Policy Analysis of Low-Carbon Energy Transition in Senegal Using a Multi-Criteria Decision Approach Based on Principal Component Analysis

Abstract

Senegal has been investing in the development of its energy sector for decades. By using a novel multi-criteria decision analysis (MCDA) based on the principal component analysis (PCA) method, this paper develops an approach to determine the effectiveness of Senegal’s policies in supporting low-carbon development. This was determined using six criteria (C1 to C6) and 17 policies selected from the review of Senegal’s energy system. In order to determine the optimal weighting of the six criteria, a PCA is performed. In our approach, the best weighted factor is the normalized version of the best linear combination of the initial criteria with the maximum summarized information. Proper weighted factors are determined through the percentage of the information provided by the six criteria kept by the principal components. The percentage of information is statistically a fit of goodness of a principal component. The higher it is, the more statistically important the corresponding principal component is. Among the six principal components obtained, the first principal component (comp1) best summarizes the values of criteria C1 to C6 for each policy. It contains 81.15% of the information on energy policies presented by the six criteria and was used to rank the policies. Future research should take into account that when the number of criteria is high, the share of information explained by the first principal component could be lower (less than 50% of the total variance). In this case, the use of a single principal component would be detrimental to the analysis. For such cases, we recommend a higher dimensional visualization (using two or three components), or a new PCA should be performed on the principal components. This approach presented in our study can serve as an important benchmark for energy projects and policy evaluation.

1. Introduction

Over the last decade, an increasing number of multinational players have become involved in the energy transition debate. The establishment of Sustainable Development Goal (SDG) No. 7 is a fantastic example of this enthusiasm, especially because we have seen a significant growth in global mobilization for this subject over the last twenty years. The energy challenge of the twenty-first century is to make a fresh transition to a more sustainable energy system marked by universal access to energy services, as well as supply security and reliability from efficient, low-carbon sources [1].

Not only has the COVID-19 pandemic harmed public health, but it has also harmed the energy transition. To begin with, the economic crisis has increased the volatility of renewable energy markets. Second, falling fossil fuel prices have damaged renewable energy’s price competitiveness even further. Third, international trade barriers have hampered renewable energy supply chains, leaving many projects stranded. Last but not least, post-pandemic recovery plans continue to rely on fossil fuel investments, making the switch to renewable energy more difficult. Energy sustainability is a persisting concern; however, the challenge has been impacted differently under the unexpected interruption of COVID-19 [2]. Despite the obstacles that the energy transition faces, the pandemic also provides opportunity [3]. West African countries with huge potential might also take advantage of this opportunity to speed up their energy transition. Senegal, which holds a significant position in the region, has set goals to increase its energy efficiency and renewable energy strategy.

In fact, imported heavy fuel accounts for 90% of Senegal’s energy (excluding biomass). It is therefore committed to moving away from power generation based on diesel and toward less expensive energy sources. Senegal has also launched an intensive endeavor to create large volumes of biofuels, primarily to power electricity-producing units, and has sugarcane-based ethanol projects. Biomass and oil products account for the majority of final consumption, while the low power usage reflects the country’s electrification rate. Senegalese public authorities have chosen an energy mix policy in the sub-electric sector, making the promotion of renewable energies and energy efficiency a major and priority goal, due to a desire to strengthen energy independence, awareness raised by the ecological crisis, and opportunities offered by the Paris Agreement on Climate Change.

By offering a thorough analysis of Senegal’s low-carbon energy transition using a novel methodology (MCDA-PCA), this research aims to improve the existing literature on policy evaluation using MCDA. All the components of the nation’s energy policies are taken into consideration when conducting the analysis of Senegal’s energy system (electrification, renewable energy, energy efficiency and climate change). This paper develops a new and relevant combined MCDA-PCA approach for evaluating policy efficacy and identifying degrees of implementation success and impact for Senegal’s low-carbon progress.

The paper is structured as follows: The first section of the paper provides background information, the paper’s goals and objectives, as well as the research’s relevance and scope. Section 2 is the literature review. The methodology used in Section 3 is a review of energy policy and policy analysis, which includes MCDA based on PCA. Section 4 contains the findings, which include the outcomes of the energy policy review, a study of existing policies and their success to date, as well as the findings. The fifth section is the discussion, which includes the interpretation and importance of the findings of our analysis. The conclusion and recommendations from the analysis, as well as suggestions for further research, are presented in Section 6.

2. Literature Review

This section briefly summarizes the pertinent literature for our study.

In the field of sustainable energy management, multi-criteria decision making (MCDM) methodologies are gaining traction. The strategies are used to solve situations involving various objectives and conflicting goals. The process of making decisions in the face of numerous objectives is addressed by multi-criteria decision making (MCDM) methods. A decision-maker must pick between quantitative and non-quantifiable factors, as well as many criteria. Because the goals are frequently at odds, the solution is greatly dependent on the preferences of the decision-maker and must be a compromise [4].

The multi-criteria decision method can help with increasingly complicated energy management issues. The goal of traditional single-criteria decision making is usually to maximize benefits while minimizing expenses. These methods allow for a better understanding of the inherent characteristics of decision problems, promote the role of participants in decision-making processes, facilitate compromise and collective decisions, and provide a good platform for understanding model and analyst perceptions in a realistic scenario. By making decisions more explicit, reasonable, and efficient, the strategies aid in improving decision quality. The employment of these tools also makes it easier to negotiate, measure, and communicate priorities.

Multi-criteria decision making (MCDM) is a technique that utilizes many methods for evaluating the sustainability of renewable energy technologies [5].

Kumar et al. [6] offer an analysis of multi-criteria decision making (MCDM) for the development of sustainable renewable energy. They get insight into various MCDM methodologies, advancements gained when comparing renewable energy applications to MCDM techniques, and hopes for the future in this field. Using the MCDM technique, a thorough study of the field of sustainable energy has been conducted.

In order to support potential Swiss energy policy making, Volkart et al. [7] used a multi-criteria decision analysis of energy system transformation pathways for Switzerland. The outcomes of the scenario quantification using an energy system model were combined with sustainability analysis based on multiple criteria for this goal.

Shem et al. [8] examined Vietnam’s potential for a low-carbon energy transition using a systematic literature review and weighted sum multi-criteria decision analysis (MCDA). Policy taxonomies that were both chronological and hierarchical were used to present and analyze the results. The results show that policies need to include measures that support renewable energy technologies and also foster the mobilization of private investment or international cooperation in order to support a road towards low-carbon development. The study also shows a strong link between the development of the clean energy sector and the accomplishment of pertinent SDGs.

For the purpose of evaluating the viability of various policy scenarios in the Korean electrical industry, Choi et al. [9] created a multi-criteria decision making (MCDM) model connected to an energy system model. Their study demonstrates that while the new transition policy is not a desirable option in terms of overall cost and emissions, it may be so if we take other factors into account. Their research comes to the conclusion that lowering coal use is the most desired approach for the energy transition based on many trade-off assessments among various criteria.

Jonek-Kowalska et al. [10] used a multi-criteria decision analysis to evaluate the effectiveness of energy policy in Central and Eastern European countries in a long-term perspective. Environmental, energy resource, economic, and energy security conditions are all considered in the evaluation process’s multi-criteria analysis. The study was conducted with a 21-year horizon in view. The results suggested that most of the countries under study have made only little progress in their efforts to reform their energy systems. The findings indicated that abandoning non-renewable fuels and simultaneously substituting them with nuclear and hydroelectric energy is the most efficient method of transforming the energy system.

Thanh et al. [11] suggested using a fuzzy MCDM model to evaluate and choose a location for a solid-waste-to-electricity facility in Vietnam. A combined compromise solution (CoCoSo) model is used to rank the candidates in the last stage, and the fuzzy analytic hierarchy process (FAHP) technique is employed to examine the relative weight of the major and secondary evaluation factors in the first stage. Their research makes a contribution for decision-makers and investors in other renewable energy projects in Vietnam and around the world in analyzing and choosing suitable areas for solid-waste-to-energy projects.

A decision-making paradigm using a fuzzy set in multi-criteria situations has been presented out by Nhi et al. [12]. They employed two techniques: the fuzzy analytic hierarchy process (FAHP) and the weighted aggregated sum product assessment (WASPAS). The first step is to calculate the weights of each criterion and sub-criterion using the FAHP model. The WASPAS model then determines the order of each choice. Based on calculations from the above techniques for determining potential places for a geothermal power plant (GPP), a final alternative is decided to have the best resolution. This study has offered a technique for choosing an appropriate location by employing a fuzzy multi-criteria decision making model.

Since principal component analysis (PCA) is a technique for reducing the dimensionality of these data sets, increasing interpretability while minimizing the loss of information by creating new uncorrelated variables that successively maximize variance, it is sometimes used in scientific work. For example, in energy studies, Bhowmik et al. [13] used a principal component analysis to study the factors that influence how society views green energy. Asbahi et al. [14] used a principal component analysis approach to assess the national energy performance via the Energy Trilemma Index.

There have been several methods used for establishing attribute weights in MCDM, as evidenced by the literature review. Common MCDM models used in energy research include the analytical hierarchy process (AHP), the weighted sum method (WSM), preference ranking organization method for enrichment evaluation (PROMETHEE), the elimination and choice translating reality (ELECTRE), the technique for order preference by similarity to ideal solutions (TOPSIS), multi-attribute utility theory (MAUT), the fuzzy analytic hierarchy process (FAHP), weighted aggregated sum product assessment (WASPAS), etc.

In this research, we enrich the literature by combining a review of energy systems with a multi-criteria decision analysis (MCDA) based on principal component analysis (PCA) in order to avoid subjectivity in the weighting of criteria and determine the optimal weighted factor for a better policy evaluation. The principal components are defined as the best linear combination of the initial criteria because they are obtained through the optimization of the maximum information contained in these initial criteria, and they are also non-correlated, which eliminates the redundancy in the initial information. The MCDM based on PCA helps to find the best weighted factor. The best weighted factor is the standardized version of the best linear combination of the initial criteria with the maximum information summarized.

3. Methodology

The methodology in this paper is based on a review of energy policy and policy analysis with multi-criteria decision method (MCDM) based on principal component analysis (PCA). Figure 1 summarizes the research process, and each step is explained in further detail within the following section.

3.1. Comprehensive Review of Senegal’s Energy Policy

The first stage was to collect and aggregate data for subsequent policy analysis, which was carried out during the energy policy review. First, a detailed evaluation of Senegal’s policies and target status as an ECOWAS member was conducted. The widely used approach of systematic review was used to provide insight from current resources of Senegal’s energy condition in order to develop a credible, dependable, and legitimate review.

From this review, 6 criteria are selected as well as 17 policies to guide the MCDA-PCA approach.

Databases such as Science Direct and Research Gate, as well as open-access search engines such as Google and Google Scholar, were used to find sources. Because many reports from international organizations were included in the evaluation, open-access searches often yielded more interesting and relevant results.

3.2. Policy Analysis with Multi-Criteria Decision Method

Unlike classic MCDA, the innovation of our research is its combination with the PCA method. The goal of MCDM based on PCA is to find the best weighted factor. The best weighted factor is the standardized version of the best linear combination of the initial criteria with the maximum information summarized.

The following steps shown in

Table 1. Renewable capacity (current and projections).

	2010	2020	2030
Installed capacity of renewable-energy-based power plants in MW (including medium and large hydro)	68	403	632
Share of renewable energies in % of total capacity installed (including medium and large hydro)	10.9%	35.6%	31.8%
Total electricity production based on renewable energies in GWh (including medium and large hydro)	253	896	1501
Share of renewable energies in the electricity mix in % (including medium and large hydro)	10%	20%	23%

were performed to conduct our MCDA:

• Step 1: Policy options selection.

With the review of Senegal’s energy system, we selected a total of 17 policies, standards, and regulations that were introduced between 2006 and 2020 (see

Table 2. Summary of selected policies.

Energy Policy	Category
2006: The National Action Plan for Adaptation to Climate Change (NAPA)	Climate change
2007: Promote the use of biodiesel (PLAN REVA)	Renewable energy
2008: LPDSE 2008–2012	Electrification, renewable energy, energy efficiency
2009:National Biogas Program (PNB-SN)	Renewable energy
2011: Law on Renewable Energy	Renewable energy, energy efficiency
2012: LDPSE 2012–2017	Electrification, renewable energy, energy efficiency
2013: Agency for Economy and Energy Efficiency (AEME)	Energy efficiency
2013: National Renewable Energy Agency (ANER)	Renewable energy
2015: Intended national determined contribution	Climate change
2015:National Plan for Energy Efficiency (NEEAP)	Energy efficiency, electrification, renewable energy
2015: National Plan for Renewable Energy (NREAP)	Renewable energy, electrification, energy efficiency
2015: Economy of energy strategy (SMES)	Energy efficiency
2015: The Rural Electrification Priority Program (PPER)	Electrification
2015:Urgent National Plan for Rural Electrification (PNUER)	Electrification
2016: Policy letter for the environment and sustainable development sector (LP/SEDD)	Climate change
2019: LPDSE 2019–2023	Electrification, renewable energy, energy efficiency
2020: National determined contribution	Climate change

).

• Step 2: Identification of objectives and criteria.

Identify the objectives and criteria that indicate the value associated with each option’s outcomes. The comprehensive review helped to identify criteria that would guide policy analysis. The purpose to be reflected within each choice while examining criteria was efficacy in measuring implementation and consequent impact of energy-related policies for Senegal’s context. The criteria serve as pillars, providing a guide for the level of execution and ensuing impacts that policies in the energy sector have attained to date. Selected criteria are:

Criterion 1 (C1): Impact level of the policy;

Criterion 2 (C2): Relevance to achieving the national energy efficiency goals;

Criterion 3 (C3): Relevance to achieving the national renewable energy targets;

Criterion 4 (C4): Private investment or international cooperation;

Criterion 5 (C5): Quantitative targets;

Criterion 6 (C6): Implementation simplicity.

• Step 3: Measure the progress of each criterion for the relevant policy.

Describe how each alternative is expected to perform in relation to the criteria. To measure the progress of each criterion for the relevant policy, each policy will be evaluated against the criteria using a scoring system derived from a range of one (low) to four (high). The “high” preferred value was the best scenario for all criteria.

• Step 4: Apply the principal component analysis to get the overall value.

The principal components are defined as the best linear combination of the initial criteria because they are obtained through the optimization of the maximum information contained in these initial criteria, and they are also non-correlated, which eliminates the redundancy in the initial information.

The proper weighted factors are determined through the percentage of the information provided by the six criteria kept by the principal components. The percentage of information is statistically a fit of goodness of a principal component. The higher it is, the more statistically important the corresponding principal component is. The detailed steps of the principal component analysis are explained in Section 3.3.

3.3. Principal Component Analysis

The main idea of principal component analysis is to summarize and visualize the information from a data set of individuals or observations defined by a variety of quantitative characteristics that may or may not be connected.

In our research, we use the PCA method in combination with the multi-criteria analysis method, a new approach that we develop for a better evaluation of energy policies. The objective of using the combined MCDA-PCA method is to avoid the subjective weightings often encountered in the use of MCDA. The combination with the PCA method thus makes it possible to obtain the best weighting factor.

As part of the analysis of Senegal’s energy policies, a principal component analysis (PCA) is carried out in order to obtain optimal weights applicable to criteria C1 to C6. The principal components obtained at the end of the PCA are weighted final scores. Our analysis is inspired by Johnston [15].

Suppose we have data of p variables on n individuals or observations. We will be working on a matrix (as shown in Equation (1)) representing the data of the n observations of the p centered variables:

$$\mathrm{Z}=\left[\begin{array}{ccc}{\mathrm{z}}_{11}& \cdots & {\mathrm{z}}_{1\mathrm{p}}\\ ⋮& \ddots & ⋮\\ {\mathrm{z}}_{\mathrm{n}1}& \cdots & {\mathrm{z}}_{\mathrm{np}}\end{array}\right]=\left[{\mathrm{Z}}_1 \dots { \mathrm{Z}}_{\mathrm{p}}\right]$$

(1)

The matrix ${\mathrm{Z}}^{\mathrm{T}}\mathrm{Z}$ is symmetric and positive semi-definite.

We will consider transforming Z into a new matrix, ${\mathrm{Z}}^{\ast }$, whose columns are the principal components of Z. These principal components are linear combinations of the initial p variables (i.e., weighted sums of the initial p variables), which are pairwise uncorrelated. The number of principal components obtained is equal to the number of initial variables.

The first principal component will have the greatest possible variance, the second being, among the principal components which are not correlated with the first, the one which has the greatest possible variance, and so on.

Finding the first of these principal components denoted ${\mathrm{Z}}_1{}^{\ast }$ amounts to finding the vector ${\mathsf{\nu }}_1$ such that the projection of the observations of the matrix ${\mathrm{Z}}^{\mathrm{T}}\mathrm{Z}$ on ${\mathsf{\nu }}_1$ has a maximum variance. By adopting the matrix notation, ${\mathrm{Z}}_1{}^{\ast }$ is the projection of $\mathrm{Z}$ on ${\mathsf{\nu }}_1$, that is, ${\mathrm{Z}}_1{}^{\ast }={\mathrm{Z}\mathsf{\nu }}_1$.

${\mathrm{Z}}_1{}^{\ast }=\left[\begin{array}{c}{\mathrm{z}}_{11}{}^{\ast }\\ \dots \\ {\mathrm{z}}_{\mathrm{n}1}{}^{\ast }\end{array}\right]$ is a vector with $\mathrm{n}$ elements, and ${\mathrm{v}}_1=\left[\begin{array}{c}{\mathrm{v}}_{11}\\ \dots \\ {\mathrm{v}}_{1\mathrm{p}}\end{array}\right]$ is a vector with p elements.

Find the first principal component or the weighted factor ${\mathrm{Z}}_1{}^{\ast }$ which resumes the maximal information (maximum variance) displayed by these p variables, which requires resolving the optimization problem in Equation (2):

$$\max \mathbb{E}\left({\mathsf{\nu }}_1{}^{\mathrm{T}}\left({\mathrm{Z}}^{\mathrm{T}}\mathrm{Z}\right){\mathsf{\nu }}_1\right) \mathrm{subject} \mathrm{to} {\mathsf{\nu }}_1{}^{\mathrm{T}}{\mathsf{\nu }}_1=1$$

(2)

In fact, as shown in Equation (3), we seek to maximize the variance.

$${\mathrm{Z}}_1{}^{\ast }: \mathrm{Var}\left({\mathrm{Z}}_1{}^{\ast }\right)=\mathbb{E}\left({({\mathrm{Z}}_1{}^{\ast })}^{\mathrm{T}}{\mathrm{Z}}_1{}^{\ast }\right)=\mathbb{E}\left({\mathsf{\nu }}_1{}^{\mathrm{T}}\left({\mathrm{Z}}^{\mathrm{T}}\mathrm{Z}\right){\mathsf{\nu }}_1\right)$$

(3)

We must also impose a constraint on the vector ${\mathsf{\nu }}_1$, because otherwise the variance $\mathrm{Var}\left({\mathrm{Z}}_1{}^{\ast }\right)$ can be as large as we want. We are therefore going to normalize the vector ${\mathsf{\nu }}_1$, which amounts to setting ${\mathsf{\nu }}_1{}^{\mathrm{T}}{\mathsf{\nu }}_1=1$.

The optimization problem can be solved using the Lagrange optimization method. Consider the Lagrangian function defined in Equation (4) by:

$$\mathcal{L}={\mathsf{\nu }}_1{}^{\mathrm{T}}\left({\mathrm{Z}}^{\mathrm{T}}\mathrm{Z}\right){\mathsf{\nu }}_1+{ \mathsf{\lambda }}_1\left(1-{\mathsf{\nu }}_1{}^{\mathrm{T}}{\mathsf{\nu }}_1\right)$$

(4)

where ${\mathsf{\lambda }}_1\in ℝ$ is a Lagrange multiplier.

When we apply a differential on the Lagrange multiplier, the following Equation (5) appears:

$$\begin{array}{ll}\left\{\begin{array}{c}\frac{\partial \mathrm{L}}{\partial {\mathrm{v}}_1}=0\\ \frac{\partial \mathrm{L}}{\partial {\mathsf{\lambda }}_1}=0\end{array}\right.& \Rightarrow \left\{\begin{array}{c}2\left({\mathrm{Z}}^{\mathrm{T}}Z\right){\mathrm{v}}_1-2{\mathsf{\lambda }}_1{\mathrm{v}}_1=0\\ {\mathrm{v}}_1{}^{\mathrm{T}}{\mathrm{v}}_1=1\end{array}\right.\\ & \Rightarrow \left\{\begin{array}{c}\left({\mathrm{Z}}^{\mathrm{T}}Z\right){\mathrm{v}}_1-{\mathsf{\lambda }}_1{\mathrm{v}}_1=0\\ {\mathrm{v}}_1{}^{\mathrm{T}}{\mathrm{v}}_1=1\end{array}\right.\\ & \Rightarrow \left\{\begin{array}{c}\left({\mathrm{Z}}^{\mathrm{T}}Z\right){\mathrm{v}}_1={\mathsf{\lambda }}_1{\mathrm{v}}_1\\ {\mathrm{v}}_1{}^{\mathrm{T}}{\mathrm{v}}_1=1\end{array}\right.\end{array}$$

(5)

We can see that the optimal solution ${\mathsf{\nu }}_1$ is the eigenvector associated with ${\mathsf{\lambda }}_1$, the largest eigenvalue of ${\mathrm{Z}}^{\mathrm{T}}\mathrm{Z}$.

For the determination of another principal component, ${\mathrm{Z}}_{\mathrm{i}}{}^{\ast }$ ($2\le \mathrm{i}\le \mathrm{p}$), it is to consider the maximization of ${\mathsf{\nu }}_{\mathrm{i}}{}^{\mathrm{T}}\left({\mathrm{Z}}^{\mathrm{T}}\mathrm{Z}\right){\mathsf{\nu }}_{\mathrm{i}}$ under the constraints ${\mathsf{\nu }}_{\mathrm{i}}{}^{\mathrm{T}}{\mathsf{\nu }}_{\mathrm{i}}=1$ (${\mathsf{\nu }}_{\mathrm{i}}$ is unitary) and $\mathrm{Cov}\left({\mathsf{\nu }}_{\mathrm{i}}, {\mathsf{\nu }}_{\mathrm{j}}\right)=0 \left(\mathrm{j}<\mathrm{i}, 2\le \mathrm{j}\le \mathrm{p}\right)$. This second constraint explains the non-correlation of the other previously determined eigenvectors with ${\mathsf{\nu }}_{\mathrm{i}}$ or the non-correlation the other previously determined components with ${\mathrm{Z}}_{\mathrm{i}}{}^{\ast }$.

The associated optimization problem (Equation (6)) will be:

$$\max \mathbb{E}\left({\mathsf{\nu }}_{\mathrm{i}}{}^{\mathrm{T}}\left({\mathrm{Z}}^{\mathrm{T}}\mathrm{Z}\right){\mathsf{\nu }}_{\mathrm{i}}\right) , \mathrm{subject} \mathrm{to} \left\{\begin{array}{c}{\mathsf{\nu }}_{\mathrm{i}}{}^{\mathrm{T}}{\mathsf{\nu }}_{\mathrm{i}}=1\\ {\mathrm{v}}_1{}^{\mathrm{T}}{\mathrm{v}}_{\mathrm{i}}=0\\ \dots \\ {\mathrm{v}}_{\mathrm{j}}{}^{\mathrm{T}}{\mathrm{v}}_{\mathrm{i}}=0\end{array}\right., \mathrm{with} \mathrm{j}<\mathrm{i}, 2\le \mathrm{j}\le \mathrm{p}$$

(6)

By still applying the resolution of the problem according to the Lagrange optimization method in Equation (7), ${\mathrm{Z}}_{\mathrm{i}}{}^{\ast }\left(2\le \mathrm{i}\le \mathrm{p}\right)$ is also determined by the vector ${\mathsf{\nu }}_{\mathrm{i}}$.

We have

$$\mathcal{L}={\mathrm{v}}_{\mathrm{i}}{}^{\mathrm{T}}\left({\mathrm{Z}}^{\mathrm{T}}\mathrm{Z}\right){\mathrm{v}}_{\mathrm{i}}+{\mathsf{\lambda }}_{\mathrm{i}}\left(1-{\mathrm{v}}_{\mathrm{i}}{}^{\mathrm{T}}{\mathrm{v}}_{\mathrm{i}}\right)+{\mathsf{\gamma }}_1{\mathrm{v}}_1{}^{\mathrm{T}}{\mathrm{v}}_{\mathrm{i}}+\cdots +{ \mathsf{\gamma }}_{\mathrm{j}}{\mathrm{v}}_{\mathrm{j}}{}^{\mathrm{T}}{\mathrm{v}}_{\mathrm{i}},$$

(7)

where ${\mathsf{\lambda }}_2,{\mathsf{\gamma }}_1,\dots ,{ \mathsf{\gamma }}_{\mathrm{j}}\in ℝ$ are Lagrange multipliers, ($\mathrm{j}<\mathrm{i}, 2\le \mathrm{j}\le \mathrm{p}).$

Then, in Equation (8), we apply a differential on the Lagrange multiplier to obtain Equations (9) and (10):

$$\left\{\begin{array}{c}\frac{\partial \mathcal{L}}{\partial {\mathrm{v}}_{\mathrm{i}}}=0 \Rightarrow \left({\mathrm{Z}}^{\mathrm{T}}Z\right){\mathrm{v}}_{\mathrm{i}}+{\displaystyle {\displaystyle \sum }_{\mathrm{j}=1}^{\mathrm{i}-1}}\frac{{\mathsf{\gamma }}_{\mathrm{j}}}{2}{\mathrm{v}}_{\mathrm{j}}={\mathsf{\lambda }}_{\mathrm{i}}{\mathrm{v}}_{\mathrm{i}}\\ \frac{\partial \mathcal{L}}{\partial {\mathsf{\lambda }}_{\mathrm{i}}}=0 \Rightarrow {\mathrm{v}}_{\mathrm{i}}{}^{\mathrm{T}}{\mathrm{v}}_{\mathrm{i}}=1\\ \frac{\partial \mathcal{L}}{\partial {\mathsf{\gamma }}_{\mathrm{j}}}=0 \Rightarrow {\mathrm{v}}_{\mathrm{i}}^{\mathrm{T}}{\mathrm{v}}_{\mathrm{j}}=0, \forall \mathrm{j}=1,\dots ,\mathrm{i}-1\end{array}\right.$$

(8)

$$\begin{array}{c}\frac{\partial \mathcal{L}}{\partial {\mathrm{v}}_{\mathrm{i}}}=0 \Rightarrow \left({\mathrm{Z}}^{\mathrm{T}}\mathrm{Z}\right){\mathrm{v}}_{\mathrm{i}}+{\displaystyle \sum }_{\mathrm{j}=1}^{\mathrm{i}-1}\frac{{\mathsf{\gamma }}_{\mathrm{j}}}{2}{\mathrm{v}}_{\mathrm{j}}={\mathsf{\lambda }}_{\mathrm{i}}{\mathrm{v}}_{\mathrm{i}}\\ \Rightarrow {\mathrm{v}}_{\mathrm{j}}{}^{\mathrm{T}}\left({\mathrm{Z}}^{\mathrm{T}}\mathrm{Z}\right){\mathrm{v}}_{\mathrm{i}}+{\displaystyle \sum }_{\mathrm{j}=1}^{\mathrm{i}-1}\frac{{\mathsf{\gamma }}_{\mathrm{j}}}{2}{\mathrm{v}}_{\mathrm{j}}^{\mathrm{T}}{\mathrm{v}}_{\mathrm{j}}={\mathsf{\lambda }}_{\mathrm{i}}{\mathrm{v}}_{\mathrm{j}}{}^{\mathrm{T}}{\mathrm{v}}_{\mathrm{i}}\\ \Rightarrow { \mathsf{\lambda }}_{\mathrm{j}}{\mathrm{v}}_{\mathrm{j}}^{\mathrm{T}}{\mathrm{v}}_{\mathrm{i}}+\left(\mathrm{i}-1\right)\frac{{\mathsf{\gamma }}_{\mathrm{j}}}{2}={\mathsf{\lambda }}_{\mathrm{i}}{\mathrm{v}}_{\mathrm{j}}^{\mathrm{T}}{\mathrm{v}}_{\mathrm{i}} \mathrm{due} \mathrm{to} \left({\mathrm{Z}}^{\mathrm{T}}\mathrm{Z}\right){\mathrm{v}}_{\mathrm{j}}={\mathsf{\lambda }}_{\mathrm{j}}{\mathrm{v}}_{\mathrm{j}} \Rightarrow { \mathrm{v}}_{\mathrm{j}}^{\mathrm{T}}\left({\mathrm{Z}}^{\mathrm{T}}\mathrm{Z}\right)={\mathsf{\lambda }}_{\mathrm{j}}{\mathrm{v}}_{\mathrm{j}}^{\mathrm{T}})\\ \Rightarrow {\mathsf{\gamma }}_{\mathrm{j}}=\frac{2}{\left(\mathrm{i}-1\right)}\left({\mathsf{\lambda }}_{\mathrm{i}}-{\mathsf{\lambda }}_{\mathrm{j}}\right){\mathrm{v}}_{\mathrm{j}}^{\mathrm{T}}{\mathrm{v}}_{\mathrm{i}}=0\end{array}$$

(9)

due to ${\mathrm{v}}_{\mathrm{i}}^{\mathrm{T}}{\mathrm{v}}_{\mathrm{j}}=0={\mathrm{v}}_{\mathrm{j}}^{\mathrm{T}}{\mathrm{v}}_{\mathrm{i}}, \forall \mathrm{j}=1,\dots ,\mathrm{i}-1$.

So, in Equation (10), we can see that

$$\frac{\partial \mathcal{L}}{\partial {\mathrm{v}}_{\mathrm{i}}}=0 \Rightarrow \left({\mathrm{Z}}^{\mathrm{T}}\mathrm{Z}\right){\mathrm{v}}_{\mathrm{i}}={\mathsf{\lambda }}_{\mathrm{i}}{\mathrm{v}}_{\mathrm{I}}$$

(10)

which proves that the vectors ${\mathrm{v}}_{\mathrm{i}} \left(1\le \mathrm{i}\le \mathrm{p}\right)$ are the other eigenvectors associated with the i-th largest eigenvalues of ${\mathrm{Z}}^{\mathrm{T}}\mathrm{Z}$, and these eigenvectors are orthogonal (non-correlated).

In our context, the n observations represent the 17 policies of Senegal and the p variables are the 6 criteria (C1 to C6). We will find the weighted factors which are equal to the principal components of the scaled data matrix.

4. Results

The results of the policy review and MCDM based on PCA are presented in this section.

4.1. The Status Quo in Terms of Policy and Targets

The Letter of Policy Development of the Energy Sector (Lettre de Politique de Développement du Secteur de l’Energie, LPDSE) governs Senegal’s energy sector. It was first published in 1997 and amended in 2002, 2008, and 2012. The LPDSE lays out bold goals for enhancing the sector’s performance over the medium and long term. The policy’s overall goal is to increase the dependability and affordability of modern electricity services in a long-term way [16].

Some of Senegal’s current energy strategies and aims are broken down into four areas, including:

• Electrification—universal access to electricity by 2025.
• Renewables—23% by 2020 as total electricity production based on renewable energies in GWh (including medium and large hydro); the targets for REs injected into the grid are 30% in 2025 and 2030.
• Energy efficiency—by 2030, network lighting penetration rate to 99%; off-grid lighting penetration rate of 1%; number of devices with labels in effect of six; percentage of industries using efficiency methods equal to 80%; and percentage of energy savings in industry equal to 25%
• Climate change—emission reductions from their predicted path are 4% and 5% in 2025 and 2030, respectively, under the unconditional option (INDC). The conditional option (INDC+) is predicted to reduce emissions by 15% to 21% over the same time period.

4.1.1. Electrification

Senegal plans to attain universal access to electricity by 2025, according to ASER projections [17], with a 60 percent rural electrification milestone in 2019. The IEA projections (IEA, 2022), as indicated in the Figure 2, confirm this. Some rapid improvement can be seen in access to energy in the country. This is due to the implementation since 2012 of the Development Policy Letter of the Energy Sector (LPDSE) and accelerated in 2015 by the National Renewable Energy Action Plan (NREAP) and the National Energy Efficiency Action Plan (NEEAP), which have enabled a large part of the population to access energy.

4.1.2. Renewables

Senegal is heavily reliant on fossil fuels to supply its energy demands, similar to many other West African nations, despite the country’s vast potential for renewable energy, especially in the area of solar energy.

The nation has recently undertaken concerted attempts to diversify its energy supply; yet, despite the high priority accorded to the promotion of renewable energies, the industry still faces numerous enduring difficulties.

Frameworks, insufficient regulatory and quality control, limited actor cooperation, insufficient finance, and a lack of technically trained employees to ensure the viability of the facilities are all obstacles to the spread of renewable energy sources. Additionally, until recently, the general population was either unaware of or underutilized the advantages received from these sources of energy [20].

Senegal’s National Renewable Energy Action Plan (2015) seeks to increase renewable energy (including hydropower share)’s of grid-connected capacity to about 31.8% by 2030 (over 630 MW). Renewable energy capacity accounted for 20% of total capacity in 2020 (280 MW) (see Table 1). Between 2015 and 2020, with the implementation of the NREAP, the installed renewable-electricity-generating capacity has increased from 2.33 to 14.69 watts per capita (Figure 3). To encourage renewable energy development and rural electrification, some renewable energy equipment (solar panel systems, solar pumping kits, biogas stoves, etc.) has been free from VAT since August 2020.

The TPED in 2018 was 4314 Ktoe. According to IEA estimations, over 40% of total primary energy demand (TPED) will be met by renewable energy resources in West Africa by 2040 [21], as indicated in Figure 4.

The risk of jeopardizing Senegal’s adherence to its pledge to reduce gas emissions by 21% by 2020 under the Paris Climate Agreement exists as a result of a significant oil and gas sector involvement. Therefore, a significant part in strengthening the energy sector will also be played by renewable energy. In Figure 5, total primary energy supply (TPES) is depicted in current and predicted scenarios, with additional low-carbon technologies taking a prominent role in 2030 and 2040.

TPES of non-renewable and renewable sources in 2013 and 2018 is shown in Figure 6. The National Action Plan for Renewable Energies (PANER) establishes the following goals for REs injected into the grid between 2025 and 2030:

• In 2025, 440 MW of electricity from various sources (PV and wind), 30% penetration rate, and no constraints to evacuate the total power, according to the results of the REs’ integration study, taking into account the planned network at the time.
• In 2030, I 550 MW of power from various sources (PV and wind), a penetration rate of 30%, the need to improve the 90 kV transmission network (Cap-Des-Biches—Kounoune), and the installation of a third 200 MVA 225/90 kV transformer [22].

4.1.3. Energy Efficiency

Senegal’s per capita energy consumption is 0.27 toe (down 5.8% from 2019), with around 276 kWh of electricity (2020). In 2020, the population grew by 2.8 percent, following the same annual rate as the previous decade.

Total primary energy consumption has increased at a high rate (+3.7 percent per year on average since 2000) and has nearly doubled to 4.7 Mtoe in 2019. Due to the COVID-19 pandemic, primary consumption fell by 3.2 percent to 4.6 Mtoe in 2020.

Oil will supply 55% of the country’s energy needs in 2020, followed by biomass (mostly wood) at 36% and coal at 7% [22]. We can see that it is fair to admit that, in the instance of Senegal, talking about energy efficiency measures seems impossible given the country’s low per capita energy use. Nonetheless, it is vital to remember that these low per capita estimates reflect the reality that a huge portion of the country’s population lacks access to conventional energy (electricity and oil products). As shown in Figure 7, the National Energy Intensity in MJ per 2017 USD has decreased from 4.19 in 2010 to 3.67 in 2019. Despite Senegal’s repeated efforts to enhance energy efficiency, the process has yet to achieve a substantial size due to a number of institutional, financial, commercial, and political barriers that prevent energy efficiency adoption. Despite the lack of an official gauge of success to date, one can be dubious of the country’s aim of 2030.

4.1.4. Climate Change

CO₂ emissions in the country were 9 MtCO₂ in 2018 and are predicted to rise to 19 MtCO₂ in 2030 and 30 MtCO₂ in 2040 (Figure 8). As part of the “Plan Sénégal Emergent” [17], the country proposed its first nationally determined contribution (NDC) in 2020 (République du Sénégal, 2020), which was based on its prior INDC (2015). Senegal pledges to cut GHG emissions by 5% by 2025 and 7% by 2030, respectively, compared to baseline levels (unconditional objective); conditional targets are 23.7 percent by 2025 and 29.5 percent by 2030. The amount of money saved will be determined by the availability of international funding, capacity building, and technology transfer.

4.2. Summary of Energy Policy Review

Senegal’s energy policy (Table 2) repertory contains qualitative and quantitative targets and objectives for climate change, efficiency, electrification, and renewables, extending from 2020 to 2040, according to a complete policy assessment. In order to compare the efficacy of present policies, 17 energy policies options from the four major energy policy areas are listed in this section.

These possibilities are considered inputs for multi-criteria decision analysis. Apart from climate change policies, the table shows that renewable energy, energy efficiency initiatives, and electrification solutions are largely considered. In Section 4.3, the effectiveness of selected policies is discussed. The following is a short summary of each policy selected.

• LDPSE 2012–2017:

The new energy strategy, supported by the Development Policy Letter of the Energy Sector (LPDSE 2012–2017) signed on 31 October 2012, pursues a number of goals after conducting an analysis of the local, national, and global contexts as well as the sectoral vision. It does this by ensuring that the nation has an adequate supply of energy that is of the highest possible quality, durability, and affordability; it also does this by implementing energy diversification strategies to lessen the country’s vulnerability to exogenous risks, particularly those related to the global oil market. Additionally, it encourages the growth of renewable energy sources, increases access to contemporary energy services for more people by assuring a more equal distribution of resources that benefits vulnerable groups and underserved areas, and encourages energy management and efficiency. The energy efficiency component of the Energy Sector Development Letter (LPDSE 2012–2017) sets the following objectives: a reduction of 10 to 20% in the public electricity bill and a saving of energy of around 40% on electricity demand in 2020 by implementing the measures listed in the electricity demand management study (DSM) [16].

• 2015: National Energy Efficiency Action Plan (NEEAP):

The National Energy Efficiency Action Plan (NEEAP) defines the following objectives in terms of energy efficiency to achieve by 2030: (i) a network lighting penetration rate of 99%; (ii) lighting penetration rate off-grid by 1%; (iii) number of devices with active tags equal to six; (iv) percentage of industries having applied efficiency measures equal to 80%; and (v) percentage of energy savings in industry equal to 25% [23].

• 2015: National Renewable Energy Action Plan (NREAP):

In this perspective and with a view to contributing to the sustainable development of Senegal, the following priority areas have been selected: increase the contribution of renewable energies in meeting national energy needs and contribute to reducing dependence on fossil fuel imports, taking into account the preservation of the environment; strengthen access to modern energy services from renewable sources, making them affordable and sustainable, with a view to ensuring the well-being of populations and the reduction of poverty; set up a support system with a view to generating energy renewables as a lever for harmonious socio-economic development; set up a communication plan and a monitoring/evaluation system [24].

• National Adaptation Action Plan (NAPA) 2006:

The objective is to carry out a participatory and integrated study and analysis of the vulnerability of the different regions of Senegal facing the negative impacts of climate change on sectors keys to our economy, which are, among others, water resources, agriculture, and coastal areas. It highlights places most at risk from climate change and outlines the Government of Senegal’s priority industries and programs. The NAPA also emphasizes how critical it is to integrate efforts to combat climate change into all pertinent national ministries, organizations, and policies. Improved national coordination among stakeholders addressing climate change challenges is a priority in order to prevent duplication of activities, maximize resources, and achieve the best results. Additional broad priority areas for adaptation include research, systematic observation, and capacity building. These categories are based on climate projections, sector-specific studies, and a study of diverse development difficulties. Senegal’s environmental, climate change, and catastrophe risk reduction laws and regulations must be developed, improved, and implemented in order to reduce climate vulnerability [25].

• 2007: (PLAN REVA) Promote the use of biodiesel:

Senegal started the National Jatropha Programme (NJP) in 2006 with the ultimate objective of producing enough diesel from jatropha to meet all of the nation’s demands beginning in 2012 [18].

• 2011: Law on Renewable Energy (SENEGAL):

This law covers uses for renewable energy sources, as well as their extraction, storage, and commercialization. It encompasses all fields of renewable energy as well as the safety and security of those fields [19]

• 2008: LPDSE 2008–2012:

The Government’s new energy policy developed in this policy letter highlights the following three major objectives: (i) ensure the energy supply of the country in sufficient quantity, under the best conditions of quality and durability and at the least cost, (ii) expand people’s access to modern energy services, and (iii) reduce the vulnerability of the country to exogenous hazards, particularly those of the world oil market.

With regard to the strategy, the Government has adopted the following main axes: development and exploitation of national energy potential, in particular in the field of biofuels and renewable energies; energy diversification through the mineral coal sector, biofuel, biomass, solar, wind, etc., for the production of electricity; the increased use of hydroelectricity within the framework of regional cooperation, especially within river basin organizations and WAPP; securing the country’s hydrocarbon supply by strengthening the local refining and cooperation with oil-producing countries; adaptation of energy infrastructure to demand based on mains public and private sector; the acceleration of access to electricity, in particular with the promotion of rural electrification and the development of energy services for the satisfaction productive and social activities; controlling energy demand and improving energy efficiency; consolidation of energy sector governance; the restructuring of the electricity sub-sector for greater efficiency and judicious involvement of the private sector; and consolidation of the policy for the management of forest resources with a view to sustainable supply of populations with domestic fuels [16].

• 2009: National Biogas Program (PNB-SN):

Senegal has had a national domestic biogas program (PNB-SN) since 2009, which encourages and oversees the technology’s spread throughout the nation [26].

• 2013: Agency for Economy and Energy Efficiency (AEME):

The AEME is responsible for implementing the policy of the State of Senegal in terms of energy, with the following main missions: the proposal of the national energy management strategy; identification, evaluation, and exploitation of potential energy savings in the various sectors of activity; advice and technical and/or financial assistance for the rationalization of energy consumption; conducting and evaluating the implementation of energy-saving programs and energy efficiency; the promotion of standards and regulations related to the rational use of energy and energy self-sufficient equipment; and the establishment of information, communication, and raising awareness among professionals and the general public [27].

• 2013: National Renewable Energy Agency (ANER):

ANER was founded by Decree No. 2016-684 on 17 May 2013, with the primary goal of promoting renewable energy across all activity sectors [28].

• 2015: Intended national determined contribution:

Emission reductions from baseline forecasts under the unconditional scenario (INDC) will be 3%, 4%, and 5% in 2020, 2025, and 2030, respectively. For the same years, the conditional scenario (INDC+) anticipates emission reductions of 7%, 15%, and 21%, respectively [29].

• 2015: Economy of energy strategy (SMES):

The project is divided into different phases. These include in particular the general introduction to the energy sector of Senegal; the presentation of the state of the art of energy management in Senegal, the analysis of the energy supply, and the energy demand situation; the energy demand projection; the benchmark of energy management policies and strategies in other countries, from definition of the strategy; the presentation of the action plan; and the upgrading of the AEME [27].

• 2015: The Rural Electrification Priority Program (PPER):

The objective was to achieve a rural electrification rate of 60% nationally in 2017, with a minimum of 30% per department, with a view to achieve universal access to electricity by 2025, one of the priorities of the Senegal Emerging Plan [30]. In 2018, the rural electrification rate was only 41.42%.

• 2015: Urgent National Plan for Rural Electrification (PNUER):

The PNUER, or “Programme National d’Urgence d’Electrification Rurale”, is a comprehensive urgency plan for the years 2015 to 2017 that is divided into different parts [31].

• 2016: Policy letter for the environment and sustainable development sector (LP/SEDD):

By incorporating sustainable development concepts into policies and boosting population resilience to climate change, the strategy intends to build a national momentum for better management of the environment and natural resources. Two strategic axes are determined in relation to this overarching goal: (1) environment and natural resource management and (2) support for sustainable development [32].

• 2019: LPDSE 2019–2023:

The following vision is stated in the Energy Development Policy Letter (LPDSE) 2019–2023, which was released in October 2019. It states that the goal is to “strengthen access for all to energy of sufficient quality and quantity at a lower cost and in a sustainable manner”, with the overall objective being to “have sustainable, quality energy that respects the environment and is the foundation for the emergence of the country” [33].

• 2020: National determined contribution:

The NDC for Senegal includes a target that is both unconditional and dependent on outside assistance. It is predicated on a BAU scenario with 2010 as the base year. Senegal proposes its mitigation by designating the years 2025 and 2030 as the midpoint and the ultimate goal, respectively [34].

4.3. Policy Examination Results

At the end of the PCA, the first principal component (comp1) is the one best summarizing the values of the criteria C1 to C6 for each policy: it retains 81.15% of the information of the energy policies presented by the criteria C1 to C6.

Figure 9 below shows the explanatory power of each of the six principal components obtained. The highest scores were obtained by the NREAP and NEEAP policies.

The final weighted “comp1” score will be used to rank Senegal’s energy policies as shown Figure 10.

The final weighted score “comp1” ranks the 2007 PLAN REVA policy as the weakest, with a score value below 1%, and the 2015 NREAP policy as the one with the highest score value.

The following supporting mechanisms and mandates received the highest grades for implementation and impact:

• 2015: NREAP;
• 2015: NEEAP;
• 2012: LDPSE 2012–2017;
• 2020: National determined contribution (NDC);
• 2015: PNUER.

The final weighted score “comp1” ranks the 2007 PLAN REVA policy as the weakest, with a score value below 1%, and the 2015 NREAP policy as the one with the highest score value (close to 100%).

• Generally, when a policy has at least one criterion whose value is superior to another, it is ranked above the other by the first principal component; this was observed for 12 of the 17 policies.

However, for policies such as NAPA, the Law on Renewable Energy, LPDSE 2019–2023, AEME, and PLAN REVA, reasoning must be applied:

• The AEME policy is ranked above the PLAN REVA policy by the first principal component because its value of criterion C6 is higher than that of PLAN REVA.
• The previous remark also applies to the classification of the NAPA and the Law on Renewable Energy policy.
• Concerning the ranking of the Law on Renewable Energy policy ahead of the LPDSE 2019–2023 policy, this is linked to its higher values of criteria C1 and C2.

The quantitative data from the first two policies, the NREAP and the NEEAP, show some demonstrable impact (criterion C1), achieving the national energy efficiency goals (criterion C2), increasing RE technology (criterion C3), and some policy implementation success (criterion C6). It has been possible to accomplish feats, particularly with regard to reducing dependence on fossil fuels, diversifying the energy mix, reducing GHG emissions and relevance to achieving the national energy efficiency goals (criterion C2) and facilitating the production of domestic energy, thanks to the establishment of the legal framework for the sustainable development of renewable energies and increasing RE technology (criterion C3). Additionally, it has made it easier to construct and control incentive schemes, particularly for the development of renewable energies. In 2018, 1.5 million Senegalese people profited from solar home systems and mini-grids, which dominate decentralized renewable energy solutions in the nation, thanks in part to the NREAP. With a rural electrification percentage that increased from 27% to 42.3% between 2012 and 2018, Senegal has 78,000 more individuals who benefit from mini-grids than most other West African countries [33].

In our analysis, it must be noted that, after considering the policies’ (2012: LDPSE 2012–2017, 2015: PNUER) initial phase, particularly the years 2015 to 2017, the following are the crucial metrics for gauging the success of the project: out of a total of 21,170 places, there are 4138 rural localities that have electricity; out of a total rural population of 8,644,376, or a 52% coverage rate, this allowed electricity to be made available to 4,460,052 persons living in electrified places; and 42% is thought to be the access rate [35]. It is important to note that in these two projects, a number of places have been electrified using renewable energies, particularly solar energy. NDC takes the fourth position followed by PNUER. Since some impact was measurable, each of these programs showed shortcomings, obstacles, and gaps in the quantitative data acquired as well as private investment and international cooperation. Less progress has been noted for lower-ranked initiatives such as PLAN REVA (2007) and AEME (2013). Both of these policies mostly point to challenges in achieving energy efficiency targets (criterion C2), renewable energy targets (criterion C3), and implementation (criterion C6).

5. Discussion

Of the seventeen policies analyzed using the MCDA-PCA approach, nine obtained a final score below 50. According to our findings, the final weighted score “comp1” ranks the 2007 PLAN REVA policy as the weakest, with a score value below 1%, and the 2015 NREAP policy as the one with the highest score value (close to 100%). Three policies received a score higher than 80 (yellow), while eight received a score of 0 to 40 (blue).

The loss of information (18.85% of the total variance) related to the use of the first principal component is, in fact, a brake on the inference of the rules related to the ranking of policies in the sense that we do not obtain an easily generalized rule for all policies. Instead, we obtain rules by classes of policies (detection of classes of policies), which minimizes the information related to the comparison of these classes of policies with each other. Generally, when a policy has at least one criterion whose value is superior to another, it is ranked above the other by the first principal component; this was observed for 12 of the 17 policies with the exception policies such as NAPA, the Law on Renewable Energy, LPDSE 2019–2023, AEME, and PLAN REVA.

Economic and technological capacities for RE growth can be provided through private and international support. A higher score for criterion C4 could indicate that RE potential and implementation status are improving. To date, the country has concentrated on financial mechanisms for renewable energy technologies, private sector incentives, and government subsidies for international cooperation. In principle, these appear appropriate, yet there is a gap in terms of implementation simplicity and impact. To indicate meaningful implementation, a policy must exhibit significant features in the majority of criteria (one through six). Effective implementation is critical for progress toward a cleaner future and, as a result, a road to zero emissions; nevertheless, there are several variables that Senegal must consider in order to accomplish successful implementation. Our analysis reveals that electrification and renewable energy projects account for the majority of projects with a score greater than 60. This is due to the country’s attempts to boost electricity generation and promote renewable energy sources. The criteria used are appropriate for the direction in which Senegal is heading and suggest that policy mechanisms are in place; nonetheless, the success of these mechanisms, as well as increased action, is required to support them. We also observe that, according to recent policies, renewable energy penetration in Senegal’s energy supply plan has become a reality. However, several flaws can be found due to a variety of factors, including the lack of or ambiguity of a household auto-consumption policy and insufficient encouragement of RE use at the household level (agricultural products conservation, livestock and fishing, grain milling, water heating, etc.). The significant takeaway from our analysis is that projects with a score of less than 50 are disproportionately focused on energy-saving strategies. Energy efficiency is critical to ensuring Senegal’s energy transition, as stated in SDG 7.3, and is one of the most important areas for improvement. Improving energy efficiency is critical to Senegal’s success in the low-carbon energy transition. Additionally, despite the country’s lofty goals in terms of energy access and renewable energy, progress in energy efficiency has been slow, as evidenced by the policies we examined, with most energy efficiency projects having obtained a result below 60 with the exception of the NEEAP, which has enabled Senegal to make progress in this area. The weaknesses discovered are a lack of norms and standards that are not being adequately applied and monitored. Incentives are insufficient to compensate for the budgetary limitations of end users. Senegal could take a personalized approach to energy efficiency in this sense by combining regulatory provisions, information actions, and incentive measures, as well as adapting best practice recommendations that have proven successful in other countries.

In general, our examination of the chosen policies reveals the following flaws: there are no statistics on the long-term viability and acceptability of previous programs, lack of a consultation network, lack of coordination among stakeholders, insufficient training and quality control centers, insufficient integration of scientific and technical research in design programs, lack of financial and technical resources to support scientific and technical research on renewable energy programming, inadequate material quality control, difficulties in ensuring small-scale local maintenance in rural areas, almost no after-sales service system, lack of consultation of beneficiary populations, virtual absence of a policy to raise awareness of the advantages of energy renewable, weakness of public subsidies for the promotion of renewable energy projects, and insufficient participation of private financial services. The limits of our research are also related to the complicated institutional and regulatory ecology, which includes too many competing programs from too many stakeholders, data scarcity, and fragmentation. In each energy political sector, there is a lack of technology transfer plans, and projects are nearly identical and are not subjected to thorough examination.

Due to the nature of the MCDM, some limitations include the use of only secondary data sources, potential bias in the data because the majority of sources were government reports and online articles, a lack of qualitative data to gauge implementation and impact to date, and limited sensitivity analysis undertaken for the MCDM.

The combination of MCDM and PCA makes it possible to measure the amount of information retained by the final measure (81.15% in our case) and to statistically infer errors related to the use of such a measure for other purposes afterwards. The loss of information (18.85% of the total variance) related to the use of the first principal component is, in fact, a brake on the inference of the rules related to the ranking of policies in the sense that we do not obtain an easily generalized rule for all policies. We rather obtain rules by classes of policies (detection of classes of policies), and we do not have enough information related to the comparison of these classes of policies between them.

6. Conclusions

In this research, we proposed a novel MCDA approach based on the PCA. Like many countries in Africa and around the world, Senegal is committed to energy transition. Due to the complex nature of the low-carbon energy transition sector, an objective evaluation of the various policies and scenarios necessitates a relevant approach allowing a judicious analysis. For our analysis, the 17 energy policies selected during the review of Senegal’s energy system among the four categories (electrification, renewable energy, energy efficiency, and climate change) were classified using the MCDA-PCA method.

The MCDA-PCA shows that the majority of initiatives with a score higher than 60 are electrification and renewable energy projects. This is a result of efforts made by the nation to increase generation capacity and encourage renewable energy sources. Our analysis has revealed that initiatives with a score of less than 50 are disproportionately concentrated on energy-saving techniques. One of the most significant areas for development is energy efficiency, which is essential to enabling Senegal’s energy transition and as specified in SDG 7.3. Senegal’s ability to make the transition to low-carbon energy depends on increasing energy efficiency. Among the 17 policies we examined, most energy efficiency projects obtained a result below 60, with the exception of the NEEAP, which has enabled Senegal to make a limited progress in this area between 2015 and 2019.

In terms of policy implications, the research of Senegal’s energy transition has revealed certain implementation stumbling blocks, primarily owing to governance and management concerns. Indeed, there is a misalignment between policy and reality in terms of achieving effectiveness, and the lack of data to track progress is an indication of a broken monitoring and assessment system. Furthermore, the task of improving institutional and regulatory frameworks, particularly in terms of coordination and transparency among the different major entities in charge of the industry, is a pressing concern. Favorable regulatory frameworks for the production of renewable energy would reduce opportunity costs, increase the effectiveness of the production of renewable energy, and foster a climate that is suitable to investment. Energy policy should have a defined aim, such as achieving energy efficiency targets (criterion C2) and increasing RE technology (criterion C3) based on the results of our analysis. The MCDA-PCA approach helps to find that Senegal has created a consistent and comprehensive energy policy over time, aided by the implementation of the Emerging Senegal Plan (PSE). By 2030, the country will have a strong chance of achieving universal access to affordable electricity, being energy self-sufficient (excluding biomass), and having a well-balanced energy mix. However, in light of the country’s demographics, economic growth, climate change challenges, the economic consequences of the COVID-19 pandemic, which has wreaked havoc on the Senegalese economy, as well as sub-regional and international circumstances, a policy focused on increased energy efficiency, coupled with bioenergy-based decarbonization of the transport and household sectors, could be a roadmap to effectively respond.

The challenge for the MCDA method is to obtain the best weighting that can allow an unbiased and objective analysis. In other words, the result of policy analysis is prominently affected by the attribute weights. This paper formalizes a strategy for integrating MCDA and PCA for the analysis of low carbon energy transition policy. The method enables the optimal weighted factor to be found. The most information-rich, standardized version of the optimal linear combination of the initial criteria serves as the best weighted factor.

The weighted factor in our context is the standardized version of the first principal component which can summarize up to 81.15% of the information provided by the six criteria. In fact, we use a standardized version of the principal component to keep the score of the policies between 0 and 1. The linear combination usually returns values which can sometimes be negative and not scaled on 100%. The standardization goal is to keep a principal component score value between 0 (as 0%) and 1 (as 100%).

Like any analysis method, the use of PCA can also have small limitations. For example, the calculation of principal components returns vectors with values in $ℝ$; this suggests the use of other transformations such as normalization to find a more appropriate interval (interval [0, 1] in our case). It is also impossible to have a component that can explain 100% of the total variance of a data set.

Future research should take into account that when the number of criteria is high, the share of information explained by the first principal component could be lower (less than 50% of the total variance). In this case, the use of a single principal component would be detrimental to the analysis. A higher dimensional visualization (using two to three components) will have to be performed, or a new PCA will have to be carried out on the principal components. As a result, the MCDA-PCA approach that is being given is a crucial step in the search for an appropriate, comprehensive analytical framework for the examination of energy-related policies from a sustainable and future-oriented perspective.