Application of TOPSIS for Multi-Criteria Decision Analysis (MCDA) in Power Systems: A Systematic Literature Review

Abstract

In this study, the authors present the results of a systematic literature review on applications of the technique for order of preference by similarity to ideal solution (TOPSIS) in power systems. The preferred reporting items for systematic reviews and meta-analyses (PRISMA) approach was used in the identification of publications used in this research. The SCOPUS database was utilized to locate the publication, and a total of 78 articles published between 2014 and 2024 were included in the review. A bibliometric analysis was performed, and reports were given on the annual number of publications and the top 10 cited journals. The main themes emerging from the content review of the publications were types of TOPSIS approaches, calculation of weights in multi-criteria decision-making (MCDM) problems, energy markets applications, renewable energy technologies assessment, heating and cooling systems combined with power systems, power system operation strategies, power system stability assessment, power system operations planning, and other power systems applications. Research trends and developments in the area were analyzed to identify the existing gaps. Proposed future research areas were identified based on trends and gaps presented.

1. Introduction

Decision-making in the field of power systems is associated with numerous factors for consideration, such as technological, economic, environmental, and social [1]. In the power systems domain, the nature of problems is often complex and multi-disciplinary. Also, the criteria for assessment of the problems encountered are often conflicting in the majority of cases. Of note, when addressing a specific problem, often there will be several potential solutions available; however, the selection of the most appropriate solution among multiple alternatives available remains a challenge, given that the criteria applied for assessment will be conflicting. Several evaluation methods applied in the selection of the best alternative often consider maximizing the benefit and minimizing the cost aspects, as the basis of evaluating options. Notably, this might present an element of subjectivity in the selection of the appropriate solution and may raise an element of doubt among decision-makers for not selecting the other alternatives. In mitigating the drawback, multi-criteria decision-making (MCDM) as a decision tool can be applied, since it will adequately reflect the preference of the decision-makers rather than the optimal solution. Also, it provides the necessary tools for decision-makers where there are multiple conflicting criteria for multiple alternatives in existence.

MCDM is a well-known branch of decision-making in the field of operations research [2]. It has gained popularity in its application within power systems lately, given the added benefit it presents in decision-making, particularly where several objectives are to be met. It is a mathematical model that considers numerous often-conflicting criteria, objectives, and stakeholders in evaluating several options [3], with the objective of selecting the feasible and preferred option amongst several potential options. It provides an allowance for a set of criteria to be defined, to evaluate alternatives through these criteria, and to enable decision-makers to make structured, comprehensive, and inclusive approaches when making decisions where complex problems exist [4]. It presents a more inclusive approach. The application areas of this method are huge [5]. Furthermore, this approach provides a better insight into the number of factors that are applied when a decision is reached, participants’ role in decision-making is promoted, facilitates collective decision-making, and provides a platform for applying methods that explicitly improve the quality of decision-making [6].

Key attributes that align with MCDM are the incorporation of multiple criteria and objectives simultaneously in solving complex problems in decision-making that will be optimal and satisfactorily, aligning with the decision-maker [7,8], handling of subjectivity and uncertainty [9] where complex and imprecise data are used through a fuzzy approach, and thus enhancing the flexibility and robustness of decision models, consideration of different stakeholders perspectives given divergent interest or priorities [10], and its allowance for a wide range of applications such as health, logistics, engineering and construction.

Several MCDM methods are in existence that are applied in evaluations, where multiple criteria or objectives that might be conflicting serve as a basis of assessment. Several scholars continue to apply the approach in solving a broad range of power system challenges. This includes applications geared toward addressing several problems in relation to the developments in the renewable energy domain [11], as well as the site selection and choice of technology in engineering [12] within the fraternity. Although there are several MCDM methods available, the application of these methods is diverse, making the selection of the best or feasible option difficult, since there could be different solutions achieved from different methods, notably when the alternatives applied are similar. The process is more similar for all the methods, as shown in Figure 1, regarding the approach to be followed when applied.

The approach of technique for order of preference by similarity to ideal solution (TOPSIS) is one amongst several MCDM methods available [13]. It was developed by Hwang and Yoon in 1981 [14,15]. It is one of the well-developed MCDM techniques and a widely used method, given its popularity. It is simple to use, and it can be applied where there are many criteria for alternatives to be evaluated. The alternatives are ranked in accordance with the shortest distance from the positive ideal solution (PIS) and the farthest distance from the negative ideal solution (NIS) in the evaluation of options, aiming to identify the best and preferred one.

The TOPSIS method has several steps described as follows:

Step 1. Defining the decision matrix as follows:

$$D=\left[\begin{array}{ccc}{r}_{11}& \dots & r_{in}\\ ⋮& \ddots & ⋮\\ r_{m1}& \dots & r_{mn}\end{array}\right]$$

(1)

where i = 1, …, m denotes the alternatives, and j = 1, …, n refers to the attribute r_ij represents the j-th attribute related to the i-th alternative.

Step 2. Normalizing the value of decision matrices as follows:

$$n_{ij}={\displaystyle \frac{{ r}_{ij}}{\sqrt{{\sum }_{j=1}^n{r^2}_{ij}}}}$$

(2)

where j = 1, …, n; and i = 1, … m.

Step 3. Calculating the weighted normalized decision matrix by multiplying the normalized decision matrix by its associated weights:

$$w_{ij}=w_{ij}\ast n_{ij}$$

(3)

where $w_{ij}$ represents the weight of the j-th attribute related to i-th alternative.

Step 4. Determining the PIS(A⁺) and NIS (A⁻).

$$A^{+}=\left\{ W_1^{+}, \dots ,W_n^{+}\right\}=\{\left(Max W_{ij}\right| j\in J), (Min W_{ij}| j\in J’\left)\right\}$$

(4)

$${ A}^{-}=\left\{ W_1^{-}, \dots ,W_n^{-}\right\}=\{\left(Min W_{ij}\right| j\in J), (Max W_{ij}| j\in J’\left)\right\}$$

(5)

where $J$ represents the positive factors, and $J\mathrm{’}$ is the negative factors.

Step 5. Calculating the distance of all alternatives to the PIS ($D_j^{+}$) and the ($D_j^{-}$) negative ideal solution (NIS).

$$D_j^{+}=\sqrt{{\sum }_{j=1}^n{\left(W_{ij}-W_j^{+}\right)}^2},i=1,\dots ,m$$

(6)

$${ D}_j^{-}=\sqrt{{\sum }_{j=1}^n{\left(W_{ij}-W_j^{+}\right)}^2},i=1,\dots ,m$$

(7)

Step 6. Calculating the relative closeness of each alternative as follows:

$${\mathrm{C}}_{\mathrm{i}}^{\mathrm{*}}={\displaystyle \frac{{\mathrm{D}}_{\mathrm{i}}^{-}}{{\mathrm{D}}_{\mathrm{i}}^{+}+{\mathrm{D}}_{\mathrm{i}}^{-}}}$$

(8)

where $C_i^{\ast }$ lies between 0 and 1; the higher the value, the better the performance.

Step 7. Ranking the preference order.

Although MCDM has been applied broadly within the engineering fraternity, as shown in the review articles sourced through the Google Scholar database, the focus on power systems has been presented as part of engineering applications, including related fields. Various renewable energy applications as part of energy planning are presented in [11], using the technique. In [16], the application of TOPSIS was applied to evaluate a wide range of applications, ranging from logistics, chemical engineering, health and safety, human resources, energy management, and others. Pandey et al. [17] presented the hybrid approach of the method that aims to solve diverse problems within several disciplines, ranging from science, mathematics, engineering, and industrial applications, through ranking approaches.

In [18], the authors present the experimental results that demonstrate the drawbacks of the TOPSIS technique as applied in several areas. Salih et al. [19] presented the challenges that can be encountered in applying techniques in various areas, including the benefits that could be derived through the application of the fuzzy approach, given several problems to be solved. Bohra et al. [20] presented a review of the method for evaluating renewable energy source (RES) site selection. Furthermore, the energy policy and planning need to be considered due to the presence of various renewable technologies, such as wind and solar PV. Jato-Espino et al. [21] presented the application of this technique in solving problems emanating within the construction industry and compared it with other methods available. Khan et al. [22] applied the approach in analyzing the developments, future frameworks, trends, and gaps that exist in the sustainable supply chain management. Yu et al. [23] presented the approach of selecting the ideal renewable energy source (RES) using different MCDM methods, including TOPSIS. In [24], the authors applied the method in the selection of the cloud service providers among several available providers. The authors in [25] presented a review of selecting different renewable technologies amongst each other in identifying the preferred one.

In accordance with previous reviews with specific focus on MCDM, extended to the TOPSIS approach [11,12,13,14,15,16,17,18,19,20,21,22,23,24,25], where a lot of work has been extensively outlined, it is evident enough to conduct a review on MCDM that incorporates TOPSIS, aiming to enhance the specific focus in electrical engineering with emphasis on power systems given continuous changes in the domain. In contrast, the purpose of the article is as follows:

• To conduct a comprehensive review of the literature on MCDM, including TOPSIS, its characteristics, and current applications through an analysis of recent and most-cited scientific papers, with an emphasis on power systems, given its diversity and recent developments.
• To identify trends and developments, including future directions of areas of application within power systems based on statistical data.
• To provide a framework and valuable resource for the researchers’ understanding of TOPSIS application in power systems, developments, and future areas of research.
• To focus the attention of researchers on what has been done and what the prospects of future development in this area are, as an encouragement.

The rest of the paper is organized as follows: Section 2 outlines the research approach undertaken in this review. Section 3 presents and discusses the results obtained in detail. Section 4 presents the analytic trends and outlines the prospects that can be studied through this approach. The conclusion and proposed direction for future research are covered in Section 5.

2. Research Approach

The study undertaken to conduct the systematic literature survey (SLR) in this research aims at presenting the MCDM applications within power systems, with a specific focus on the TOPSIS approach. The SLR methodology is characterized by three distinguishable features, being the delineation of the research strategy, the identification of inclusion and exclusion criteria of the publications, and, finally, the aim to seek, gather, and synthesize the relevant available publications to be considered for research. The process, as summarized schematically in Figure 2, fully describes the systematic literature review process adopted in reporting the approach, termed the PRISMA approach in accordance with Moher et al. [26]. This method allows the screening of publications, refinement of inclusion and exclusion based on quality, and selection of relevant criteria as per criteria, and finally, the summarized publication of choice will be presented.

2.1. Identification

The Scopus database was used as a first step in the identification of the research articles to be used in this research. The “article, title, and abstract” in the “search within” was chosen, the “search document” option was selected, and the string “multi-criteria decision” was inserted in the search field. A total of 1586 documents were returned by the search. Following the initial search, the fields that were not initially considered were added. The choices that followed are as follows: “subject area”, “engineering”, “document type”, “article”, and in the keywords, “multi-criteria decision”, “TOPSIS”, and “power systems” were chosen; for the “source type”, “journal” was chosen. The outcome of the process returned 286 articles.

2.2. Screening

For this stage, only articles in English were considered in this regard. Additionally, the consideration was on articles that are published in the window period 2014–2024. This makes the provision of the latest publication part of the survey. Also, the duplicated articles were filtered out. Articles based on relevance, based on abstract analysis, were identified for review. The exclusion was on those deemed not to be relevant. Based on this approach, a total of 137 articles remained within consideration for review.

2.3. Eligibility

The download of the full text for all the articles screened was conducted, aiming to review the content and analyze it in establishing the research context. Based on the research context of the manuscript, the eligibility for review was identified among the articles downloaded. Although there were other articles that were never available upon download, those happened to be excluded from the refined list. Some articles that were “the reviews” were also excluded from the eligibility. Upon conclusion of the exercise, 78 articles were eligible for review and analysis, as part of the research. Although only English articles are considered, future consideration in expanding the database of articles is recommended for inclusion as part of the future research

2.4. Extraction

The 78 eligible articles identified as part of the process described above were analyzed as part of the final stage of the extraction process. Notably, the major themes, including sub-themes, were identified as highlighted in Figure 3 below. The details are fully described within Section 3.

Search string “multi-criteria decision analysis” and “TOPSIS” and “power systems”. The search was limited to English articles only.

3. Results and Discussion

3.1. Bibliographical Analysis

3.1.1. Number of Publications per Year

The annual number of publications per year for the 78 selected articles is plotted in Figure 4. These numbers, covering the period from 2014 to 2024, present a steady growth in the number of publications. However, exponential increase is noticeable in recent years starting from 2020, as shown by the number of publications reaching 8 as opposed to an average of 3 in the past years. A big jump in the last few years is evidently visible in the graphical representation. This shows recent interest in the approach lately. The trendline shows the initial number of publications of 2, with an average growth of 4.6 annually, as depicted by the equation $Y=4.6877x+2$.

3.1.2. The Top 10 Most-Cited Journals

From the list of publications reviewed, an analysis conducted presents the list of the top 10 most cited journals. The list is shown in

Table 1. List of the top 10 most-cited journals considered in this study.

Journal	Number of Citations
Energies	11
IEEE Access	9
Electric Power Systems Research	6
Energy	6
Journal of Energy Storage	5
Applied Energy	4
International Journal of Electrical Power and Energy Systems	4
Electric Power Components and Systems	3
Global Energy Interconnection	3
International Journal of Electrical and Computer Engineering	3
International Transactions on Electrical Energy Systems	3

, where the highest is Energies, with 10 publications sourced, followed by IEEE Access with 9 publications, and Electric Power Systems Research and Energy Journals, each contributing 6 publications to the list.

The overall outcome of the results obtained from the review conducted based on publications allowed an approach of grouping them into five distinguishable categories.

3.2. Types of TOPSIS Approaches

An overall assessment of the qualifying publications suggested that they could be grouped into standard TOPSIS, fuzzy TOPSIS, where fuzzy logic is applied in handling the vagueness and uncertainty in decision-making, and variants or modifications of TOPSIS that can be flexibly applied in decision-making based on several objectives, often classified as hybrid applications. These are grouped into entropy weight–TOPSIS combination, AHP–TOPSIS combination, weighted sum model–TOPSIS combination, combined weight–TOPSIS combination, hybrid particle swarm–TOPSIS approach, affinity propagation–TOPSIS combination, average weight ordered distance–TOPSIS combination within the review.

The approach was also used in determining the weights that could be applied in solving multi-criteria problems through the other MCDM approaches. In solving some of the latest technological developments, such as renewable energies, the approach was considered as part of the review, including heating and cooling systems, grouped into these specific themes.

3.2.1. TOPSIS Evaluation Approach

The technique of TOPSIS as an independent MCDM approach is also presented in solving several power system problems. The applications are summarized and discussed in the section below.

Table 2. List of TOPSIS applications in solving various problems, including the criteria applied.

Problem to be Solved	Ref.	Author	Criteria
Managing electricity prices and power system reliability	[27]	Abdollahi et al.	peak reduction, energy consumption, load factor, demand response incentive value
Improvement of the power line inspection approach	[28]	Cui et al.	communication faults, sensor failure, power system anomaly
Mitigation of extreme weather events	[29]	Wang et al.	physical perspective, human perspective, cyber perspective
Improvement of power system quality	[30]	Skrikakolapu et al.	cost function of the current control, minimization of the switching frequency
Solving the power system congestion problem	[31]	Salehizadeh et al.	congestion, intermittent generation, emissions reduction, security
Evaluating power system security assessment	[32]	Yu et al.	intuitionistic fuzzy numbers
Increasing the observability of the power system	[33]	Singh et al.	rotor angle and frequency of voltage busbars, reactive power deficiency, maximum loading of transmission lines
Identifying and resolving voltage problems	[34]	Spertino et al.	voltage quality, grid losses, online tap changers’ lifespan increase
Balancing the supply and demand loading	[35]	Alikhani et al.	type of load, customer bill, peak reduction, energy reduction, load factor, peak valley
Stabilization of the power system	[36]	Hashemi et al.	voltage, output power of generators, var limits of var sources, fault clearing time, loading of the network
Maximization of the generation supply capacity	[37]	Shen et al.	electricity constraints, electricity contracts, plant constraints
Improving the power system hosting capacity	[38]	Diaaeldin et al.	power loss, active power delivered, annual energy loss
Control of the power system in emergencies	[39]	Xie et al.	active output load, power shortage, active power shortage level, inertia
Managing the system security	[40]	Jiaang et al.	improved node index, node tightness index, improved node intermediate index, improved voltage crossing index
Assigning adequate power levels across the network	[41]	Elattar et al.	fault current, FCL size, TVSI, TVD
Maximizing the transmission line utilization	[42]	Riazaeian-Marjani et al.	active power losses intended, power flow control, economic aspect of power flow
Combating the effects of harmonics due to non-linear loads	[43]	Ebrahimi et al.	network reconfigurations, distribution generation dispatch, demand side management, reactive power compensation
Alleviating network congestion	[44]	Salehizadeh et al.	costs, number of curtailed customers, importance of curtailed load
Improving the reliability of the power system	[45]	Leckbich et al.	costs, system average duration index, system frequency interruption index
Smoothing the fluctuations of renewable energies connected to the grid	[46]	Yao et al.	air to methanol ratio, pressure ratio, isentropic efficiency of gas
Improving the power system operations and planning	[47]	Jani et al.	operating costs, voltage deviation, air emission applied
Enhancement of the frequency stability of the power system	[48]	Lui et al.	costs, revenue, performance
Addressing a power-flow problem in a regulated market	[49]	Ramesh et al.	economic, technical, operational, and security aspects

summarizes the full details of problems to be solved, as sourced from the publications considered.

In managing high electricity prices, including power system reliability, Abdollahi et al. [27] used the approach in prioritizing and selecting the appropriate demand management programs aimed at addressing this challenge. They applied peak reduction, energy consumption, load factor, and demand response incentive value as the criteria. In improving the power line inspection approach, the multirotor inspection strategy was evaluated through this approach by the authors [28]. They applied the criteria of communication fault, sensor failure, and power system anomaly in the assessment. In mitigating the extreme weather events on the power system, Wang et al. [29] applied the approach in proposing a resilient framework plan from four municipalities in China regarding the response to the threats and compared them against each other. The approach considered physical, cyber, and human perspectives in the evaluation as the criteria.

As a measure to improve the power quality of the power system, Srikakolapu et al. [30] applied the approach to identify the proper switching state of the distribution static compensator (DSTATCOM). The criteria applied were the cost function of the current control and the minimization of the switching frequency. In solving the power system congestion problem, the authors in [31] applied the approach to establish the optimization strategy of the power system, considering congestion, intermittent generation, emissions reduction, and security as the criteria within the study.

In evaluating the power system security assessment, Yu et al. [32] applied the approach in evaluating the system through an optimized approach called weighted hamming distance. The criteria were based on intuitionistic fuzzy numbers generated. Singh et al. [33] aimed at increasing the observability of the power system through the optimal placement of phasor measurements units (PMUs) on the critical buses that will be identified through the TOPSIS approach. Redundancy of measurements, rotor angle and frequency monitoring of generator buses, reactive power deficiency, and maximum loading limit under transmission line outage were applied as indices. In identifying and resolving the voltage problems on low voltage (LV) networks, several optimal setups on voltage control devices commissioned were studied and identified by the authors in [34]. Voltage quality, grid losses, and online tap changer (OLTC) lifespan increase were applied as the criteria for evaluation.

To balance the load between supply and demand, the authors in [35] applied this approach to evaluate and select the demand management program that can be applied. They applied criteria of type of load, customer bill, peak reduction, energy reduction, load factor, and peak to valley within the assessment. Aiming to stabilize the power system, the optimization of the voltage and first swing together was studied by Hashemi et al. [36] through this approach. They applied voltage, output power of generators, var limits of var sources, fault clearing time, and loading of the network as indices. Aiming at maximizing each supply area generation connected to the grid in China, Shen et al. [37] applied this approach to optimize the hydro plant that supplies the electricity system. They applied the criteria for security constraints, electricity contracts, and plant constraints as part of the approach.

With the objective of improving the power system hosting capacity, the authors in [38] applied the method in evaluating the deployment of soft open points within the network in conjunction with distribution reconfiguration. They applied power loss, active power delivered, and annual energy loss as the criteria. With the objective of being able to control the power system under emergencies, Xie et al. [39] applied the study to identify the optimal candidate under-frequency scheme for the partition networks. They applied active power output, load, power shortage, active power shortage level, and inertia constant as the criteria. The authors in [40] proposed an approach for identifying vulnerable nodes within a distribution system. This ensures that the system is managed optimally and securely. They applied the criteria of improved node index, node tightness index, improved node intermediate index, and improved voltage crossing index.

With the objective of minimizing fuel costs, the environmental emissions of the generation units, and the total losses over the HVAC, multi-terminal direct current (MDC), and voltage source converter (VSC) stations leading to assigning adequate power levels across the power system, Elattar et al. [41] applied the approach in selection of optimal candidate point amongst a number of options available within the power system that consists of VSCs and high voltage direct current (HVDC) technologies. They applied the criteria of losses, fault current, fault current limiter (FCL) size, TVSI, and TVD. In maximizing the benefit of transmission line utilization where there is a high level of penetration of wind, the authors in [42] applied the method in evaluating the placement of the FACTS device, interflow power flow control (IPFC). Active power losses, intended power flow control, and the economic aspects of power flow were applied as the criteria. With the aim of combating the effect of harmonics due to the presence of non-linear loads within the power network, Ebrahimi et al. [43] used the approach to optimize the network to minimize the total harmonic distortion of the system. Network reconfiguration, distributed generation (DG) dispatch, demand-side management, and reactive power compensation were applied as indices.

To alleviate the network congestion, the authors in [44] applied the method to calculate the degree of severity. They applied costs, the number of curtailed customers, and the importance of curtailed load in the assessment as the criteria. Leckbich et al. [45] applied the method to identify the optimal location of reclosers and switches within the power system, aimed at improving the reliability of the power system. They applied the system average duration index (SAIDI), the system average interruption frequency index (SAIFI), and costs as the criteria. With the aim of smoothing the fluctuations of renewable energy connected to the power system, Yao et al. [46] applied the approach to evaluate the performance of compressed air technology in combination with thermochemical energy on the power systems’ performance. Air-to-methanol ratio, pressure ratio, and isentropic efficiency of gas were applied as the criteria.

With the objective of improving power system operations and planning, Jani et al. [47] applied the method to validate the results obtained in selecting the optimal site for placement of the energy storage system (ESS). Operation cost, voltage deviation, and air emission were applied as the criteria. In trying to enhance the frequency stability of the power system, Liu et al. [48] applied the approach to evaluate the strategy of optimizing the power system considering different energy storage systems. They applied costs, revenue, and performance as indices for evaluation. In addressing the power flow problem in a regulated network, the approach was applied to evaluate the varied load levels and compare them against each other [49]. The criteria of the valuation were on economic, technical, operational, and security aspects.

3.2.2. Fuzzy TOPSIS

A fuzzy system allows the use of linguistic variables with which decision-makers are more at ease than with using precise numerical values. Sometimes, it is difficult to represent the real values of the rating of alternatives and the weights of the criteria; as a result, the fuzzy value is adopted [50]. It is more suitable for addressing complex problems, and it is well defined in that it can deal with incomplete, uncertain knowledge and information. Some of the problems highlighted below were solved through the approach. In mitigating the challenges that could be experienced due to large-area voltage collapse, Wu et al. [51] applied the approach of fuzzy TOPSIS to rank the voltage buses within the Chinese utility, considering the criteria of proportionality and system functionality. In selecting the preferred sites for placing the electric vehicle (EV) chargers, Guo et al. [52] applied the methodology to identify and select the charging sites and rank them among each other through economic, social, and environmental indices. The proper site selection for the placement of EV chargers is considered to play a pivotal role in the EV ecosystem. To evaluate and select the best strategy for managing three hybrid wind farms located closer to each other, Dhiman et al. [53] applied the approach to manage the challenge brought by the variation of the wind speed. They applied criteria of wind wake, wind curtailment, and forced outage for the selection.

3.2.3. Variants of TOPSIS Assessment Applications

Variants or modifications of TOPSIS can be flexibly applied in decision-making based on several objectives. These are normally intended to improve the TOPSIS approach and enhance the quality of decision-making. Several methods discussed below were applied to solve a few power system problems as described.

Improved TOPSIS Assessment Applications

The traditional TOPSIS approach uses the Euclidean distance between the different schemes and the positive and the negative solutions. However, these often lead to poor accuracy and stability given the approach. To mitigate the challenge, the Mahalanobis distance and the Tanimoto correlation coefficient are applied to replace the Euclidean distance in improving the TOPSIS approach. Here, the method is applied to address the reliability challenge the power system experiences due to peak loading. Chen et al. [54] applied the approach to establish a model that is used to evaluate the adjustable potential of large industrial loads connected to the system regarding their participation in the peak shaving exercise. The criteria applied were the characteristics of operation and production, users’ electricity characteristics, and willingness to be interrupted. To evaluate the reasonable quantity of power supplied by photovoltaic (PV) systems that can be absorbed by the grid without any challenge, the approach was applied by the authors in [55] to assess the reasonable power that can be supplied. They applied the criteria of daily energy generation, maximum ratio of power, and maximum ratio of energy generation for an assessment. The improvement of TOPSIS considered the exclusion of the negative ideal solution step. In meeting the national energy strategy of China, 25 different regions’ electricity potential for electricity substitution was compared against each other to select the best performing in accordance with the criteria of energy consumption, economic development, and environmental production [56]. The TOPSIS improvement was done by replacing the negative ideal solution and positive ideal solution with the concept of the connection degree.

Entropy Weight–TOPSIS Combination Assessment

Entropy is used to measure the uncertainty of the alternatives when the weight of the criteria is determined. It measures the degree of differentiation [57]. The higher the degree of dispersion, the higher the degree of differentiation, and the more value can be derived. The approach uses objective information from the criteria and calculates weights in an objective manner. The calculation process of the entropy method is more convenient, and the weight of each evaluation index obtained is easy for decision-makers to understand, although the approach is data-dependent, implying that if the amount of data is small, the results will be inaccurate. The nature of the following problems was evaluated with the combination of TOPSIS and the entropy approach. To ensure that all the provinces in China achieve the objectives of the carbon peaking and carbon neutrality goals, Liu et al. [58] applied the entropy approach to objectively assign weights to the indices and the TOPSIS method to compare the provinces regarding their power system compliance in this regard. They applied the criteria on generation, network, load, and storage, and compared them with each other. To manage the uncertainty of wind farm performance, the best bidding strategy was evaluated through the approach by the authors in [59] in Denmark. They applied charging and discharging, including scheduling of the wind farm day-ahead in the electricity market, as the criteria.

In managing the uncertainty of wind, Falsafi et al. [60] applied the method to evaluate several available demand response programs and selected the best program for consideration. The criteria applied in this approach were cost, emission, and multi-objective functions. In moving away from fossil to clean energy in the Philippines, the authors in [61] applied the approach by evaluating the optimal capacity of the load supplied by the hybrid renewable systems. They applied the criteria of lowest energy cost, capital cost, fuel savings, and occupied area within the approach.

Weighted Sum Model–TOPSIS Combination

The weighted sum model combines both qualitative and quantitative criteria in the evaluation of alternatives. The alternatives will be considered because of the advantages and disadvantages when a decision is made [62]. Various factors will be weighed when deciding. The method has an added advantage of being simple, flexible, and transparent. Although it has disadvantages such as reliance on subjective judgment, lack of precision, and inability to capture non-linear relationships, the approach, used in combination with TOPSIS, solves the power system-related problems as presented. In ensuring that the placement of flexible alternating current transmission systems (FACTS) devices addresses the system requirements optimally within a power system, Chinda et al. [63] evaluated the optimum placement of the two “dyna-flow controllers” on the system. They applied the method with the criteria of power loss, fuel cost, fuzzy index, and line flow index. In managing the challenge of electrification flexibility, hybrid electrification was considered in Sierra Leone, where the authors evaluated different hybrid electrification modules consisting of PV as part of the solution [64]. They applied the economic, technical, and environmental criteria.

Combined Weight–TOPSIS Combination

The optimized combination of the subjective and objective weights is applied as part of the evaluation within the TOPSIS method to avoid the one-sidedness of a single method and thus make the weights more reasonable [65]. This section presents an approach to solving power system problems and ensuring the smooth and seamless operation of the power system. To this end, Liu et al. [66] evaluated different auxiliary service models within European countries using this approach. The development of auxiliary service transactions, reasonable compliance with trading mechanisms, benefits of the transaction mechanism were applied as the criteria of assessment. This further ensures seamless and accurate auxiliary service operation of the power market.

Affinity Propagation–TOPSIS Combination Assessment

Affinity propagation is an efficient data clustering method. In data clustering, hidden patterns in a dataset are extracted [67]. It is commonly used in computer science for tasks such as indoor localization and improving Wi-Fi-based positioning accuracy. It presents an advantage where the user does not need to specify the number of clusters but does need to specify a ‘sample preference. Although the disadvantage is that it is slow and memory-heavy, thus making it difficult to scale to large datasets. However, in this section, the approach is presented in combination with TOPSIS, aiming to find an alternative optimal black-start scheme that will greatly enhance the restoration of the system in the event of failure. Leng et al. [68] evaluated different candidate black-start schemes to select the best one that can be considered in the event of a power failure. The criteria applied in the selection were rating capacity, unit state, startup power, ramping ratio, and number of switch operations.

3.2.4. Hybridization of the TOPSIS Method in the Assessment

The TOPSIS method was applied in conjunction with other MCDM methods in solving several power system-related problems. The methods as applied are discussed in the subsections below.

Analytical Network Process–TOPSIS Combination Assessment

The analytic network process (ANP) is a generic form of AHP that allows for more complex and interdependent relationships in the hierarchy [69]. It enables feedback and interactions between and within clusters, making it a more comprehensive decision-making tool. It has the advantage of the ability to handle complex decision-making problems with multiple criteria, subjective inputs, and interdependent relationships among criteria. It is more suitable for risk and uncertainty applications. Here, it was applied in an attempt to increase wind and solar power by 2030 and achieve carbon neutrality in China by 2060. Chen et al. [70] evaluated and selected different types of flexible power sources that could be commissioned in China through this approach when combined with the benefit, opportunity, cost, and risk (BOCR), including TOPSIS. The criteria applied were benefits, opportunity costs, and risks within the approach.

Analytic Hierarchy Process–TOPSIS Combination

The analytic hierarchy process (AHP) is normally used for ranking the alternatives based on expert opinions where multiple criteria exist. It is more suitable for solving complex problems using the hierarchical level (from higher to lower) in an MCDM, to compare the criteria pairwise to make a final decision. Its advantage is its ability to manage different classes of quantitative and qualitative criteria. Conversely, the disadvantage is that the objects to be evaluated should not be too many, otherwise the hierarchical structure will be too complicated. Here, the method is combined with TOPSIS in evaluating several complex problems of that nature within the power systems (

Table 3. List of the extended fuzzy TOPSIS, variants of TOPSIS, and hybrid TOPSIS methods in solving various power-system problems.

Area		Problem to Be Solved	Ref.	Criteria
Fuzzy TOPSIS		Mitigating large-area voltage collapse	[51]	proportionality, system functionality
		Selection of EV charger placement	[52]	economic, social, environmental
		Selection of the best strategy for managing the wind farm	[53]	wind wake, wind curtailment, forced outage
Variants of TOPSIS	Improved TOPSIS	Addressing the reliability challenge of the power system	[54]	characteristics of operation and production, users’ electricity characteristics, willingness to be interrupted
		Assessment of the reasonable quantity of power supplied by a photovoltaic (PV) system	[55]	daily energy generation, maximum ratio of power, maximum ratio of generation
		Meeting the energy strategy of China	[56]	energy consumption, economic development, environmental production
	Entropy Weight and TOPSIS Combination	Achieving carbon peaking and neutrality goals	[57]	generation, network load, storage
		Managing the uncertainty of the wind farm performance	[59]	charging and discharging, scheduling of the wind farm day-ahead in the electricity market
		Managing the uncertainty of wind	[60]	costs, emissions, multi-objective function
		Moving away from fossils to clean energy	[61]	lowest energy cost, capital cost, fuel savings, occupied area
	Weighted Sum Model and TOPSIS Combination	Placement of FACTS devices	[63]	power loss, fuel cost, fuzzy index, line flow
		Managing the challenge of electrification flexibility	[64]	economic, environmental, technical
	Combined Weight and TOPSIS Combination	Smooth and seamless operation of the power system	[66]	auxiliary service transaction, reasonable compliance with the trading mechanism, benefits of the transaction mechanism
	Affinity Propagation and TOPSIS Combination Assessment	Selection of an optimal alternative Black-Start scheme	[68]	rating capacity, unit state, startup power, ramping ratio, number of switch operations
Hybridization of TOPSIS	ANP-TOPSIS Combination	Increasing wind and solar power by 2030 and achieving carbon neutrality in China	[70]	benefits, costs, opportunity, risk
	AHP-TOPSIS Combination	Managing adequate power distribution in power systems	[71]	cost, voltage drop, power losses
		Accommodate the penetration of renewable energy	[72]	market structure, market performance, market conduct
		Ensuring that the excess wind at the ports is utilized for power generation	[73]	wind resource, specific climate, noise
		Establishing the optimal financial structure within the electricity market	[74]	volume traded, risk, mean price, revenue assessment

). In managing adequate power distribution in power systems, Zedack et al. [71] evaluated the placement of photovoltaic and battery energy storage in optimizing the energy purchased. They applied the approach where the criteria were cost, voltage drop, and power losses. To accommodate the penetration of renewable energy, several shared energy systems (SES) markets were evaluated against each other through this approach [72]. The criteria applied were market structure, market performance, and market conduct. In ensuring that the excess wind at the ports is utilized for power generation, the authors in [73] applied the method to evaluate the location of wind farms within ports, including several available generator types to be considered. They applied wind resources, specific climate, and noise as indices. In establishing the optimal financial structure within the electricity market, Oprea et al. [74] evaluated the models in selecting the best model among several constructed ones. They applied criteria of electricity volume traded, risk, mean price, and revenue in the assessment.

3.3. Calculation of Weights in MCDM Problems

As an added benefit, TOPSIS can be applied to establish the weights of the criteria that could be applied in any MCDM approach. This section highlights some aspects in that regard. In accordance with Xing et al. [75], TOPSIS can be used for assigning weights to various alternatives or criteria as opposed to ranking alternatives. Furthermore, there are no strict requirements in criterion type, criterion amount, sampling size, or data distribution. The authors further identified four common types of criteria that are normally applied through TOPSIS, viz., benefit, costs, intermediate criterion (closer to value), and interval criterion (closer to a specific interval). Aiming to optimize the flexibility of the power system with high renewables penetration and improve the stability of the power system when the pumped storage is ramped up, Zhang et al. [76] calculated the flexibility and stability scores through the TOPSIS approach within the normalization of the matrices. In ensuring the effective operation of the power system, the remaining useful life (RUL) of the lithium-ion batteries connected to the system was evaluated by Long et al. [77]. As part of the evaluation, TOPSIS was applied for the selection of the criteria that formed part of the evaluation of the remaining RUL of the lithium-ion batteries. The selected parameters were data volume, correlation coefficient, fluctuation degree, and subjectivity.

3.4. Energy Markets Applications

The reform of the energy markets is driven by the economic incentive of competitive electricity pricing and encouraging competition within the industry. To this end, this is a developing area, although mature in some regions, as evident from the research conducted.

In minimizing the impact of congestion on the power system, particularly on the low voltage (LV) network with high penetration of renewables, Vaughan et al. [78] evaluated different tariff designs aimed at minimizing the impact of congestion on the power system. The criteria of system sustained operations, economic efficiency, equity, and consumer consideration were applied. With the objective of dealing with complex constraints within a power system, Guo et al. [79] evaluated various dynamic economic emissions dispatch (DEED), clustered into different combinations, with the aim of selecting the best combination. They applied criteria of cost, emission, and time in the evaluation.

In accelerating the introduction of renewable plants into the electricity market, Soltaniyan et al. [80] applied the approach in the selection of the method that allows better planning and introduction of the plant into the power system. They applied the criteria of total load, average learner index, social welfare, congestion costs, sum of units declared costs, Nash index, and consumer payments within the evaluation. To adjust and manage the PV penetration on the power system, the approach was applied to construct an optimal electricity pricing model for the PV users connected to the grid, including ranking the PV users’ customers’ reaction [81]. The criteria applied were shared photovoltaic energy, contributors to operators’ revenue, and installed capacity.

The electricity market is seen as a pivotal mechanism that can be applied in balancing the power system, taking into account the source–network–load scenario. It is also a mechanism that can be applied as a benefit in both the generator and the consumer through an optimal pricing mechanism put in place. The analysis and evaluation attest to the fact that the energy market is a potent and effective mechanism that has the potential to manage the power system’s economic and environmental aspects.

3.5. Renewable Energy Technology Assessment

Renewable energy, referred to as clean energy, is produced from natural resources such as wind, sunlight, and geothermal heat [82]. The ongoing discussion on climate change has led to several countries setting targets aiming to achieve net-zero emissions. To propel the idea, the introduction of renewable energy sources in energy generation remains the most favorable and realistic approach. The application of this approach has been extensively considered in several studies related to these technologies, as outlined in this section.

In addressing the challenge of remote provisioning of power, heating, and pumping water, the optimal performance of the tri-generator was evaluated based on the approach [83]. The criteria of electric power generation, freshwater production, and heating power (system outputs) were applied in the evaluation. Li et al. [84] aimed to manage costs and ensure that there are enough renewable plants connected to the system; they applied the method to optimize the load distribution whilst minimizing the emissions in a network that has wind plants connected. They applied criteria of minimum power, maximum power, and megawatts per hour in the evaluation. In trying to improve the power system sustainability, Peirow et al. [85] applied the approach to evaluate different hybrid renewable energy combinations that supply the hospital load. The criteria applied were technical, environmental, economic, and energy security factors.

As part of the response to climate change and improving the level of renewable penetration, Lim et al. [86] applied the method to prioritize the deployment of renewable energy sources across the 15 regions in South Korea. The criteria applied were solar radiation, wind speed, energy demand, and natural gas demand. In an attempt to manage the power system constraints associated with the economic and environmental distribution of large-scale renewables, He et al. [87] applied the approach to evaluate the practical distribution of photovoltaic systems within the grid. The criteria applied were investment and earning ability, production and operation ability, power-grid coordination ability, energy-conservation and emission-reduction ability, sustainable development ability, and society-serving ability. In an attempt to develop low-carbon electricity, Jin et al. [88] established an optimum solution that compares the wind power fluctuation and the low-carbon trading price to influence the electric dispatch mode. They applied criteria of fuel cost and emission cost as part of the evaluation. In reducing the effect of greenhouse gas emissions through the introduction of renewable energy technologies, Kumar et al. [89] applied the approach to select the best bidding approach to optimize the buyer and suppliers’ revenue in the renewable market in India. The cost and benefit criteria were applied within the approach.

Although renewable plants are intermittent and non-dispatchable, they pose operational challenges such as stability for the power system. Their presence provides an indispensable benefit of providing clean and cheaper energy, thus offering inherent prospects for combating the effects of greenhouse gas (GHG) emissions.

3.6. Heating and Cooling Systems Combined with Power Systems

Heating, ventilation, and air conditioning (HVAC) are parts of the energy consumption in a building. Solar energy, radiative cooling, and radiative heating are among the different physical properties that are used for heating and cooling buildings [90]. To reduce fossil fuel consumption and greenhouse gas emissions, the authors applied the approach to optimize the combined cooling, heating, and power (CCHP) system that contains the thermal and solar storage [91]. They applied economic, environmental, and energy criteria in the evaluation. To improve energy efficiency and reduce emissions, He et al. [92] applied the approach to rank alternatives of different CCHP systems. They applied the criteria of resource, environment, economic, society, and technology as part of the evaluation.

In trying to optimize the heat recovery steam generator (HRSG), Ghafharri et al. [93] applied the approach to optimize the fire tube heat recovery system. They applied the criteria of total annual costs, steam mass flow rate, and energy efficiency as part of the evaluation. In ensuring the sustainability of the manufactured electronic device used for cooling systems, Loganathan et al. [94] evaluated different materials used in manufacturing. They applied the criteria of melting point, latent heat, density, specific heat (solid), specific heat (liquid), and thermal conductivity as part of the assessment. The authors in [95] evaluated various optimization techniques applied to hybrid thermal systems (HTSs) for domestic heating consisting of solar, electric heat pumps, and fuel-based (LPG or diesel) in order to select the best optimization technique aimed at lowering carbon, reducing emissions, and saving costs for the customers.

3.7. Power System Operation Strategies

In achieving a resilient power system where numerous considerations, such as costs, operational efficiencies, and system security, must be addressed, several strategies must be implemented. In this section, the achievement of power system strategy through the TOPSIS assessment is presented. Adetunji et al. [96] conducted a study evaluating the strategic approach of introducing distributed energy into the power system, considering the reduction of emissions and costs. The criteria of power loss index, voltage deviation, emission costs, voltage stability index, operating costs, and installation costs were applied in the assessment.

To evaluate and select the best strategy for managing three hybrid wind farms located closer to each other, Dhiman et al. [53] applied the approach in managing the challenge brought by the variation of the wind speed. They applied criteria of wind wake, wind curtailment, and forced outage for the selection. In addressing the reliability challenge the power system experiences due to peak loading, Chen et al. [54] applied an improved TOPSIS method in establishing a model that is used to evaluate the adjustable potential of large industrial loads connected to the system regarding their participation in the peak shaving exercise. The criteria applied were characteristics of operation and production, users’ electricity characteristics, and willingness to be interrupted.

In ensuring that all the provinces in China achieve the objectives of the carbon peaking and carbon neutrality goals, Liu et al. [58] applied the entropy approach to objectively assign weights to the indices and the TOPSIS method to compare the provinces regarding their power system compliance in this regard. They applied the criteria on generation, network, load, and storage, and compared them with each other. In reducing the effect of greenhouse gas emissions through the introduction of renewable energy technologies, Kumar et al. [89] applied the approach to select the best bidding approach in optimizing the buyers’ and suppliers’ revenue in the renewable market in India. The cost and benefit criteria were applied within the approach.

3.8. Power System Stability Assessment

Power system stability refers to the phenomenon of the power systems, given an initial operating condition, having the ability to regain an equilibrium position when subjected to a system fault, with most of the system variables bounded so that the entire system remains intact [97]. Here, TOPSIS was applied in evaluating various aspects of the power system to ensure that the stability of the system is achieved. Aiming to reduce losses, optimize the techno-economic benefit, and improve the voltage profile on the network, Kurundkar et al. [98] evaluated the optimal placement of reactive power devices in the system to achieve the intended objective through this approach. The criteria applied were power loss reduction, maximization of the stability margin of voltage, and minimization of the deviation of voltage within the approach. As a measure to improve the power quality of the power system, Srikakolapu et al. [30] applied the approach in the identification of the proper switching state of the distribution static compensator (DSTATCOM). The criteria applied were the cost function of the current control and minimization of the switching frequency.

Aiming to stabilize the power system, the optimization of the voltage and first swing together was studied by Hashemi et al. [36] through this approach. They applied voltage, output power of generators, var limits of var sources, fault clearing time, and loading of the network as indices. With the objective of improving the power system hosting capacity, the authors in [38] applied the method to evaluate the deployment of soft open points within the network in conjunction with distribution reconfiguration. They applied power loss, active power delivered, and annual energy loss as the criteria.

To alleviate the network congestion, the authors in [44] applied the method to calculate the degree of severity. They applied costs, the number of curtailed customers, and the importance of curtailed load in the assessment as the criteria. To enhance the frequency stability of the power system, Lui et al. [48] applied the approach to evaluate the strategy of optimizing the power system considering different energy storage systems. They applied costs, revenue, and performance as indices for evaluation. In mitigating the challenges that could be experienced due to large area voltage collapse, Wu et al. [51] applied the approach of fuzzy TOPSIS to rank the voltage buses within the Chinese utility, considering the criteria of proportionality and system functionality.

3.9. Power System Operation Planning

The objective of the power system planning and operation is to determine the investment schedule of the generation plants and the interconnection links that ensure a reliable and efficient system based on the predicted demand [99]. With the objective of improving the power system operation and planning, Jani et al. [47] applied the method to validate the results obtained in the selection of the optimal site for the placement of ESS. Operation cost, voltage deviation, and air emission were applied as the criteria to ensure the smooth and seamless operation of the power system. To this end, Liu et al. [66] evaluated different auxiliary service models within European countries using this approach. The development of auxiliary service transactions, reasonable compliance with the trading mechanism, and benefits of the transaction mechanism were applied as the criteria of assessment within the approach.

In finding an alternative optimal black-start scheme that will greatly enhance the restoration of the system in the event of failure, Leng et al. [68] evaluated different candidate black-start schemes in selecting the best one that can be considered in the event of power failure. The criteria applied in the selection were rating capacity, unit state, startup power, ramping ratio, and number of switch operations. The authors applied the method in constructing the evaluation indicators, including comparing three power systems planning schemes [100] with each other. The scheme’s objective is to be adaptive to the volatility presented through additional generation from renewable plants.

3.10. Other Power System Applications

The application of TOPSIS stretches in several areas within the power system beyond the areas already presented. Although they are not grouped in accordance with the main themes as presented, they are outlined in this section.

To integrate thermal generation units into the power system, the TOPSIS algorithm was applied to evaluate the optimal set of solutions achieved through the combined heating and cooling plant when integrated into the power system to select the optimal ones [101]. The criteria applied for evaluation were cost, heat, and emissions.

Aiming to achieve the objective of energy management and designing an integrated system, TOPSIS was applied to evaluate the optimal size of the hybrid energy systems that will be supplying the residential area [102]. The economic, technical, and environmental criteria were applied in the evaluation. To achieve the real-time economic emission power dispatch amongst several available methods, the approach was applied to extract the best compromise optimal solution within a thermal generator [103]; fuel cost and environmental impact were applied as the criteria for evaluation.

Aiming to identify the best manufacturing plant that manufactures electrical devices, different manufacturing plants located in different towns were evaluated against each other through TOPSIS [104]. The criteria for evaluation were conducted through the imprecise or vague data applied within the approach. In achieving wide-range and long-distance electricity transmission, heat, and natural gas energy supply, the approach was applied in evaluating the comprehensive benefit of the model design [105]. Economic, environmental, and social criteria were applied in the approach.

4. Research Trends and Prospects for Future Research

Potential areas for future research were identified based on the analysis of the 78 papers downloaded from the Scopus database. VOSviewer version 1.6.19 is a software tool for constructing and visualizing bibliometric networks. These networks may, for instance, include journals, researchers, or individual publications, and they can be constructed based on citation, bibliographic coupling, co-citation, or co-authorship relations. The software also offers text mining functionality that can be used to construct and visualize co-occurrence networks of important terms extracted from the body of scientific literature [106]. The key words’ strengths of co-occurrence were calculated based on a predefined criterion.

Results

In addition to areas identified where TOPSIS can be applied, authors have investigated and identified new and potential areas that could be worthwhile applying the approach.

Table 4. List of keywords with at least five occurrences, with the number of occurrences and the total link strengths for each of those keywords shown.

Cluster	Key Word	Link Strength	Occurrences
1 (red)	Decision-Making	95	33
	Decision Theory	16	5
	Economics	14	5
	Electric Power System Control	16	6
	Electric Utilities	22	7
	Ideal Solutions	19	6
	Investments	31	9
	Optimization	15	6
	Power	28	14
	Sustainable Development	19	6
	TOPSIS	84	37
2 (green)	Electric Load Flow	18	6
	Genetic Algorithms	35	9
	Multi-Objective Optimization	78	25
	Optimal Systems	20	5
	Pareto Optimal Solutions	36	9
	Particle Swarm Optimization	22	6
3 (blue)	Carbon	19	5
	Costs	36	8
	Emission Control	20	5
	Power Markets	25	8
	Solar Energy	16	5
	Wind Power	47	11
4 (yellow)	Digital Storage	22	5
	Electric Power Systems	85	28
	Energy Storage	23	7
	Renewable Energy	50	16
	Sensitivity Analysis	23	8

presents the keywords that occur at least five times. There were 28 words that met at least the criteria of re-occurrences. These words are used to formulate the clusters of eligibility of publications’ focus, in accordance with the VOSviewer approach.

The network visualization map is shown in Figure 5. It can be observed from the visualization map that the keywords are represented in accordance with the clusters. These clusters (1,2, 3, and 4, respectively) are represented in colors (red, green, blue, and yellow) on the map as shown in Figure 5, the network visualization map.

With reference to Figure 5 and Table 4, it could be observed that there is a total of 11 keywords in cluster 1 (red). The keyword “TOPSIS” has the highest re-occurrence of 37 and strength of 84, followed by “decision-making” with the re-occurrence of 33 and strength of 95. Cluster 2 (green) contains 6 keywords, with “multi-objective optimization” followed by “genetic algorithm” and “pareto optimal solution” having the highest re-occurrence of 29 and the strength of 78; however, both “genetic algorithm” and “pareto optimal solution” have a re-occurrence of 9 each, with total strengths of 35 and 36, respectively.

Cluster 3 (blue) has 6 keywords, with “wind power” having the highest strength and re-occurrence at 47 and 11, respectively. Similarly, cluster 4 (yellow) contains 5 keywords, with “electric power systems” having a strength of 85 and a re-occurrence of 28.

Figure 6, the overlay visualization map of the keywords, shows the years in which the development of the research focuses over a period ranging from 2012 to 2022. The earlier years are represented by a darker blue color and the later years by yellow. It can be observed that the keywords in clusters 3 and 4 tend to be of interest regarding research in this area, in recent years. However, there is also a strong indication that the keywords in clusters 1 and 2 still exhibit a reasonable level of interest regarding the research focus.

In accordance with the overlay visualization map and network visualization map, electric load flow, energy storage, multi-objective optimization, renewable energy including solar energy, and carbon have been of interest within the research area in the recent years of focus. This is in line with the objective of reducing greenhouse gas emissions through the development of a less carbon-intensive power system.

Power markets and digital storage are another set of areas of interest lately, since many power industries worldwide are moving from the historic vertically operated power utilities to a more diversified market-driven approach. As demonstrated by the theme of energy markets discussed, there have been considerable publications in this area.

The power system optimization and stability aspects have evolved over the years, as evident in earlier years’ publications. The emphasis has been on congestion management, security, and power quality aspects geared toward operating the system optimally, reliably, and sustainably. However, there has still been a focus on these themes in recent years, underscoring the need for continuous research in this area.

5. Conclusions and Proposed Direction for Future Research

In this study, the authors conducted a systematic review of TOPSIS applications within power systems. The PRISMA approach was adopted, and the SCOPUS database was utilized for the publications to be included in the review. The analysis of the compiled bibliometric data under review allowed the authors to present the number of publications per annum, the cumulative number of publications over a review period, and the top-cited journal articles.

From the extracted publications, the summary of the analysis presented the application of the methodology in evaluating TOPSIS for solving power systems problems. Major themes, including sub-themes, were extracted to demonstrate the value that the approach could present. These were categorized into the following areas: calculation of weights, energy markets applications, renewable technologies assessments, heating and cooling systems combined with power systems applications, power systems operation strategy, power system stability assessment, power system operation planning, and others that could not be grouped into the themes extracted. Also, various forms of TOPSIS were extracted as applied in solving several problems outlined. Notably, the research gaps and trends were evaluated, allowing identification of future potential areas of application of the methodology in contribution to the existing knowledge within this area of application. The graphical abstract, as shown in Figure 7, summarizes the approach, analyses, and the directions for future research as presented in this study.

Based on the analysis of gaps identified using VOSviewer in conjunction with the qualitative analysis of the reviewed material as extracted from the derived themes, this assisted in the identification of potential research areas based on the extracted themes, which are as follows:

• The rollout of electric vehicles (EVs) is a means of combating the effects of greenhouse gas emissions and is part of synergistic technological development. More site identification is needed in preparation for the mass rollout of the charging infrastructure.
• In planning for a suitable location where high voltage direct current (HVDC) lines are constructed to transmit renewable energy from a remote location to the desired load centre, often the potential terminating points will consist of several locations characterized by conflicting criteria. To mitigate the bias of ideal terminating point selection, the approach can be applied to select the preferred terminating point, as well as to rank the other potential points among them.
• Deployment of modern advanced technologies in the form of grid software and grid hardware technologies, such as FACTS devices, could potentially offer the benefit of managing and operating the power system reliably and sustainably for the foreseeable future. Identification of these optimal sites within the power system could be done through this approach.
• Studies aimed at combating the effects of extreme weather events due to the global warming phenomenon can help avert the risk and threats that might affect the power system. Applying the method to identify vulnerable power corridors within the integrated power system will be beneficial.
• Several maintenance strategies are applied in executing the maintenance plans aimed at ensuring the extension and operational reliability of various power system components. However, selecting the preferred plan often creates a challenge for utility maintenance personnel. To assist the selection approach, the methodology can be applied given its effectiveness in assessing the preferred choice amongst several potential options in place. These will improve the prioritization of critical components for maintenance selection.
• The sustainability of the power system can be assessed through this approach, where environmental, social, and economic elements can be factored into the evaluation as the criteria considered. This will assist in ensuring that the sustainable power system is maintained.
• Planning for the power system substation location often creates a challenge of selecting the preferred location, given that there will be several potential sites in existence. All the sites will be characterized by several conflicting criteria that can be applied for selection. To eliminate the element of bias in the selection process and remove doubts for the decision-makers about not selecting other potential locations, the approach will assist in mitigating the drawback highlighted.