strengthening local community engagement is the most suitable solution to the adoption of renewable energy in SA. A sensitivity analysis model was conducted to validate the robustness of the results. The findings validate the consistency of the methods, with the ranking of the barriers and mitigation strategies remaining stable under various scenarios. This study presents a context-specific causal analysis of barriers and an objective prioritisation of mitigation strategies in South Africa using an integrated hybrid DEMATEL and CRITIC–TOPSIS approach, providing policymakers and decision-makers with valuable insights to develop strategic plans and policies that address the identified barriers.

1. Introduction

Renewable energy (RE) technologies are rapidly gaining popularity worldwide as a sustainable alternative for generating electricity. This is because they are environmentally friendly [1] and cost-effective [1,2], and their development has a positive impact on the economic development [3]. South Africa is among the countries that have vowed to adopt renewable energy technologies for electricity generation to lower and ultimately eliminate reliance on fossil fuels for electricity generation [4]. South Africa has set a target to limit its carbon dioxide (MtCO₂) emissions to 350–420 million tonnes by 2030 and to achieve net-zero emissions by 2050 [5].

Situated at the southernmost tip of Africa, South Africa is the 25th largest country in the world by land area and has a population of close to 60 million [6]. South Africa is considered an economic hub for Africa due to its industrialised economy and advanced infrastructure. It is not surprising that it is also the largest carbon dioxide emitter on the continent, accounting for at least 33% of the emissions. The burning of coal for electricity generation is the biggest contributor to the country’s CO₂ emissions [3].

South Africa is geographically well-positioned to receive abundant wind and solar energy in its Cape provinces [3]. The annual 24 h global solar radiation average is approximately 220 W/m² [7], coupled with an annual average wind speed of over 4 m/s [8,9]. These conditions collectively position the country well to harness these renewable energy resources for electricity generation, reducing reliance on conventional power generation and mitigating CO₂ emissions.

South Africa’s commitments to a diversified and sustainable energy mix date back to the 1998 White Paper for Energy Policy. The policy laid the foundation for restructuring the energy sector by integrating alternative energy sources. This commitment was further advanced through the 2003 White Paper on Renewable Energy policy, developed by the then Department of Minerals and Energy (now the Department of Electricity and Energy (DEE)). The policy defined the government’s targets for renewable energy development, emphasising the optimal use of the country’s abundant renewable energy resources, such as solar, wind, biomass, and hydro, for on-grid and off-grid power generation, as well as for non-electric technologies [10].

By the end of 2024, the country had generated more than 80% of its electricity from Eskom’s coal-fired power plants, while renewable energy accounted for only about 12% of total generation. This renewable share consisted of solar photovoltaic (PV), concentrated solar power (CSP), and onshore wind energy procured through the government-led Renewable Energy Independent Power Producer Procurement Programme (REIPPPP) [11]. The persistent dominance of coal in the electricity mix underscores the need to accelerate the development of utility-scale renewable energy sources. Therefore, this study aims to address the following research problem: the challenges that hinder the development and integration of utility-scale renewable energy into South Africa’s energy mix are rarely analysed and documented in the literature. The research problem is addressed by first analysing how renewable energy barriers are interrelated and assessing their influence in advancing the integration of renewable energy. Secondly, the study evaluates how these barriers can be managed to effectively integrate renewable energy into South Africa’s energy supply.

There has been greater interest in renewable energy studies among policymakers and academics in South Africa. The study by Todd and McCauley analysed key policy barriers hindering the large-scale implementation of renewable energy technologies in South Africa [12]. The study by Mirzania et al. [13] examined the constraints affecting South Africa’s energy transition goals. Baumli and Jamasb [14] evaluated barriers to private participation in renewable energy projects across Africa, including South Africa. However, a significant gap still exists in the literature regarding the identification and analysis of multidimensional barriers to renewable energy in South Africa. Specifically, there are still limited peer-reviewed studies on the prioritisation of renewable energy barriers and ranking of their solutions using multi-criteria decision-making techniques. Much of the literature analyses these barriers in isolation and fails to identify and critically assess the most significant barriers to renewable energy integration, including their possible interrelationships, thereby overlooking their complex, multidimensional, and interdependent nature [8]. Addressing this gap is crucial because, until these substantial barriers to renewable energy adoption are studied, understood, and strategically removed, the pursuit of a clean and sustainable energy transition will not be realised [15].

This study makes a significant contribution to the existing body of knowledge on renewable energy in South Africa in multiple ways. It develops an integrated hybrid multi-criteria decision-making (MCDM) framework to effectively analyse the barriers and their solutions.

• Firstly, a Decision-Making Trial and Evaluation Laboratory (DEMATEL) MCDM technique is employed to prioritise renewable energy barriers in South Africa, ensuring a more comprehensive and holistic approach to decision-making.
• Secondly, mitigation strategies are identified and tailored to specifically address each barrier, an approach rarely seen in the literature, where solutions tend to be generalised.
• Lastly, an integrated CRITIC-TOPSIS framework is developed to objectively assign weights to criteria and prioritise the mitigation strategies for effectively addressing these barriers.

The subsequent sections of this research paper are organised as follows: Section 2 provides the literature review on barriers and multi-criteria decision-making methods. The research methodology for this study is detailed in Section 3. The findings are presented in Section 4. Section 5 presents the sensitivity analysis for DEMATEL and TOPSIS, and Section 6 discusses the findings in the context of South Africa. Lastly, the conclusion and contributions of the study are presented in Section 7.

2. Literature Review

The adoption of renewable energy into the energy system is a global issue. Different countries face unique challenges and barriers [16,17]. Numerous studies across various developing countries have employed multi-criteria decision-making methods to assess and overcome barriers to renewable energy, including those conducted in Malawi [4], China [18], and Spain [19]. The subsequent subsections first discuss various MCDM techniques and their application in assessing renewable energy barriers globally, followed by renewable energy barriers contextualised to the South African market, as examined by Molepo et al. [8], and conclude by outlining mitigation strategies to address each barrier.

2.1. Multi-Criteria Decision-Making

Multi-criteria decision-making is a multidimensional technique applied widely to determine solutions to problems with multiple objectives and conflicting criteria [20]. MCDM is the most applied analysis and decision-making tool suited to solving complex problems with high uncertainty [21,22]. They consider multiple criteria to obtain a more informed, sustainable, and robust decision [20].

A range of MCDM methods and tools with different purposes have been developed to solve real-world problems in different areas [23]. They are applied broadly across various fields, such as energy management [24], finance and business [25], agriculture, and computer science [23], supply chain, engineering, technology, healthcare, and environmental management [26]. There are numerous and varied MCDM methods with different properties, and their selection and application are dependent on the nature of the decision-making problem [27]. Each MCDM method has its own strengths and limitations. The selection of the most suitable method depends on the nature of the decision problem, the application, the level of complexity, the availability of data, and decision-makers’ preferences [28,29].

Various MCDM methods are applied across different research areas. According to da Ponte et al. [21], the most common MCDM methods applied across many research areas are AHP, ELECTRE, PROMETHEE, TOPSIS, and Grey Theory.

The summary of previous studies conducted across the globe to analyse renewable energy barriers using a range of MCDM techniques is presented in

Table 1. Literature review on barriers that applied MCDM techniques.

Country	Year	Research Study	Top Barriers	MCDM Method	Reference
Türkiye	2026	Evaluation of barriers to integration of renewable energy technologies	Technical and Infrastructural	Single-Valued Spherical Fuzzy SWARA	[30]
Saudi Arabia	2025	Evaluating challenges to renewable energy development	Market and Financial	Fuzzy DEMATEL	[31]
Ghana	2025	Analysing barriers to the adoption of renewable energy	Policy, Economic/Financial and Institutional	Delphi & Fuzzy Synthetic Evaluation (FSE)	[32]
China	2025	Assessing underlying barriers and solutions to the development of solar photovoltaic	Financial and Technological	Grey-AHP	[18]
India	2025	Evaluating barriers impeding the implementation of renewable energy technologies	Economic	Interval-valued Pythagorean fuzzy AHP (IVPyF-AHP)	[33]
Bangladesh	2024	Identifying challenges of the waste-to-energy transition	Economical	BMW	[34]
Tunisia	2024	Assessment of barriers to renewable energy systems	Political and Economic	Integrated Stepwise Assessment Ratio Analysis and DEMATEL (SWARA-DEMATEL)	[35]
Saudi Arabia	2024	Analysis of barriers and solution strategies to the renewable energy transition	Political and Economic	Fuzzy AHP	[36]
Spain	2024	Identify key factors influencing the adoption of decentralised renewable energy technologies	Technical and Economic	ANP	[19]
Malawi	2023	Evaluation of barriers and solutions to the development of RE	Economic and Political-governmental	AHP and Fuzzy TOPSIS	[4]
India	2023	Identify potential challenges hindering the pace of sustainable development	Policy	DEMATEL	[37]
Pakistan	2023	Assessment and prioritisation of barriers to alleviate energy poverty	Financial	AHP	[38]
Egypt	2023	Assessment of risks affecting the deployment of photovoltaic systems	Economic	Type-2 neutrosophic numbers (T2NN-CRITIC-EDAS)	[39]
Ethiopia	2022	Analyse and rank barriers to the development of solar PV	Policy	ISM and MICMAC	[40]
Ghana	2022	Assess and prioritise renewable energy barriers and development strategies	Technical and Economic	CRITIC and Fuzzy TOPSIS	[41]
India	2022	Recognise and rank barriers and their impact on RE technologies development	Policy and Political	Modified Delphi and AHP	[16]
Pakistan	2022	Mapping barriers of RE development against energy literacy dimensions	Technical	PESTEL	[42]
Iran	2022	Identifying risks affecting the sustainable development of solar PV plants	Distribution and Marketing	DEMATEL-ANP	[43]
China	2021	Exploring the best methods for overcoming barriers to the development of the wind energy industry	Institutional (policy and regulatory)	DEMATEL-NK	[44]
Ghana	2020	Exploring barriers to renewable energy development	Political and Regulatory	MULTIMOORA EDAS	[15]
Iran	2020	Identification and removal of barriers preventing the use of renewables	Economic and Technological	BOCR-ANP	[45]

. The data shows an application of a wide range of MCDM methods to analyse renewable energy barriers and solutions. Hybrid and fuzzy AHP are the most applied techniques, followed by hybrid DEMATEL.

2.2. Barriers

The barriers were adopted from a study by Molepo et al. [8], in which interviews were conducted with renewable energy industry experts to contextualise the barriers to the South African market. The list of these barriers is tabulated in

Table 2. Renewable energy barriers in South Africa. Adopted from [8].

Barrier Category	Barrier Code	Barrier	Barrier Description	Reference
Agreement	A1	Lack of agreement	Lack of agreement and alignment between key institutions responsible for driving renewable energy projects.	[46]
Knowledge	K1	Lack of coordination	Lack of coordination between public and private institutions.	[47]
Technological	T1	Challenges of grid capacity/connection	Lack of grid capacity in areas where RE resources are abundant halts RE project development.	[18]
	T2	Lack of technically/technologically skilled labour	Shortage of technically skilled workforce in the RE industry.	[32]
	T3	Lack of Research and Development facilities	The limited research in renewable energy hinders the opportunities to advance renewable energy projects.	[48]
	T4	Intermittent/unreliable supply of renewable energy sources	Renewable energies are intermittent; their generation availability is often compared to that of conventional energy sources, making REs unfavourable.	[31]
	T5	Lack of local manufacturing facilities	Lack of infrastructure, such as local manufacturing facilities, contributes to the slow pace of RE development.	[41]
	T6	Shortage of technical training institutes	Lack of technical training facilities to upskill human resources.	[49]
Economic	E1	Lack of access to funding	Renewable energy projects require high capital costs, and there is limited access to funding.	[50]
Social	S1	Lack of public awareness and acceptance	The public is unaware of the benefits of renewable energy projects; hence, they do not accept them in their communities.	[33]
Political	P1	Lack of policy continuity	Change of government ministers/political leadership impacts RE targets, policy, and commitments.	[35]
	P2	Nepotism	RE projects and licences are awarded with favouritism, reducing interest from investors and other IPPs.	[51]
	P3	Ineffective bureaucratic permit procedures	The government’s process of acquiring permits and licences is long and complex, creating delays to RE projects.	[48]
Others—Market	OM1	Underdeveloped supply chain and logistics	Underdeveloped supply chain and logistics, and challenges with the procurement of renewable energy key components.	[52]
	OM2	Market uncertainties	The unstable global economic markets have a negative impact on RE development.	[52]
Others—Geographical & Environmental	OG1	Challenges with transportation systems/infrastructure	Poor road infrastructure leads to higher transportation and project costs and extended completion timelines.	[53]
	OG2	The challenges of acquiring public/communal land	The process of acquiring land and pre-environmental assessment is challenging.	[31]

. The barriers are grouped into the following categories: Agreement, Knowledge, Technological, Economic, Social, Political, Market, and Geographical. The barriers are assigned codes based on these categories, along with a corresponding description.

2.3. Proposed Mitigation Strategies

A systematic approach is proposed to address the barriers identified in Table 2 by identifying mitigation strategies from the literature and systematically mapping them to each barrier. The summary of the proposed mitigation strategies and their corresponding codes are presented in

Table 3. Proposed solutions to the barriers in South Africa.

Barrier	Solution Description
A1—Lack of agreement	S1—Development of clear regulatory frameworks [32]
K1—Lack of coordination	S2—Coordination, knowledge-sharing, and cooperation among key institutions [1]
T1—Challenges of grid capacity/connection	S3—Invest in grid infrastructure upgrade [18]
T2—Lack of technically/technologically skilled labour	S4—Training and capacity building [54]
T3—Lack of Research and Development facilities	S5—Support Research and Development facilities [36]
T4—Intermittent/unreliable supply of renewable energy sources	S6—Invest in battery energy storage systems [54]
T5—Lack of local manufacturing facilities	S7—Promoting local production capacity [55]
T6—Shortage of technical training institutes	S8—Invest in vocational education and training at the tertiary level [55]
E1—Lack of access to funding	S9—Improved funding mechanisms [33]
S1—Lack of public awareness and acceptance	S10—Public awareness-raising [50]
P1—Lack of policy continuity	S11—Create stable long-term policies [31]
P2—Nepotism	S12—Enforcement of regulatory frameworks and independent reviewers [56]
P3—Ineffective bureaucratic permit procedures	S13—Streamline regulatory processes [36]
OM1—Underdeveloped supply chain and logistics	S14—Develop infrastructure to support supply chain and logistics [33]
OM2—Market uncertainties	S15—Diversify the market [32]
OG1—Challenges with transportation systems/infrastructure	S16—Invest in transportation infrastructure [1]
OG2—The challenges of acquiring public/communal land	S17—Strengthen local community engagement [18]

. The mitigation strategies are tailored to address each barrier specifically, a rare approach in the literature where solutions tend to be generalised. This approach enables the formulation of effective and targeted policy interventions that address each individual barrier.

3. Methodology

The prioritisation of barriers and mitigation strategies is completed using the DEMATEL AND CRITIC-TOPSIS techniques. The survey questionnaire used to collect data for this research was divided into two sections. The questionnaire was developed based on the barriers in Table 2 and mitigation strategies in Table 3. The questionnaire was shared with the renewable energy industry experts in the South African market. The purpose was to rank the influence of the barriers on each other and rate how each of the proposed mitigation strategies/solutions satisfies the criterion. The sample for this research study was selected using a non-probability, purposive expert sampling technique. This type of sampling technique is based on non-random criteria and factors. In this case, the selection criterion was for renewable energy industry experts with 5 or more years of experience in utility-scale wind and solar PV projects in South Africa. Validation of relevance began with a systematic literature review to identify global renewable energy barriers [57], followed by expert validation through semi-structured interviews to contextualise these barriers for the South African market [8]. Lastly, triangulation with the literature is conducted in this research by comparing the findings with peer-reviewed studies. Presented in Figure 1 is the illustration of the steps taken to analyse the barriers and solutions to renewable energy integration using MCDM techniques.

3.1. DEMATEL Analysis

The Decision-Making Trial and Evaluation Laboratory (DEMATEL) is an MCDM method used to analyse complex interrelationships among factors in multifaceted decision-making problems [58]. Introduced in 1971 at Battelle Memorial Institute of Geneva, the DEMATEL technique is gaining popularity and is widely used to solve many real-world problems [59,60]. It is a vital tool with the ability to visualise causal links and analyse the interdependence among criteria/factors [61]. DEMATEL solves complex decision-making problems by providing a structural model utilising decision-makers’ knowledge [62]. One of the main advantages of DEMATEL compared to other techniques is its potential to yield credible outcomes with minimal data from an expert questionnaire, rather than a public one [62,63]. Its shortcoming is the inability to determine hierarchy and the weights of individual criteria [64]. The DEMATEL method was selected for this study due to its ability to incorporate experts’ judgements [65]. Moreover, as highlighted by Richards et al. [46], renewable energy barriers are highly interdependent and cannot be analysed in isolation. DEMATEL provides an integrated approach to examining such interdependencies and is particularly effective at addressing problems involving causal relationships among multiple factors [65,66].

The rating of the influence of the barriers on each other was done using a 5-point Likert scale ranging from 0 to 4. The ratings were represented as follows: 0 for no influence, 1 for very low influence, 2 for low influence, 3 for strong influence, and 4 for very high influence. A DEMATEL analysis was conducted as follows.

Step 1: Construct the combined/average matrix of all participants’ feedback (A).

The first step is to construct an average matrix (A) of all participants’ responses using Equation (1), where i and j represent the row and column indices, respectively. $a_{ij}$ represents the average factor of index i and j, $X_{ij}^k$ represents an integer score k assigned by experts, and H represents the number of experts.

$${A=a}_{ij}={\displaystyle \frac{1}{H}}\sum _{k=1}^H{X}_{ij}^k$$

(1)

Step 2: Construct the normalised initial direct-relation matrix (D).

The normalised initial direct-relation matrix is calculated through the normalisation of the combined/average matrix (A) using Equations (2) and (3). Normalisation is the crucial step for maintaining consistency and comparability in the analysis.

$$D=A\times S$$

(2)

where S is

$$S=Min\left[{\displaystyle \frac{1}{Max{\sum }_{j=1}^n{a}_{ij}}}\right.,\left.{\displaystyle \frac{1}{Max{\sum }_{i=1}^n{a}_{ij}}}\right]$$

(3)

Step 3: Construct the total relation matrix (T).

The next step is the calculation of the total relation matrix using Equation (4), where I denotes the identity matrix. The total relation matrix is vital in visualising the complex interdependencies within the barriers.

$$T=D\left({I-\left.D\right)}^{-1}\right.$$

(4)

Step 4: Compute the R and C matrices.

In this step, the sum of the rows (R) and the sum of the columns (C) are calculated from the total relation matrix (T) using Equations (5) and (6), respectively. From this, the computation of $R + C$ and $R - C$ is carried out to obtain the prominence and relation. $R + C$ indicates the degree of relation between the barriers [67]. The highest value of $R + C$ indicates the strongest relation with other barriers, while the lowest value indicates the weakest relation with other barriers. $R-C$ indicates the kind of relation between the barriers. The positive value of $R - C$ indicates cause (influence on other barriers), while the negative value indicates effect (influenced by other barriers) [67].

Calculate the sum of elements in a row; R

$$R={\left(\sum _{i=1}^n{t}_{ij}\right)}_{1\times n}$$

(5)

Calculate the sum of elements in a column; C

$$C={\left(\sum _{j=1}^n{t}_{ij}\right)}_{1\times n}$$

(6)

Step 5: Draw a cause-and-effect diagram.

In this step, the DEMATEL causal diagram is developed from $R+ C$ and $R - C$ calculations to visualise the cause-and-effect relationship between the renewable energy barriers.

Step 6: Calculate the threshold value to obtain the digraph. The threshold limit of influence is obtained using Equation (7), where $t_{ij}$ represents the total direct and indirect influence of barriers in row i on barriers in column j, and n² represents the number of barriers.

$$\alpha ={\displaystyle \frac{{\displaystyle {\sum }_{i=1}^n}{\sum }_{j=1}^n{t}_{ij}}{n^2}}$$

(7)

3.2. CRITIC-TOPSIS Analysis

The mitigation strategies are prioritised using the Technique for Order of Preference by Similarity to Ideal Solution (TOPSIS) technique. However, given that weight determination is the most important component of MCDM methods [68], the Criteria Importance Through Inter-criteria Correlation (CRITIC) technique was employed to objectively assign weights to the evaluation criteria.

It is imperative to analytically assess and select the evaluation criteria, ensuring theoretical background and practical relevance [69]. The selection of criteria in this study is supported by previous literature, which consistently demonstrates that decision criteria in renewable energy systems require the consideration of multiple dimensions spanning technical, social, economic, and environmental factors [14,70,71]. The evaluation criteria are presented in

Table 4. Evaluation criteria for renewable energy strategies.

Code	Evaluation Criteria	Cost/Benefit
C1	Cost effectiveness	Cost
C2	Technical feasibility	Benefit
C3	Social acceptance	Benefit
C4	Implementation time	Benefit
C5	Environmental impact	Benefit

. Experts were asked to rate how each of the strategies in Table 3 satisfies the criteria in Table 4. The results were analysed using a combined CRITIC-TOPSIS technique to determine the weights of the evaluation criteria and then rank the alternatives/strategies. The advantage of the CRITIC method is that it eliminates the need for pairwise comparison in determining weights [72]. This reduces reliance on experts and, therefore, diminishes bias and subjective judgement, increasing the reliability of the decision-making process [69].

3.2.1. Weight Calculation

Criteria Importance Through Inter-criteria Correlation (CRITIC) is one of the several MCDM weighting techniques that compute objective weights for criteria. The CRITIC method determines the relative importance of criteria based on objective values [69]. Proposed in 1995, the CRITIC method establishes criterion weights by assessing the standard deviation and interrelationship between criteria [73].

The survey questionnaire results were aggregated into a single decision matrix X, and the following steps were used to compute the criteria weights using the CRITIC method’s stepwise technique.

$$\mathrm{X}=\left[\begin{array}{cccc}{x}_{11}& x_{12}& \dots & x_{1j}\\ x_{21}& x_{22}& \dots & x_{2j}\\ ⋮& ⋮& \ddots & ⋮\\ x_{i1}& x_{i2}& \dots & x_{ij}\end{array}\right]$$

Step 1: Normalise the decision matrix X (where i = alternative and j = criterion) using Equation (8) for the benefit criterion and Equation (9) for the cost criterion. In Equations (8) and (9), $x_j^{-}=\mathrm{m}\mathrm{i}\mathrm{n}(x_1,x_2,\dots ,x_m)$ and $x_j^{+}=\mathrm{m}\mathrm{a}\mathrm{x}(x_1,x_2,\dots ,x_m)$ respectively.

$${x^{*}}_{ij}={\displaystyle \frac{x_{ij}-x_j^{-}}{x_j^{+}-x_j^{-}}}$$

(8)

$${x^{*}}_{ij}={\displaystyle \frac{x_j^{+}-x_{ij}}{x_j^{+}-x_i^{-}}}$$

(9)

Step 2: Calculate the standard deviation of each criterion using Equation (10). Standard deviation is a statistical measure that shows how much a data point deviates from the mean value [73]. A higher standard deviation value indicates that the criterion has a stronger ability to differentiate between alternatives and therefore is more significant in the decision-making process, while a lower standard deviation value indicates that the criterion is assigned a low priority [69].

$${\sigma }_j=\sqrt{{\displaystyle \frac{1}{m}}{\displaystyle \sum _{i=1}^m}{\left(x_{ij}-{\overline{x}}_j\right)}^2}$$

(10)

Step 3: Calculate the correlation matrix using Equation (11). The correlation matrix plays an important role in quantifying the degree of interdependence among every criterion considered in analysing mitigation strategies for renewable energy barriers [73]. It is computed by comparing the performance of each criterion with that of other criteria [69]:

$${\rho }_{jk}={\displaystyle \frac{{\sum }_{i=1}^m\left(x_{ij}-{\overline{x}}_j\right)\left(x_{ik}-{\overline{x}}_k\right)}{\sqrt{{\sum }_{i=1}^m{\left(x_{ij}-{\overline{x}}_j\right)}^2}. \sqrt{{\sum }_{i=1}^m{\left(x_{ik}-{\overline{x}}_k\right)}^2} }}$$

(11)

Step 4: Calculate the amount of information using Equation (12). The information content measure represents the degree of redundancy or overlaps between each criterion and all other criteria. The higher the information content value, the less redundant the criterion is with respect to other criteria, and therefore the more important it is in the decision-making process [69].

$$C_j={\alpha }_j.{\displaystyle \sum _{k=1}^n}(1-{\rho }_{jk})$$

(12)

Step 5: Calculate the criteria weights using Equation (13). The weights are assigned to the criteria considering the criterion determined from the standard deviation and the inter-criterion correlation determined from the information content [69,73].

$$w_j={\displaystyle \frac{C_J}{{\sum }_{J=1}^n{C}_j}}$$

(13)

3.2.2. Prioritisation of Mitigation Strategies

Technique for Order of Preference by Similarity to Ideal Solution (TOPSIS) is a multifaceted MCDM method applied to evaluate and determine the best alternatives based on their similarity to the ideal solution and dissimilarity to the worst solution [26]. The TOPSIS method was formulated in the 1980s by Yoon and Hwang and has since gained popularity in solving problems with many criteria for alternatives to be evaluated [28,70]. This method is very efficient, easy to compute, and has been applied broadly to solve MCDM problems in different study areas such as manufacturing, finance, sustainable engineering, supply chain, and education [28,74].

TOPSIS analysis was conducted to rank the proposed mitigation strategies using the following prescribed steps:

Step 1: Construct the normalised decision matrix using Equation (14). Normalisation is critical in the TOPSIS steps; it transforms the original data into unitless values, eliminating unit inconsistencies and ensuring compatibility in the calculation across different alternatives [73].

$$r_{ij}={\displaystyle \frac{x_{ij}}{{\sum }_{j=1}^n{x^2}_{ij}}}$$

(14)

Step 2: The weighted normalised matrix is calculated following Equation (15), where $r_{ij }$ is the normalised decision matrix calculated in Equation (14) and $w_j$ is the weight calculated in Equation (13) using the CRITIC method.

$$v_{ij}={ r}_{ij }\times w_j$$

(15)

Step 3: Compute the degree of proximity to the ideal solution, for positive ideal ($A^{+})$ and negative ideal ($A^{-}$), using Equations (16) and (17) respectively. A positive ideal represents the best possible performance for each solution, whereas a negative ideal represents the poorest possible performance [69].

$$\begin{gathered} A^{+}=\left\{v_1^{+},\dots ,v_n^{+}\right\}, \\ where v_j^{+}\left\{\mathit{max}\left(v_{ij}\right) if j \in J;\mathit{min}\left(v_{ij}\right) if j \in J^{\prime }\right\} \end{gathered}$$

(16)

$$\begin{gathered} A^{-}=\left\{v^{-}, v^{-},\dots ,v^n\right\}, \\ where v_j^{-}\left\{\min \left(v_{ij}\right) if j\in J;\max \left(v_{ij}\right) if j \in J^{\prime }\right\} \end{gathered}$$

(17)

Step 4: Calculate the separation measures/Euclidean distances for the positive and negative ideal using Equations (18) and (19) respectively. ${S_i}^{+}$ indicates the distance between an alternative and the ideal best solution while ${S_i}^{-}$ indicates the distance between an alternative and the ideal worst solution.

$${S_i}^{+}=\sqrt{{\displaystyle \sum _{j=1}^n}{\left(v_{ij}-{v_j}^{+}\right)}^2}$$

(18)

$${S_i}^{-}=\sqrt{{\displaystyle \sum _{j=1}^n}{\left(v_{ij}-{v_j}^{-}\right)}^2}$$

(19)

Step 5: Calculate the relative closeness to the solution using Equation (20). A higher $C_i$ value indicates that the alternative is closest to the ideal solution and therefore considered the best [71].

$$C_i={\displaystyle \frac{S_i^{-}}{S_i^{-}+S_i^{+}}}$$

(20)

Step 6: Rank solutions.

The mitigation strategies are ranked based on their relative closeness values. The calculated value of $C_i$ ranges between 0 and 1. The higher the $C_i$, the better the solutions.

4. Results

This section presents the results of the survey questionnaire, structured into three subsections: respondents’ profile in Section 4.1, ranking of barriers in Section 4.2 and prioritisation of mitigation strategies in Section 4.3. The survey questionnaires were sent to a diverse group of professionals in the renewable energy industry, representing public and private organisations. Fourteen positive responses were received and analysed. DEMATEL is not statistically based, and a small sample size is required to achieve statistical significance [62]. According to various studies conducted, the following sample sizes are sufficient for the DEMATEL technique: between six and twenty-five [75], between ten and fifteen [76], and at least twelve [77]. Therefore, a sample size of fourteen was deemed sufficient for this study.

4.1. Respondents Profile

The experts’ profiles were analysed in terms of years of experience and level of qualification. The analysis of the experts’ profiles is depicted in Figure 2a,b. As shown in Figure 2a, the experts’ level of experience ranges from 5 to 20 years, with 79% having 11–20 years of experience. The level of qualifications presented in Figure 2b shows that 64% of experts possess postgraduate degrees, with 43% at the Master’s level. The inclusion of highly experienced experts with high levels of qualifications strengthens the credibility and validity of the study.

The respondents’ profiles are further analysed based on their institutional distribution. Figure 3 presents the distribution of participants by institutional affiliation, indicating that 86% are from private institutions and 14% from public institutions. Further analysis also considers the distribution based on institutional focus, as illustrated in Figure 4. The experts, drawn from diverse professional backgrounds and experiences, provided valuable insights into the barriers hindering renewable energy integration in South Africa.

4.2. Prioritisation of Barriers Using DEMATEL

This section presents the ranking and prioritisation of the barriers hindering renewable energy integration in South Africa, using the DEMATEL technique. The results of the combined matrix obtained using Equation (1) are presented in

Table A1. Average matrix (A).

	A1	K1	T1	T2	T3	T4	T5	T6	E1	S1	P1	P2	P3	OM1	OM2	OG1	OG2
A1	0.0	3.5	2.5	2.2	2.7	1.1	2.5	2.7	2.6	1.6	2.2	2.4	3.0	1.8	2.6	1.4	2.4
K1	3.2	0.0	2.8	2.4	2.3	1.0	2.2	2.2	2.4	1.8	2.5	2.1	3.1	1.9	2.7	1.1	1.9
T1	3.1	2.3	0.0	2.1	2.5	3.3	2.4	2.4	3.3	0.9	1.9	2.5	1.9	1.8	2.6	1.4	1.9
T2	2.3	2.3	1.6	0.0	3.0	2.6	2.9	2.9	1.4	1.2	1.4	1.7	0.9	1.4	1.5	0.8	0.7
T3	2.5	2.4	1.9	3.3	0.0	2.7	3.1	2.7	2.0	1.9	1.4	1.6	1.4	2.4	2.2	1.6	1.0
T4	1.2	1.6	3.4	2.2	1.7	0.0	1.6	1.7	3.1	2.2	1.0	1.0	1.1	1.4	2.1	0.6	1.0
T5	2.5	2.4	1.9	3.2	2.4	2.1	0.0	2.6	2.0	1.9	1.4	1.6	1.4	2.4	2.1	1.5	1.0
T6	2.5	2.4	1.9	3.2	2.7	2.1	2.8	0.0	2.0	1.9	1.4	1.6	1.4	2.4	2.2	1.6	1.1
E1	2.5	2.6	2.5	1.6	2.5	2.4	2.7	2.4	0.0	1.5	1.9	2.5	2.1	1.9	2.9	1.3	1.4
S1	2.2	2.0	0.9	1.9	2.1	1.6	1.8	1.9	1.2	0.0	1.3	1.5	1.4	1.2	1.5	0.8	1.9
P1	2.8	2.6	2.7	1.5	2.4	1.0	2.2	2.2	2.1	1.4	0.0	3.4	2.9	1.9	2.9	1.5	1.7
P2	2.6	2.6	2.3	2.3	2.6	0.9	2.4	2.4	2.9	2.1	2.9	0.0	2.7	1.9	2.4	1.8	1.8
P3	2.8	2.8	2.1	1.5	1.6	1.5	1.4	1.5	1.9	0.9	2.0	2.3	0.0	1.5	2.2	1.4	1.7
OM1	2.2	2.1	1.4	2.3	2.0	1.3	1.9	1.9	1.9	0.9	1.1	1.8	1.9	0.0	2.1	2.4	0.6
OM2	2.6	2.6	3.0	2.4	2.9	1.9	2.6	2.8	3.1	1.9	2.7	2.6	2.9	2.2	0.0	2.1	1.6
OG1	1.6	1.4	1.1	1.6	1.5	0.7	1.3	1.4	1.9	0.8	1.6	1.9	2.1	2.6	2.2	0.0	0.9
OG2	1.7	1.7	1.3	0.7	0.9	0.9	0.6	0.7	1.6	2.5	1.7	1.7	1.8	0.6	1.0	0.7	0.0

in Appendix A and the results of the normalised initial direct-relation matrix calculated using Equation (2) are presented in

Table A2. Normalised initial direct matrix (D).

	A1	K1	T1	T2	T3	E1	E2	E3	E4	S1	P1	P2	P3	OM1	OM2	OG1	OG2
A1	0.000	0.088	0.063	0.055	0.068	0.029	0.063	0.068	0.064	0.039	0.055	0.059	0.075	0.045	0.066	0.034	0.059
K1	0.081	0.000	0.070	0.061	0.057	0.025	0.055	0.055	0.061	0.045	0.063	0.054	0.077	0.048	0.068	0.027	0.047
T1	0.077	0.057	0.000	0.054	0.063	0.082	0.061	0.061	0.082	0.021	0.047	0.063	0.047	0.045	0.064	0.036	0.047
T2	0.057	0.057	0.039	0.000	0.075	0.066	0.073	0.073	0.036	0.030	0.034	0.043	0.023	0.034	0.038	0.020	0.018
T3	0.063	0.059	0.047	0.082	0.000	0.068	0.077	0.068	0.050	0.048	0.036	0.041	0.034	0.059	0.055	0.039	0.025
T4	0.030	0.039	0.084	0.055	0.043	0.000	0.041	0.043	0.077	0.055	0.025	0.025	0.027	0.036	0.052	0.016	0.025
T5	0.063	0.059	0.047	0.081	0.059	0.054	0.000	0.064	0.050	0.048	0.036	0.041	0.034	0.059	0.054	0.038	0.025
T6	0.063	0.059	0.047	0.081	0.068	0.054	0.071	0.000	0.050	0.048	0.036	0.041	0.034	0.059	0.055	0.039	0.027
E1	0.063	0.064	0.063	0.039	0.063	0.061	0.068	0.061	0.000	0.038	0.048	0.063	0.054	0.048	0.073	0.032	0.036
S1	0.055	0.050	0.023	0.048	0.052	0.039	0.045	0.047	0.030	0.000	0.032	0.038	0.036	0.030	0.038	0.020	0.047
P1	0.070	0.064	0.068	0.038	0.061	0.025	0.055	0.055	0.052	0.036	0.000	0.084	0.073	0.047	0.073	0.038	0.043
P2	0.066	0.064	0.057	0.057	0.066	0.021	0.059	0.061	0.072	0.052	0.073	0.000	0.068	0.047	0.061	0.045	0.045
P3	0.070	0.070	0.054	0.038	0.041	0.038	0.036	0.038	0.048	0.023	0.050	0.057	0.000	0.038	0.055	0.034	0.043
OM1	0.055	0.054	0.034	0.057	0.050	0.032	0.047	0.047	0.048	0.021	0.029	0.045	0.048	0.000	0.054	0.061	0.016
OM2	0.064	0.066	0.075	0.059	0.073	0.048	0.066	0.070	0.077	0.048	0.068	0.064	0.073	0.055	0.000	0.052	0.039
OG1	0.041	0.036	0.027	0.039	0.038	0.018	0.032	0.034	0.048	0.020	0.039	0.048	0.054	0.064	0.055	0.000	0.023
OG2	0.043	0.043	0.032	0.018	0.021	0.021	0.016	0.018	0.041	0.063	0.043	0.043	0.045	0.016	0.025	0.018	0.000

. The results of the total relation matrix computed using Equation (3) are presented in

Table A3. Total relation matrix (T).

	A1	K1	T1	T2	T3	T4	T5	T6	E1	S1	P1	P2	P3	OM1	OM2	OG1	OG2
A1	0.28	0.35	0.30	0.31	0.33	0.23	0.31	0.32	0.32	0.22	0.26	0.29	0.30	0.26	0.32	0.19	0.22
K1	0.34	0.26	0.30	0.30	0.31	0.22	0.30	0.30	0.30	0.22	0.26	0.28	0.30	0.25	0.31	0.18	0.21
T1	0.34	0.32	0.24	0.30	0.32	0.28	0.31	0.31	0.33	0.20	0.25	0.29	0.27	0.25	0.31	0.19	0.21
T2	0.27	0.27	0.23	0.20	0.28	0.22	0.27	0.27	0.24	0.18	0.20	0.22	0.20	0.20	0.24	0.15	0.15
T3	0.31	0.30	0.26	0.31	0.24	0.25	0.31	0.30	0.28	0.21	0.22	0.25	0.24	0.25	0.29	0.18	0.17
T4	0.23	0.24	0.26	0.24	0.24	0.15	0.23	0.23	0.26	0.19	0.18	0.20	0.20	0.19	0.24	0.13	0.15
T5	0.30	0.29	0.26	0.30	0.29	0.23	0.22	0.28	0.27	0.21	0.22	0.24	0.23	0.24	0.27	0.18	0.17
T6	0.31	0.30	0.26	0.30	0.30	0.23	0.30	0.23	0.28	0.21	0.22	0.25	0.24	0.25	0.28	0.18	0.17
E1	0.32	0.32	0.29	0.28	0.31	0.25	0.31	0.30	0.24	0.21	0.24	0.28	0.27	0.25	0.31	0.18	0.19
S1	0.24	0.23	0.19	0.22	0.23	0.18	0.22	0.22	0.20	0.13	0.17	0.19	0.19	0.17	0.21	0.13	0.16
P1	0.33	0.32	0.30	0.28	0.31	0.22	0.30	0.30	0.30	0.21	0.20	0.30	0.29	0.25	0.32	0.19	0.20
P2	0.34	0.33	0.29	0.30	0.32	0.22	0.31	0.31	0.32	0.23	0.28	0.23	0.29	0.25	0.31	0.20	0.21
P3	0.29	0.28	0.25	0.24	0.25	0.20	0.24	0.24	0.25	0.17	0.22	0.24	0.19	0.21	0.26	0.16	0.18
OM1	0.26	0.26	0.22	0.25	0.25	0.18	0.24	0.24	0.24	0.16	0.19	0.22	0.22	0.16	0.25	0.18	0.14
OM2	0.36	0.35	0.33	0.33	0.35	0.26	0.34	0.34	0.35	0.24	0.29	0.31	0.32	0.28	0.28	0.22	0.21
OG1	0.23	0.22	0.19	0.21	0.21	0.15	0.20	0.20	0.22	0.14	0.18	0.20	0.21	0.21	0.23	0.11	0.13
OG2	0.19	0.19	0.16	0.16	0.17	0.13	0.16	0.16	0.18	0.16	0.16	0.17	0.17	0.13	0.17	0.11	0.09

. The total relation matrix reflects both the direct and indirect influences among the barriers. The results for $R+C$ and $R-C$ are shown in

Table 5. Results of DEMATEL analysis.

Barrier Code	R	C	$\mathit{R}+\mathit{C}$	$\mathit{R}-\mathit{C}$	Ranking	Impact
A1	4.814291	4.955528	9.769819723	−0.14123693	1	Effect
K1	4.648703	4.818613	9.467316014	−0.16991096	3	Effect
T1	4.697084	4.318756	9.015840405	0.378328443	6	Cause
T2	3.771889	4.505784	8.277673402	−0.73389542	11	Effect
T3	4.384599	4.679843	9.064442286	−0.29524363	5	Effect
T4	3.539515	3.608078	7.147592511	−0.06856338	14	Effect
T5	4.192037	4.547442	8.739479149	−0.35540438	9	Effect
T6	4.294474	4.520852	8.815325913	−0.22637764	8	Effect
E1	4.555967	4.571942	9.127908501	−0.0159753	4	Effect
S1	3.270449	3.303985	6.5744341	−0.03353649	15	Effect
P1	4.622518	3.728066	8.350584228	0.894451354	10	Cause
P2	4.736179	4.171971	8.908150366	0.564207413	7	Cause
P3	3.856595	4.141613	7.998207897	−0.2850184	12	Effect
OM1	3.655512	3.803006	7.458517846	−0.14749385	13	Effect
OM2	5.149422	4.585921	9.735342806	0.563501177	2	Cause
OG1	3.249922	2.881069	6.130991539	0.368853191	16	Cause
OG2	2.658154	2.954839	5.612992806	−0.2966852	17	Effect

. The description of the barrier codes is in Table 2 of Section 2.2. The barrier with the highest $R+C$ value is the most prominent barrier, and the one with the lowest $R+C$ value is the least prominent barrier [78].

Drawing from the results in Table 5, five barriers (T1, P1, P2, OM2, and OG1) are identified as cause barriers. According to the DEMATEL analysis, these barriers have an influence on other barriers [67]. The remaining twelve barriers (A1, K1, T2, T3, T4, T5, T6, E1, S1, P3, OM1, and OG2) are identified as effect barriers. These are barriers that are influenced by others [67].

The top three critical barriers ranked based on the highest $R+C$ prominence are a lack of agreement between key stakeholders responsible for driving renewable energy projects (A1)—there is no agreement and alignment between key stakeholders in key institutions; market uncertainties (OM2)—global market fluctuations lead to exchange rate/currency instability and price volatility, which discourages investment in renewable energy; and a lack of coordination between public and private institutions (K1)—the lack of coordination, partnership, knowledge-sharing and collaboration between the government and non-government institutions results in poor innovation necessary to advance renewable energy projects.

The visualisation of the causal diagram is shown in Figure 5. Each barrier is plotted in one of the four quadrants. The positive $R-C$ values lie above the x-axis, while the negative values lie below the x-axis. Additionally, barriers to the left of the mean line are considered to have less influence, and those to the right of the mean line are considered to have significant influence [79].

The four quadrants are classified as follows: core (Q1), driving (Q2), independent (Q3), and impact barriers (Q4). The barriers in Q1 are identified as cause barriers and are the most influential, with high prominence and high relation [67,80,81]. The barriers in Q2 are also identified as cause barriers, and they are the determinants with low prominence and high relation [80,81,82,83]; however, their existence must not be overlooked. The barriers in Q3 are identified as effect barriers, with low prominence and low interrelations [67,80,81,83]. The barriers in Q4 are also identified as effect barriers, with high prominence but low interrelations [67,80,81,83].

The critical root-cause barriers (placed in the first quadrant) that need focus are market uncertainties (OM2), challenges with grid capacity/connection (T1), nepotism (P2), and lack of policy continuity (P1). South Africa’s heavy reliance on international markets for renewable energy technologies makes it susceptible to global uncertainties. These uncertainties affect project cost and duration. The challenge of grid connection is currently a burning issue in South Africa. The rich resource provinces have no grid availability, which has stalled many projects [3]. The inconsistency of the REIPPPP discourages IPPs and investors from participating in the government procurement programme [84]. South Africa is prone to frequent leadership changes, which affect current policies and commitments. Solving these barriers will, in turn, reduce the severity of several others.

The critical effect barriers placed in the fourth quadrant, include a lack of alignment (A1), a lack of coordination (K1), a lack of access to funding (E1), a lack of technically skilled labour (T2), a lack of Research and Development facilities (T3), a lack of local manufacturing facilities and a shortage of technical training institutes (T6). These barriers are a consequence of the other barriers and do not need to be solved directly. Resources must not be focused on addressing these barriers; however, they should be monitored and managed [67], and if they are reduced, it means the strategy applied to address the critical root-cause barriers is working.

The challenges of acquiring public/communal land (OG2), lack of public awareness and acceptance (S1), intermittent/unreliable supply of renewable energy sources (T4), ineffective bureaucratic permit procedures (P3), and underdeveloped supply chain and logistics (OM1) are placed in the third quadrant and are independent barriers. These barriers have limited relations with other barriers. The challenge with transportation systems/infrastructure (OG1) is the only barrier placed in the second quadrant. This barrier is a driving factor. Its impact is limited; however, it should not be overlooked because if left to operate independently, it is still significant [67].

The threshold limit of influence value of 0.243 is obtained using Equation (7) and all values that are less than this threshold value are removed from the total relation matrix. This is to ensure that all minor effects are eliminated from the total relation matrix, and the significant ones are kept and used to construct the influential network relation map [62,85].

The interrelation diagram for the most influential barriers is shown in Figure 6. The influence of each barrier is represented by an arrow with a different colour. The direction of the arrow represents the influence of the barrier on the other. An incoming arrow represents the direction of the influence from another barrier, while the outgoing arrow represents the direction of the influence to the other barriers. The lack of agreement (A1), a lack of coordination (K1), the challenges of grid capacity/connection (T1), and market uncertainties (OM2) have multiple arrows pointing outwards from them. Furthermore, in Figure 5, A1 and K1 barriers are positioned in Q4, indicating that they are the most influential effect barriers, while T1 and OM2 are positioned in Q1, indicating that they are cause barriers with the most significant relationship with other barriers.

4.3. Prioritisation of Mitigation Strategies Using CRITIC-TOPSIS

This section presents the calculated weights using the CRITIC technique. Each weight represents the relative importance of each criterion. The results of the correlation matrix $({\rho }_{jk}$), standard deviation (${\sigma }_j$), information content ($C_j$), and weights of criteria ($w_j)$ obtained using the stepwise technique of the CRITIC method are presented in

Table 6. Calculated weights.

	C1	C2	C3	C4	C5	${\mathit{\sigma }}_{\mathit{j}}$	${\mathit{C}}_{\mathit{j}}$	${\mathit{w}}_{\mathit{j}}$
C1	1.000	0.651	−0.652	−0.094	0.007	0.257	1.049	0.199
C2	0.651	1.000	−0.744	0.024	−0.139	0.230	0.967	0.183
C3	−0.652	−0.744	1.000	0.042	0.136	0.240	1.250	0.237
C4	−0.094	0.024	0.042	1.000	0.504	0.264	0.929	0.176
C5	0.007	−0.139	0.136	0.504	1.000	0.310	1.082	0.205

. The positive and high values of the correlation matrix between criterion (i.e., C1 & C2 and C4 & C5) indicate redundancy, while the low negative correlations (C1 & C3 and C2 & C3) indicate that criteria provide diverse and unique information [73].

Criteria C3 (social acceptance) has the highest information content ($C_j$) value indicating its significance in the decision-making process, and it also has the highest weight ($w_j$), indicating that it is the most influential factor for evaluating renewable energy strategies. These weights are going to be used in the subsequent steps to evaluate and prioritise mitigation strategies.

Seventeen mitigation strategies identified in Section 2.3 are ranked and prioritised in this section following the TOPSIS technique. Presented in

Table 7. Results of TOPSIS analysis.

Code	C1	C2	C3	C4	C5	${\mathit{S}}_{\mathit{i}}^{+}$	${\mathit{S}}_{\mathit{i}}^{-}$	${\mathit{C}}_{\mathit{i}}$	Ranking
S1	0.048	0.045	0.051	0.043	0.056	0.029	0.030	0.508	7
S2	0.050	0.046	0.053	0.040	0.055	0.029	0.029	0.497	9
S3	0.050	0.056	0.044	0.041	0.046	0.037	0.029	0.445	12
S4	0.041	0.044	0.056	0.043	0.048	0.025	0.030	0.543	4
S5	0.047	0.048	0.059	0.036	0.046	0.027	0.028	0.512	6
S6	0.058	0.051	0.042	0.037	0.046	0.043	0.023	0.351	15
S7	0.043	0.041	0.060	0.037	0.036	0.033	0.026	0.443	13
S8	0.048	0.048	0.051	0.044	0.048	0.029	0.027	0.485	10
S9	0.055	0.042	0.049	0.038	0.042	0.040	0.015	0.279	17
S10	0.039	0.031	0.068	0.037	0.052	0.029	0.036	0.548	3
S11	0.046	0.039	0.056	0.043	0.056	0.027	0.030	0.526	5
S12	0.045	0.042	0.054	0.049	0.055	0.027	0.033	0.551	2
S13	0.043	0.041	0.051	0.047	0.054	0.029	0.030	0.507	8
S14	0.043	0.042	0.061	0.044	0.036	0.030	0.028	0.482	11
S15	0.051	0.044	0.050	0.039	0.044	0.036	0.019	0.351	16
S16	0.038	0.036	0.058	0.033	0.041	0.034	0.027	0.439	14
S17	0.045	0.037	0.075	0.045	0.057	0.021	0.044	0.676	1

are the results of the weighted normalised matrix $\left(v_{ij}\right)$, Euclidean distance $({S_i}^{+} \& {S_i}^{-})$ and relative closeness ${(C}_i)$ obtained using Equations (14)–(20). The weighted normalised matrix represents the integrated performance of each mitigation strategy across all five criteria. Criterion C1 is a cost criterion, and C2–C5 are specified as benefit criteria. In the weighted normalised matrix, the optimal mitigation strategies are characterised by lower weighted values for cost criteria and higher weighted values for benefit criteria, both of which contribute positively to ranking [73]. A lower value of ${S_i}^{+}$ indicates that the mitigation strategy is closer to the best solution, and conversely the higher value of ${S_i}^{-}$ indicates that the mitigation strategy is farther from the worst solution. The ranking of the most influential mitigation strategy is based on the highest C_i value. As shown in Table 7, S17 is ranked first with the highest $C_i$ value of 0.676, followed by S12 with 0.551 and S10 with 0.548.

The visualisation of the ranking of the mitigation strategies is presented in Figure 7, highlighting the most and least influential strategies based on the Ci value. The results show that strengthening local community engagement (S17) is the most effective strategy for overcoming barriers to renewable energy adoption in the South African market. This emphasises the importance of involving community members in the early stages of renewable energy project development [31]. The engagement will improve community acceptance of these projects.

The enforcement of regulatory frameworks and the use of independent reviewers (S12) is the second most significant strategy, with a Ci value of 0.551. Enforcing regulations and involving independent reviewers is essential for improving and maintaining the credibility of the REIPPPP. Enforcing regulatory frameworks will ensure that frameworks and procedures are properly implemented, thereby reducing delays in appointing preferred bidders. Moreover, the inclusion of independent reviewers will mitigate risks and enforce transparency in the public procurement programme.

Public awareness-raising (S10) ranked third among the most important mitigation strategies for overcoming barriers to renewable energy in South Africa. The government must initiate programmes to promote renewable energy literacy. Campaigns and training programmes must be accessible to the public to raise awareness and encourage acceptance of renewable energy projects. The programmes must be targeted at the young population in institutions of higher learning and the broader public to ensure that no one is left out of the just energy transition. These programmes will eliminate misconceptions about renewable energy and provide a clear understanding of the benefits of adopting these technologies [32].

5. Sensitivity Analysis

A sensitivity analysis was conducted to assess the reliability and robustness of the decision-making results. Reliability is the degree to which the measure of a construct is consistent or dependable [86]. Sensitivity analysis is a popular tool in MCDM, used to validate the robustness of the technique or framework [87]. The analysis examines how variations in input parameters affect the decision-making model’s results [88]. This can be achieved by either varying the weights assigned to experts or by changing the weights of the criteria by a certain percentage [89]. Sensitivity analysis was conducted on the DEMATEL and TOPSIS techniques to determine how variations in weights affect the prioritisation of barriers and strategies. In the DEMATEL technique, the weight of the barrier with the strongest influence is systematically varied across predetermined ranges to assess the sensitivity of the results [88]. This approach helps in evaluating the extent to which the outcome depends on a single influential barrier. For the TOPSIS technique, the weights are systematically varied among the benefit and cost criteria to ensure a balanced and unbiased assignment [69]. This approach enables the analysis to test the stability of the ranking results by observing the sensitivity of the outcomes to changes in each criterion.

5.1. Sensitivity Analysis for Barriers

This study conducted a sensitivity analysis to assess the robustness and consistency of the DEMATEL method by varying the top-ranked/most influential barrier. The lack of agreement (A1) barrier was systematically varied from 0.1 to 0.9, to measure its impact on other barriers, while the weights of the other barriers were proportionally adjusted to ensure that the total weight remained equal to 1 [90]. This resulted in the development of nine new cases or scenarios to compare with the original case. This approach enabled the assessment of how changes in the relative importance of barrier A1 impact the overall causal structure and barrier rankings within the DEMATEL technique. The results of the sensitivity analysis, showing all cases, are presented in Figure 8. In all the cases (original and nine new cases), the ranking of the following barriers remained stable: the lack of agreement (A1), a lack of public awareness and acceptance (S1), and a lack of local manufacturing facilities (T5). These findings confirm the consistency and robustness of these barriers across different variations, concluding that the lack of knowledge (A1) is a critical barrier. Therefore, more focus should be placed on addressing this barrier, ensuring that there is agreement among key stakeholders across key institutions. There was minimal variation in the challenges of grid capacity (T1) and market uncertainties (OM2) across all cases, indicating their consistency and low sensitivity to variations in weights. The moderate fluctuations in K1, T2, T3, P1, OM1, and OG1 indicate that their rankings are moderately sensitive to methodological changes. The barriers T4, T6, E1, P2, P3, and OG2 showed significant fluctuations across all scenarios, highlighting their unstable ranking, which is dependent on weight variations.

5.2. Sensitivity Analysis for Mitigation Strategies

The sensitivity analysis of the TOPSIS technique was performed by varying the weights of the criteria to assess the stability of the prioritised solutions. The following adjustments were made, resulting in four cases/scenarios: In the first case, all criteria were assigned the same or equal weights. In case 2, the cost (C1) criterion was assigned a 20% weight, and the benefit (C2–C5) criteria were assigned an 80% weight. In case 3, only criterion C1 was assigned a 20% weight, and in case 4, only the benefit criteria were assigned a 20% weight. Presented in Figure 9 is the comparison of the ranking for all seventeen mitigation strategies under the different scenarios (case 0 is the original weights calculated by CRITIC in Section 3.2.1). A higher Ci ranking indicates the preferred strategy, while a lower ranking indicates the worst strategy. Strategies S3, S4, S5, S6, and S12 demonstrate stability across various cases, underscoring their robustness in mitigating barriers to renewable energy. There is a noticeable shift in strategies S9, S13, and S14, emphasising their sensitivity to weight adjustment.

6. Discussion

Data analysed through DEMATEL and CRITIC-TOPSIS techniques revealed the most significant barriers and mitigation strategies, respectively. This section interprets these barriers in the context of South Africa, examining the broader theoretical implications for integrating renewable energy in the country.

6.1. Barriers

The study analysed the barriers to renewable energy faced by Independent Power Producers in South Africa. These barriers are categorised into six categories: Agreement, Knowledge, Technological, Economic, Social, and Political. The findings of the MCDM framework and triangulation with the literature reveal the complex, multidisciplinary, and interrelated barriers that collectively hinder the integration of renewable energy into the energy system [91].

These categories of barriers are not mutually exclusive and do not necessarily exist in a linear sequence. Instead, they are often interlinked and overlapping, underpinning one another in complex ways [8].

6.1.1. Lack of Agreement

The agreement barrier, which emerged as the most dominant effect barrier, is rooted in both political and knowledge-related barriers, and its impacts cascade into economic and technological barriers. This was apparent when the lack of agreement between stakeholders (Eskom and the government) resulted in the delay of signing PPAs for the REIPPPP Bid Window (BW) 4 projects [92]. These disruptions, rooted in a lack of policy continuity and coordination between stakeholders (third-ranked barrier), significantly eroded investor confidence, limiting access to financing mechanisms and prompting several technology providers to withdraw from the South African market. Several years later, the impact is still felt and has escalated into the challenges of market uncertainties.

6.1.2. Market Uncertainties

Market uncertainties (second-most prominent barrier), on the other hand, are shaped by landscape-level dynamics, including policy continuity and economic pressures, which create an unstable environment for investment and hinder the acceleration of renewable energy projects. In the South African context, renewable energy development is particularly exposed to external market conditions, as projects rely heavily on imported technologies and equipment. These dependencies increase vulnerability to global currency fluctuations and volatile equipment prices, which can significantly elevate project costs. These uncertainties weaken investor confidence, increase financial risk and may delay project implementation, even in cases where technological feasibility and resource availability are favourable.

6.1.3. Lack of Coordination

The results also highlighted a lack of coordination as the third most prominent barrier, which continues to impede the large-scale integration of renewable energy in South Africa. Firstly, poor collaboration among ESKOM, DMRE, IPPs, and NERSA was evident in the awarding of BW 6 projects, where all wind projects could not proceed due to a lack of grid connection capacity in the Cape provinces [93].

Secondly, weak partnerships among institutions of higher learning, R&D facilities, industry, and government led to limited innovation and inadequate skills transfer. These insufficiencies not only hinder project implementation but also increase reliance on foreign technologies and expertise, reducing the long-term sustainability of renewable initiatives.

These findings highlight that challenges related to grid connection (currently a burning issue in SA’s renewable energy industry), along with policy continuity and financing, are not purely technical or financial but are rooted in institutional weaknesses among key stakeholders. Surprisingly, market uncertainty plays a more influential role than expected, ranking above some structural barriers. This suggests that investor perceptions and regulatory unpredictability are as critical as physical infrastructure constraints in driving renewable energy development and integration in South Africa.

Collectively, these findings reinforce the theoretical argument that a narrow focus on technical or economic constraints is insufficient to understand the dynamics of energy transitions.

The findings of this study vary from those of previous studies. According to the literature, the top three most significant barriers are within the economic/financial, political/policy, and technological/technical categories [52,94,95]. These studies have shown that the primary barriers to renewable energy development in developing countries are largely financial, including high investment costs and limited access to funding due to unstable economic conditions [32,96]. According to Mostafaeipour et al. [96] and James et al. [32], these financial limitations hinder the deployment of renewable energy projects in these countries. In addition, Chisale and Lee [4] and Pathak et al. [16] noted that weak political commitment and high levels of corruption and nepotism further impede the acceleration of renewable energy development. These governance challenges negatively influence policy formulation and implementation, delaying renewable energy expansion. Similarly, inadequate technological capacity and dependence on imported technologies further delay renewable energy projects in many developing countries [87].

These variations reflect the contextual differences in the maturity and structures of renewable energy markets across various regions. In South Africa, the renewable energy sector has evolved through the REIPPPP, attracting private investment and deploying advanced technologies. The primary challenges are rooted in governance and institutional coordination, largely shaped by the country’s centralised electricity system dominated by the state utility, Eskom.

This demonstrates that renewable energy barriers are highly contextual, and their impacts vary across regions depending on the predominant economic, social, political or geographical conditions, etc.

6.2. Mitigation Strategies

The results of the CRITIC-TOPSIS identified mitigation strategies to overcome the barriers to renewable energy faced by Independent Power Producers in South Africa. Although some of these strategies have already been implemented in the country, the renewable energy industry continues to face challenges.

6.2.1. Strengthening Local Community Engagement

The areas abundant in solar and wind resources in South Africa are often located in remote rural regions and on privately owned agricultural land, which, in some cases, is collectively owned by the community. Without the consent of these landowners, these projects become unfeasible. Community engagements should commence in the early stages of project development and be maintained throughout the entire project lifecycle. These engagements should be facilitated through transparent, structured, and legally grounded processes that involve local governance structures, such as community representatives and councillors.

These engagements will ensure that the community understands the socio-economic benefits of these projects, such as job creation, infrastructure development, skill enhancement, and project ownership, thereby fostering trust and transparency between developers and the community [31]. In South Africa, the REIPPPP provides established frameworks to ensure that the socio-economic benefits of these projects are both feasible and effectively implemented.

6.2.2. The Enforcement of Regulatory Frameworks and Independent Reviewers

The enforcement of regulatory frameworks and the involvement of independent reviewers are crucial to mitigating the challenges and inconsistencies experienced in the REIPPPP programme, such as delays in releasing bid windows, the awarding of preferred bidder status, and projects failing to reach financial close.

This will ultimately enhance and maintain the credibility of the REIPPPP, as noted by Kruger and Alao [97], and mitigate risks and delays, thereby enforcing transparency in the public procurement programme [13], which in turn creates a conducive environment for investment and innovation.

6.2.3. Raising Public Awareness

According to Alyamani et al. [36], public awareness is crucial for ensuring the success of renewable energy projects. In South Africa, as coal-fired power stations are gradually phased out and replaced by renewable energy plants [13], it becomes crucial for the government to implement public awareness campaigns that educate the public about the benefits of renewable energy for sustainable development. Such campaigns will not only inform the public about the environmental and economic benefits of clean energy but will also build trust and foster public acceptance of the transition. By ensuring that the public is informed and engaged, the government can strengthen societal support for its vision of transitioning towards renewable energy successfully [98].

These findings are inconsistent with the literature. Previous studies identified the top-ranked mitigation strategies as providing subsidies and incentives to make renewable energy projects more cost-competitive [1,18,38]. The study by Akpahou et al. [53], highlighted the importance of developing comprehensive, consistent policies to support renewable energy development. In addition, improving and strengthening access to funding mechanisms and expanding financing options were identified as critical strategies [36,94,99]. Lastly, investing in and supporting Research and Development (R&D) to drive innovation and address region-specific challenges can lead to technological breakthroughs, improved project efficiency, and significant cost reductions, thereby accelerating renewable energy development and integration [100,101].

7. Conclusions

South Africa has abundant renewable energy resources capable of generating clean, affordable, and sustainable energy. However, despite this potential, the country has not harnessed these resources to their full extent due to various barriers, including Agreement, Knowledge, Technical, Economic, Social, Political, Market, and Geographical constraints. This research proposed a hybrid multi-criteria decision-making framework to prioritise renewable energy barriers and mitigation strategies in South Africa. The prioritisation and cause-and-effect relationship assessment of the barriers hindering the adoption of renewable energy were conducted using the DEMATEL technique. The top three barriers were a lack of agreement, market uncertainties, and a lack of coordination between key institutions. These barriers collectively undermine the acceleration of renewable energy in South Africa. The lack of agreement between NERSA and NTCSA on grid allocation reflects broader policy discontinuity that undermines coordinated decision-making. This misalignment contributes to significant grid connection constraints. These constraints delay project implementation and create market uncertainty, which in turn erodes investor confidence and ultimately limits access to financing. These findings highlight that challenges related to grid connection (currently a burning issue in SA’s renewable energy industry), along with policy continuity and financing, are not purely technical or financial in nature but are rooted in institutional weaknesses among key stakeholders. Therefore, addressing technical barriers alone is insufficient without first addressing underlying coordination issues, which stem from a lack of policy continuity. Surprisingly, market uncertainty plays a more influential role than expected, ranking above some structural barriers. This suggests that investor perceptions and regulatory unpredictability are as critical as physical infrastructure constraints in driving renewable energy development and integration in South Africa.

Furthermore, the study offers a thorough and comprehensive CRITIC-TOPSIS framework for implementing strategies to overcome these barriers. The key strategies include strengthening community engagement, enforcing regulatory frameworks, appointing independent reviewers for the procurement programme, and raising public awareness. Implementing these strategies would enable South Africa to overcome the challenges associated with renewable energy development, thereby accelerating the integration of renewable energy into its energy systems.

This study contributes to the existing body of literature and industry knowledge in multiple ways. Firstly, it enhances the theoretical understanding of the barriers to renewable energy by providing context-specific causal analysis of barriers and objective prioritisation of mitigation strategies in South Africa using an integrated hybrid DEMATEL and CRITIC–TOPSIS approach. While previous studies have applied hybrid MCDM techniques such as DEMATEL-ANP [43] or CRTITIC-TOPSIS [41] in isolation or partial combination, this study advances the methodology by fully integrating causal analysis, objective weighting, and ranking within a single coherent framework.

Secondly, the findings of this research offer valuable insights for stakeholders in both the public and private sectors, enabling them to develop strategic plans and policies that address the identified barriers.

While the experts provided valuable insights into the barriers to renewable energy integration, the composition of the panel may have introduced bias. In particular, the dominance of private-sector perspectives may have led to the underrepresentation of public institutional stakeholders. Future research studies could address this methodological limitation by expanding expert engagement to include a more balanced representation of stakeholders across both the public and private sectors. In addition, the use of expert judgements can introduce bias, despite adopting a comprehensive mixed-methods approach that minimises subjectivity through the CRITIC technique. Therefore, future studies could extend this work by employing iterative consensus-building techniques such as the Delphi method, incorporating fuzzy set theory to handle uncertainty in judgements, using neutrosophic methods to handle ambiguity, or validating results through comparisons across different MCDM methods.