Methodologies for Technology Selection in an Industry 4.0 Environment: A Methodological Analysis Using ProKnow-C

Abstract

In an ever-evolving digital environment, organizations must adopt advanced technologies for real-time big data processing to maintain their competitiveness and growth. However, selecting appropriate technologies is a challenge, particularly for small and medium-sized enterprises (SMEs). This study develops a literature review to analyze the methodologies used in the selection of technologies, with a special focus on those associated with the Industry 4.0. Knowledge Development Process-Constructivist (ProKnow-C) method, which was used to build a bibliographic portfolio, examining approximately 3400 articles published between 2005 and 2024, from which 80 were selected for a detailed analysis. The main methodological contributions come from research articles, the ScienceDirect database, the Expert Systems with Applications Journal, studies conducted in Turkey, and publications from the year 2023. The results highlight the predominant use of multi-criteria techniques, emphasizing hybrid approaches that combine various decision-making methodologies. In particular, the analytic hierarchy process (AHP) and TOPSIS methods were employed in 51.25% of the analyzed cases, either individually or in combination. It is concluded that technology selection should be based on flexible and adaptive approaches tailored to the organizational context, aligning long-term strategic objectives to ensure business sustainability and success.

1. Introduction

In today’s dynamic business landscape, information systems constitute a fundamental tool for business organizations. These technologies empower organizations to manage their operations more efficiently, interact more effectively with suppliers and customers, and compete successfully in the market [1]. This increasing reliance on technology is driven by the rapid pace of technological evolution, making the adoption of new technologies an imperative for companies seeking to increase their market value [2]. The implementation of innovative technological solutions, such as process automation, data analysis, and artificial intelligence, optimizes operational efficiency by reducing costs, improving productivity, and streamlining decision-making. This allows companies to become more competitive and quickly adapt to market changes [3].

Moreover, the use of e-commerce platforms, social networks, and digital marketing tools facilitates expansion into new markets by eliminating geographical barriers and enabling access to customers in various regions, contributing to the diversification of revenue sources by opening new sales channels and reaching previously untapped customer segments [4]. Effective technology management enhances business performance by providing competitive advantages, integrating processes, and fostering innovation—key factors for success in the digital era [5]. Technology is a crucial factor in competitiveness. It enables companies to differentiate themselves, increase profitability, and establish strategic alliances, fostering long-term success in dynamic markets [6].

Industrial production is undergoing a radical transformation driven by the Fourth Industrial Revolution and its digital technologies [7]. In this context, the concept of Industry 4.0 emerges as the most tangible manifestation of this revolution, fostering smart manufacturing in an innovative environment [8]. This paradigm shift facilitates the circular design of product life cycles while enhancing the implementation and management of complex ecosystems. These ecosystems provide real-time information and enable autonomous interactions among machines, systems, and objects [9].

The definition of Industry 4.0 that best aligns with the focus of this study is the one proposed by Rupp et al. [10]. This perspective offers a comprehensive, practical application of emerging technologies in production environments, which supports the methodological approach centered on technology selection. These authors conceive Industry 4.0 as the implementation of cyber-physical systems for the creation of smart factories. This vision relies on technological pillars like big data and data analytics, advanced robotics, the Internet of Things, additive manufacturing, augmented and virtual reality, cloud computing, and advanced cybersecurity. These technologies collectively enable real-time information exchange across the value chain [11], giving rise to smart factories capable of optimizing processes, improving efficiency, and personalizing production.

The landscape of Industry 4.0 is continuously evolving, driven by the advancement of several emerging technologies. Key among these is artificial intelligence, including machine learning and its subfield deep learning, digital twins, blockchain, 5G, edge computing, and the metaverse, which are increasingly being adopted by companies across various sectors [12]. This adoption not only reflects a trend toward modernization but also opens new opportunities in key areas such as advanced automation, data-driven decision-making, simulation, and collaboration in virtual environments [13].

The effective integration of the technological pillars and emerging technologies of Industry 4.0 can generate a significant competitive advantage for companies [14]. This advantage results in tangible benefits that drive growth and market success, such as increased efficiency, cost reduction, quality improvement, the creation of new business models, and greater flexibility and adaptability [15]. However, addressing these challenges requires the development of new skills in information technology and data analytics [16]. Companies that overcome these challenges can achieve a significant competitive advantage by establishing themselves as leaders in their respective sectors [17]. However, addressing these challenges requires the development of new skills in information technology and data analytics [18]. Moreover, to achieve effective digital transformation, organizations must adjust their business models and optimize all operational processes related to products and services [19].

In this context, higher education institutions play a crucial role in preparing future professionals and aligning them with labor market demands. This requires not only updating curricula but also improving the educational infrastructure to meet the requirements of Industry 4.0 [20].

Likewise, it is essential to create adequate training programs, develop standards and regulations that promote innovation and security, and foster a culture of collaboration and continuous learning [21]. Collaborative approaches are essential to maximizing the potential of Industry 4.0 and mitigating its risks [22]. In this regard, the triple helix model is crucial, as it fosters interaction among three key actors: businesses, governments, and universities [23]. Choosing the right approach is crucial for developing effective strategies [24].

Technology selection represents one of the most complex challenges for organizations, as it involves choosing and investing in a technological field that offers competitive advantages among multiple alternatives [25]. To make more informed decisions, the creation of key performance indicators in these processes is essential, as they provide quantifiable data on the performance of selected technologies [26]. These indicators enable the evaluation of factors such as the return on investment, scalability, compatibility with existing systems, sustainability, and the level of acceptance by employees and customers, among many others [27,28]. An appropriate technological choice optimizes processes, enhances efficiency, fosters innovation, and facilitates adaptation to change [29].

In this regard, adaptability is particularly important in highly volatile industrial markets, where rapid change is the norm. The COVID-19 pandemic provided a stark example of how companies that quickly adopted technological innovations, such as online platforms, e-commerce, and logistics solutions, were able to remain competitive, while those that did not were often forced to exit the market [30]. This challenge is even more critical for SMEs, whose adaptability and competitiveness often hinge on effective technology selection [31].

While most of the analyzed studies on technology selection in the context of Industry 4.0 focus on developing decision-making frameworks and roadmaps for practical implementation within organizations, this study offers a distinct contribution by employing the ProKnow-C method to conduct a systematic and rigorous analysis. Unlike previous approaches, ProKnow-C enables the structured and transparent selection of relevant scientific literature through well-defined inclusion and exclusion criteria. This methodological rigor supports a critical evaluation of the new methodological proposals or the evolution of existing approaches, uncovers gaps in the literature, highlights emerging trends, and identifies opportunities for improvement. Consequently, this study provides a solid foundation for guiding future research in this area.

1.1. Groups in Technology Selection

Based on a review of the literature, seven main groups are proposed to structure the technology selection process within the field of engineering. These groups are summarized in Figure 1: (i) problem definition, (ii) determination of evaluation criteria, (iii) identification of candidate technologies, (iv) assignment of weights to the criteria, (v) evaluation of alternatives, (vi) prioritization and selection of the optimal alternative, and (vii) recommendation and sensitivity analysis. This approach enables systematic evaluations, comparisons, and the adoption of technological alternatives.

• Problem definition: In this group, the primary objective of the decision is established by identifying the organization’s needs, the challenges of the production environment, and any technical, economic, or strategic constraints that must be considered. A clear and precise problem definition is essential to ensure coherence throughout the evaluation process and to align criteria with the objectives of the manufacturing system [32,33,34].
• Identification of candidate technologies: A systematic exploration is conducted to identify potential Industry 4.0-related technologies applicable to the organization’s specific context. This process may include literature reviews, expert consultations, analyses of relevant use cases, and an initial screening to exclude non-viable options, thereby ensuring a representative and context-relevant set of alternatives [35,36,37].
• Determination of evaluation criteria: This step involves selecting the criteria to be used for evaluating the candidate technologies. The criteria may span technical, economic, organizational, environmental, or social dimensions and should be defined according to the problem’s context. Validation with experts and hierarchical categorization of the criteria help ensure their relevance and applicability [38,39,40].
• Assignment of weights to the criteria: As not all criteria are equally important, weights are assigned using techniques such as AHP, SWARA, or other structured methods. This step requires expert participation and mechanisms to verify the consistency of judgments, ensuring that the assigned weights appropriately reflect strategic priorities [41,42,43].
• Evaluation of alternatives: A decision matrix is constructed in which candidate technologies are quantitatively assessed against each criterion. This information can be normalized and processed using multiple-criteria decision-making (MCDM) techniques such as TOPSIS, VIKOR, and PROMETHEE, among others. The result is a comparative evaluation of the alternatives based on their multidimensional performance [44,45,46,47].
• Prioritization and selection of the optimal alternative: Based on the scores obtained, a ranking of technologies is produced, enabling identification of the most suitable option. This group may also involve a risk analysis, verification of technical feasibility, and discussions with key stakeholders to ensure that the final decision is grounded in a comprehensive perspective [37,48,49].
• Recommendation and sensitivity analysis: Finally, a report is prepared that details the process, the results obtained, and recommendations for implementation. The sensitivity analysis is used to examine how variations in weights or criterion values affect the outcome, contributing to validation of the decision’s robustness and providing feedback for future applications [50,51,52].

A single study may address multiple groups of the technology selection process; therefore, the percentages are not mutually exclusive and do not add up to 100%. This approach allows for a more accurate representation of the thematic coverage of each publication. Significant strengths were identified in the current literature in the group (v) evaluation of alternatives (80%) and (vi) prioritization and selection of the optimal alternative (70%). Most studies focus on the final phases of the process, particularly on the application of multi-criteria decision-making methods—such as AHP, TOPSIS, and VIKOR, among others—to assess and select the best alternative. This reflects a strong interest in structured decision-making methodologies, which are beneficial for technically sound, quantitatively justified decisions.

Group (iii), determination of evaluation criteria (75%), also shows a high concentration of studies. This highlights a solid understanding of the factors that influence technology adoption, such as the cost, performance, sustainability, or flexibility—elements essential to well-substantiated decision-making.

Regarding weaknesses or gaps, group (iv), the assignment of weights to the criteria, exhibits lower coverage (47.5%). Fewer than half of the studies explicitly address how weights are assigned, indicating a lack of in-depth analysis of the relative importance of the criteria involved. In many cases, weights are assigned arbitrarily or without empirical justification, which may compromise the validity of the final decision.

Group (vii), recommendation and sensitivity analysis, is among the least addressed (37.5%). This suggests that the robustness of technology decisions is rarely evaluated, despite the need to validate decisions under different scenarios due to uncertainty or contextual changes. The absence of a sensitivity analysis may undermine the reliability of the proposed recommendations.

Vulnerabilities were also identified in the initial groups: (i) problem definition (65%) and (ii) identification of candidate technologies (60%). Although these percentages are not particularly low, they indicate that 35% to 40% of the studies do not adequately contextualize the problem or explore a sufficiently broad range of available technological options. Poor problem formulation can lead to misdirected decisions, while a limited identification of alternatives may compromise the quality of the final solution.

1.2. Contributions and Limitations

The main contributions of this study include the following:

• Providing a clear understanding of how organizations determine which technologies best suit their specific needs.
• Identifying and analyzing methodologies used by companies for technology selection.
• Examining emerging trends and key factors influencing technological decisions, such as alignment with strategic objectives, integration with existing systems, and adaptability to future changes.
• Clarifying the technology selection process, providing a more detailed understanding
• Providing a reference framework to help companies adopt technologies that enhance their long-term competitiveness and efficiency.

The main limitations of this study include the variability in technology selection methodologies across geographical contexts, local regulations, and technological infrastructures may limit the generalizability of findings. Furthermore, the diversity of industrial sectors complicates the identification of common patterns. Methodologically, the focus on English-language articles and the use of ProKnow-C, which relies on title and keyword searches, may have resulted in the exclusion of some relevant studies.

The remainder of this study is organized as follows: Section 2 presents and validates the methodology used. Section 3 provides a literature analysis of the technology selection, with a particular focus on the Industry 4.0 context, based on 80 studies that form the bibliographic portfolio. Section 4 analyzes and discusses the literature review results using a network and density map. Finally, Section 5 presents the main conclusions and proposes future lines of research.

2. Methodology

To support the knowledge construction process, the ProKnow-C method was used, as proposed by Ensslin et al. [53] and recently validated by De Carvalho et al. [54]. ProKnow-C is a structured approach for developing a relevant bibliographic portfolio, involving steps such as identifying keywords, searching databases, filtering and selecting articles, and conducting a bibliometric analysis [55,56]. This method was chosen for its systematic and comprehensive approach to a literature review, enabling a rigorous and transparent process for identifying key publications relevant to this study.

ProKnow-C has been used by various authors in different contexts. For instance, it has been employed to characterize performance indicators in energy management and Industry 4.0, exploring its theoretical foundations and research opportunities [57]. Similarly, it has been used to develop a bibliographic portfolio of solar energy generation forecasting, analyze trends, and provide a theoretical framework for future research, particularly in Latin America [58]. Additionally, it has been applied to define the state of the art in lithium-ion battery recycling and to build a database of relevant publications for future research in this field [59].

Filtering Process

This section outlines the construction of a highly relevant scientific bibliographic portfolio, employing an adaptation of the ProKnow-C methodology as developed by [60]. The initial stage, stage (1), involves the definition of keywords. This step is designed to precisely capture the research focus.

To ensure comprehensive coverage, keywords like ‘technology selection’, ‘Industry 4.0 selection’, and ‘technology prioritization’ were identified, reflecting the study’s core concepts. The search was also expanded to complementary areas, such as project or portfolio selection that integrates technologies, to analyze selection strategies in these contexts. This led to the inclusion of additional terms like ‘multicriteria decision-making methods’, ‘project selection’, ‘portfolio selection’, and ‘strategic selection’, ensuring that the collected literature is as complete and relevant as possible for further analysis.

In stage (2), an exhaustive search, multiple databases were explored, rather than restricting the search to a few specific sources. This strategy enabled a broader and more diverse identification of relevant articles for the research topic, increasing the coverage of potentially significant studies and enhancing the quality of the bibliographic portfolio.

The adherence test, stage (3), is crucial for verifying the suitability of the selected keywords, ensuring that the collected literature aligns with the study’s focus and contributes to enhancing the quality of the bibliographic portfolio.

To refine the search, filters were applied in the database. First, the search was limited to articles published between 2005 and 2024, capturing the evolution of the research area and ensuring its relevance to the current context. Additionally, a language filter restricted the search to English documents, given the prevalence of relevant literature in this language. Furthermore, tailored search equations were applied in the databases, strategically combining the predefined keywords to improve the relevance of the information. The search equations used are as follows:

• ‘Technology Selection’ OR ‘Industry 4.0 Selection’ OR ‘Technology Prioritization’.
• (‘Project Selection’ AND ‘Technolog* Selection’) OR (‘Project Selection’ AND ‘Portfolio Selection’ AND ‘Technolog*’).
• (‘Strategic Selection’ AND ‘Technolog* Selection’) OR (‘Multi Criteria Decision Making’ AND ‘Technolog* Selection’) OR (‘Strategic Selection’ AND ‘Industry 4.0 Selection’ AND ‘Technolog* Selection’) OR (‘Multi Criteria Decision Making’ AND ‘Industry 4.0 Selection’ AND ‘Technolog* Selection’).
• (‘Multi Criteria Decision Making’ AND (‘Technolog* Selection’ OR ‘Technolog* Prioritization’ OR ‘Industry 4.0 Selection’ OR (‘Project Selection’ AND ‘Technolog*’) OR (‘Portfolio Selection’ AND ‘Technolog*’))) OR (‘Strategic Selection’ AND (‘Technolog* Selection’ OR ‘Technolog* Prioritization’ OR ‘Industry 4.0 Selection’ OR (‘Portfolio Selection’ AND ‘Technolog*’) OR (‘Project Selection’ AND ‘Technolog*’))).

Numerous articles were identified using the search engines Web of Science, SCOPUS, and Google Scholar, accessing platforms such as the Wiley Online Library, ScienceDirect, SpringerLink, Emerald Insight, Taylor & Francis Group, and IEEE Xplore Digital Library, among others (Figure 2).

Approximately 3400 documents were identified in this initial review, advancing to stage (4), title screening. At this stage, an acceptance criterion was applied based on the presence of at least one predefined keyword from the search equations in the document title and the indication that the study employed a methodology for selecting technologies, portfolios, or projects involving a specific technology. This preliminary filter was essential to ensure the alignment of the articles with the research focus and to narrow the document set to those most relevant to the study

With 363 documents remaining after the title selection process, stage (5) involved a full reading of the abstract. This stage verified whether the central topic explicitly presented a methodology for selecting technologies, projects, or portfolios involving technologies, particularly those related to Industry 4.0. The analysis focused on identifying methodologies based on optimization, MCDM methods, and other relevant techniques. Consequently, 215 articles were discarded, leaving 148 selected documents.

In stage (6), the availability of the documents for full review was verified. During this process, 16 were discarded due to access restrictions or costs, as some databases were redirected to scientific journals where the selected studies were inaccessible. As a result of this filter, 132 documents remained.

In stage (7), full-text reading of the articles, the objective was to verify that each study presented a detailed methodology and clearly explained its application. The relevance of the methodologies, selection criteria, and obtained results was also evaluated. Consequently, 52 studies that did not meet the review criteria were discarded, ensuring that the final selection included only high-quality and relevant documents. Finally, in stage (8), definition of the bibliographic portfolio, 80 articles were selected as highly relevant to the subject under study (Figure 3).

3. Literature Review

This section provides a comprehensive analysis of the literature on technology selection, focusing on technologies applicable to Industry 4.0, key trends, and factors influencing business technological decisions. This study provides a reference framework for understanding the field’s current state and identifying emerging research opportunities.

For each study, the quartile (Q) according to the Scimago Journal Rank (SJR) is reported, along with its research domain, main contribution, solution methodology, and type of selection. With the aim of systematically organizing and analyzing the existing knowledge, the literature was grouped into four categories: (1) multi-criteria decision-making techniques (studies 1–63); (2) optimization techniques (studies 64–70); (3) literature reviews on technology selection methodologies (studies 71–74); and (4) proposals for new strategies in technology selection (studies 75–80). In total, 80 documents were classified in

Table 1. Bibliographic portfolio of studies.

Ref./Quartile	Domain	Contribution	Methodology	Type of Selection
1/Conference	Selection of radio access technologies in vehicular networks [38].	Multi-criteria framework for optimizing cooperative intelligent vehicular networks.	Technique for order preference by similarity to ideal solution (TOPSIS)	TS
2/Conference	Evaluation and prioritization of renewable energy technologies [39].	Guide for sustainable energy planning and identification of multi-criteria technologies.	TOPSIS	TS
3/Q1	Decision-making in medication dispensing processes [61].	Identification of seven critical factors for technology adoption.	TOPSIS and AHP	TS
4/Q2	Identification of 3D printing technologies for prosthesis manufacturing [41].	Expert panel for weighting and ranking alternatives.	TOPSIS, AHP, and PROMETHEE	TS
5/Q1	Evaluation of 3D printing technologies for dental models [62].	Network meta-analysis for comparing seven printing technologies by precision.	Literature review	TS
6/Q1	Multi-criteria methods for assessing the potential of the social Internet of Things [63].	Evaluation of technologies for modeling and predicting human behavior.	TOPSIS and entropy weighted method	I4.0
7/Conference	Selection of localization technologies [64].	Multi-criteria framework for assessing technological alternatives in a balanced and objective manner.	TOPSIS and geometric mean method	I4.0
8/CiteScore 4.5	Hybrid methodology for the selection of industrial robots [42].	Classification of robots based on proximity to the ideal solution and multiple criteria.	TOPSIS and best worst method (BWM)	I4.0
9/Q1	Selection of energy storage technologies [65].	Structured guide for optimizing technology selection under uncertainty.	TOPSIS and triangular fuzzy	TS
10/Q3	Selection of technology in the petrochemical industry [35].	Identification and prioritization of key factors in selection using multi-criteria methods.	TOPSIS, Delphi method, and Shannon entropy	TS
11/-	Selection of lighting technologies [45].	Multi-criteria decision model with technical and sustainable criteria.	Elimination and choice expressing reality (ELECTRE III)	TS
12/Q3	Project management and decision-making in information technology [66].	Support system for evaluating and prioritizing projects using multi-criteria and multi-attribute criteria.	TOPSIS, multi-attribute decision-making (MADM) and multi-objective decision-making (MODM)	PRS
13/Conference	Selection of technologies to optimize logistics and supply chain management [44].	Multi-criteria method for evaluating technological alternatives under uncertainty and ambiguity.	Fuzzy TOPSIS	TS
14/Book chapters	Selection of technologies for food waste management [67].	Fuzzy multi-criteria method for evaluating technologies under uncertainty.	Fuzzy TOPSIS	TS
15/Q2	Evaluation of technological impact on manufacturing strategies [33].	Hybrid method for calculating weights and classifying technological influence.	Fuzzy TOPSIS and Fuzzy AHP	I4.0
16/Q2	Selection of technologies for effective manufacturing operations management [32].	Framework for prioritizing technologies based on economic, social, and environmental criteria.	Fuzzy TOPSIS and Fuzzy AHP	I4.0
17/Q4	Identification of key digitalization elements for efficient operations [68].	Framework for identifying key factors and metrics.	Fuzzy TOPSIS and Fuzzy AHP	I4.0
18/-	Selection of technologies for competitive markets [40].	Framework for structuring and defining evaluation indicators.	Fuzzy TOPSIS and Fuzzy AHP	TS
19/Conference	Optimal selection of technological strategies using multi-criteria approaches [51].	Prioritization of strategies based on human resources and organizational factors.	AHP and VIseKriterijumska Optimizacija I Kompromisno Resenje (VIKOR)	I4.0
20/Q4	Selection of renewable energy technologies [69].	Fuzzy multi-criteria method for classifying categories and criteria.	Fuzzy TOPSIS and PROMETHEE	TP
21/Q1	Evaluation of technological projects in the aerospace industry [52].	Fuzzy, hybrid, and structured multi-criteria framework in two modules.	Fuzzy TOPSIS and analytic network process (ANP)	TP
22/Conference	Selection and prioritization of technological projects in business portfolios [70].	Design of a multi-criteria system integrating strategic objectives and critical success factors.	Fuzzy TOPSIS, PRISMA, and ANP	PRS
23/Q1	Selection of emerging technologies for R&D at the national level [71].	Uncertainty management in decision-making without criterion decomposition.	DAHP (Dual AHP or AHP bayesiano)	TP
24/CiteScore. 4.3	Technological adoption in manufacturing [49].	Classification of key technologies with expert validation.	AHP	I4.0
25/Q1	Selection of communication technologies for electrical substations [36].	Decision framework with technical, economic, and adaptability criteria.	AHP	TS
26/Conference	Implementation of technologies in additive manufacturing [72].	Hybrid approach for prioritizing criteria and selecting technologies.	AHP and Fuzzy VIKOR	I4.0
27/Conference	Selection of sustainable energy technologies [73].	Reducing uncertainty in perception and weight assignment.	Fuzzy TOPSIS, AHP, and Fuzzy VIKOR	TS
28/Q1	Selection of technologies to reduce emissions in the maritime sector [74].	Hybrid multi-criteria evaluation of technologies under uncertainty and incomplete data.	AHP, VIKOR, and Fuzzy AHP	TS
29/Conference	Selection of internet access technologies [75].	Multi-criteria framework for assessing technologies based on cost, safety, and coverage.	AHP and DecideIT	TS
30/Q1	Selection of renewable energy technologies [76].	Hybrid method for weighing technical, economic, and social criteria.	AHP and combinative distance-based assessment (CODAS)	TS
31/-	Selection of technologies for energy management [77].	Key criteria and framework for technology selection and decision-making.	AHP and literature review	TS
32/Q1	Forecasting and prioritization of investments in emerging technologies [78].	Quantitative method based on a multi-criteria tool.	AHP and cluster analysis of factor scores	TS
33/Conference	Selection and evaluation of manufacturing technologies [79].	Hybrid methodology considering technology, competition, and management.	AHP and quality function deployment (QFD)	I4.0
34/Q2	Prioritization of technological projects under uncertainty and expert judgment [80].	Arithmetic means for evaluating projects with five criteria and four alternatives.	AHP and q-spherical fuzzy rough set	I4.0
35/Q1	Evaluation of critical factors for technology implementation [48].	Prioritization of key factors and industrial sensitivity analysis.	AHP and ANP	I4.0
36/Conference	Integration of digital technologies in industrial processes [34].	Technology selection framework with a social and business approach.	Action research with iterative strategic analysis	I4.0
37/Q1	Evaluation and selection of R&D projects [81].	Hierarchy of criteria in five levels and scoring guide.	AHP, data envelopment analysis (DEA), and utility additive discriminant (UTADIS)	PRS
38/Q2	Selection of technologies in manufacturing [47].	Three-stage hybrid model for weighting relevant factors.	Fuzzy AHP and Fuzzy ANP	I4.0
39/Q1	Ranking and selection of strategic projects [82].	Modeling relationships between projects and financial objectives of the balanced scorecard.	Decision-making trial and evaluation laboratory (DEMATEL)	PRS
40/Q1	Selection of emerging technologies using multi-criteria methods [46].	Hybrid model integrating patent co-citation analysis to identify technological areas.	DEMATEL, Fuzzy Delphi, ANP, and network relationship map (NRM)	TS
41/Q1	Selection of R&D project portfolios [83].	Hybrid method for managing interdependence and feedback in decisions.	DEMATEL, modified Delphi method, and ANP	PS
42/Q2	Optimal selection of industrial robots based on application [37].	Comparison of multi-criteria methods for selecting criteria in manufacturing.	TOPSIS, simple additive weighing, linear programming technique, VIKOR, ELECTRE III, and net flow method	I4.0
43/Conference	Enterprise interoperability to optimize performance using Lean indicators [84].	Technology prioritization framework with diagnostics and performance evaluation.	ELECTRE I and PROMETHEE II	I4.0
44/Q1	Selection of robots for industry and manufacturing [85].	Group decision-making for robot selection with weights and uncertainty.	ELECTRE II, VIKOR, and group decision making	I4.0
45/Q1	Technology selection with relative efficiency assessment [86].	Multi-criteria methodology with common weights that reduces computational cost.	DEA	TS
46/Q1	Selection of wafer-cutting technology for integrated circuits [87].	Decision model with static, dynamic, and expert-driven factors.	Fuzzy Logic	TS
47/-	Selection of remanufacturing technologies [88].	Evaluation of technologies considering key criteria and uncertainty.	Fuzzy Logic	TS
48/Q3	Selection of technologies for industrial wastewater treatment [89].	Evaluation framework for decision-making, analyzing five technologies across ten criteria.	Fuzzy VIKOR	TS
49/Q1	Implementation of technologies in SMEs [90].	Identification of key factors for implementation using a multi-criteria approach.	Fuzzy complex proportional assessment	I4.0
50/Q1	Optimization of last-mile logistics with technology selection [50].	Weighting of criteria for selecting the best technology.	Fuzzy ANP and Fuzzy ADAM	I4.0
51/Q1	Selection of technologies for solid waste disposal [91].	Incorporating uncertainty into long-term decisions and calculating technology rankings.	BWM	TS
52/Q1	Selection and prioritization of strategies for technology implementation [43].	Hybrid approach for identifying key attributes.	BWM, Fuzzy, tomada de decisão interativa multicritério (TODIM), and interval-valued intuitionistic fuzzy (IVIF)	I4.0
53/Q1	Selection of energy storage technologies [92].	Multi-criteria model considering specific storage requirements.	Probabilistic dual hesitant fuzzy set	TS
54/Conference	Selection of advanced battery technologies for electric vehicles [93].	Multi-criteria model for evaluating emerging technologies based on key performance factors.	Weighted product model	TS
55/Q2	Selection of R&D projects through technology foresight [94].	MCDM model for evaluating and weighting criteria.	Stepwise weight assessment ratio analysis	PRS
56/Q3	Selection of renewable energy technologies [95].	Integrated model for evaluating and weighting criteria and classifying alternatives.	Evaluation based on distance from average solution (EDAS)	TS
57/Conference	Selection of access technologies in heterogeneous wireless networks [96].	MCDM model based on operator and user satisfaction.	Satisfaction-based	TS
58/Q1	Selection of energy conservation and emission reduction technologies [97].	Model based on rank centrality and differences among experts.	Large-scale group decision-making	TS
59/Conference	Selection of high-tech projects in an uncertain environment [98].	Algorithm for calculating preferences considering all criteria.	Weighted generalized hesitant fuzzy power average	PRS
60/Conference	Selection and prioritization of R&D projects in the aerospace industry [99].	Structured questionnaire for experts and decision support system for technology selection.	Scoring algorithm	PRS
61/Q1	Selection of Internet of Things technologies in urban transportation [100].	Method for analyzing relationships, weighting criteria, and prioritizing applications.	Fuzzy cognitive map, BWM, and additive ratio assessment	I4.0
62/Conference	Selection of cloud computing technology [101].	Method for identifying criteria, managing uncertainty, and conducting multi-criteria evaluation.	IVIF and MULTIMOORA	TS
63/Conference	Selection of cloud services [102].	Methodology for classifying cloud services based on non-functional properties.	REMBRANDT method	TS
64/Q3	Portfolio selection analysis for R&D projects [103].	Multi-criteria model for evaluating projects considering interrelations and synergies.	Nonlinear optimization	PS
65/Book chapters	Selection of digital transformation project portfolios [104].	Model with multiple constraints, interdependencies, and uncertainty management.	Optimization and integer programming	I4.0
66/Q3	Selection of technologies in the context of digital transformation [105].	Hybrid framework integrating key dimensions through a quantitative approach.	AHP, QFD, and mixed integer programming	I4.0
67/Book chapters	Selection of technological projects [106].	Optimization model for maximizing benefits and evaluating scenarios under uncertainty.	Stage-Gate	I4.0
68/Q1	Selection and impact of technologies in manufacturing companies [107].	Technology evaluation, portfolio construction, and operational improvements.	Bayesian network and machine-learning algorithms	I4.0
69/Q1	Analysis and selection of technologies in advanced manufacturing [108].	MCDM approach with common weights and an efficient algorithm.	DEA and mixed integer linear programming	TS
70/Conference	Projects in information and communication technologies [109].	Agent-based approach for modeling uncertainty with integer linear programming.	Fuzzy Agent	PRS
71/Q3	Decision-making for the selection and implementation of technologies [110].	Clustering of key criteria with multi-factor evaluation.	Literature review	I4.0
72/-	Structured decision-making for technology selection in manufacturing [111].	Technology identification and strategic roadmap development.	Literature review	I4.0
73/Q2	Evaluation and selection of technologies using multi-criteria analysis [112].	Evaluation of trends, industrial technologies, and their sub-areas.	Literature review	TS
74/Q2	Selection and evaluation of R&D projects in the communication sector [113].	Evidence-based approach for R&D prioritization and evaluation.	Literature review	PRS
75/Technical Report	Selection of technologies for big data analysis [114].	Framework for classifying technologies, identifying key criteria, and guiding technological adoption.	S.T.A.D.T. selection framework (SSF)	I4.0
76/Conference	Selection of automation technologies in manufacturing [115].	Framework for technology analysis, computational cost optimization, and investment criteria incorporation.	New methodology	TS
77/CiteScore 1.7	Selection of technologies for manufacturing [116].	Quantitative methodology for selecting technology providers based on evaluation dimensions.	New methodology	I4.0
78/Conference	Selection of technology portfolios in the aerospace industry [117].	Methodology for resource optimization in commercial and governmental environments.	Strategic planning and prioritization	PS
79/Q1	Innovation management and selection of R&D projects [118].	MCDM model considering technology and customer value.	AHP and new methodology	PRS
80/Q1	Selection of technology for remanufacturing [119].	MCDM model considering economic benefits, environmental impact, and technological synergies.	AHP and new methodology	PS

Technology selection (TS), Industry 4.0 (I4.0), Project selection (PRS), Technology prioritization (TP), Portfolio selection (PS).

. Abbreviations for the type of selection are as follows: Technology selection (TS), Industry 4.0 (I4.0), Project selection (PRS), Technology prioritization (TP), Portfolio selection (PS).

Analysis of the Bibliographic Portfolio

The portfolio encompasses a wide range of methodologies, reflecting the diversity of approaches and strategies in the academic literature on technology selection. This heterogeneity responds to the varying requirements and contexts of each researcher, who tailor methodologies to their specific needs. A significant portion of the analyzed documents falls within category (1), focusing on MCDM techniques. These studies not only apply existing MCDM methodologies but also integrate different approaches to better adapt to each context, considering the available alternatives, evaluated criteria, and objectives.

MCDM methods enable the evaluation of alternatives by comparing technical, economic, environmental, and social criteria [120]. In this study, they stand out for their ability to integrate multiple factors and criteria, facilitating the effective selection of technologies aligned with specific objectives and contexts.

A key finding is the use of hybrid MCDM approaches, combining different multi-criteria techniques. Typically, one methodology determines the weights of evaluation criteria for technological alternatives, while another selects the most suitable option based on these weighted criteria.

Furthermore, some studies integrate more than two methodologies, reflecting an interest in comprehensive and detailed technology-selection approaches. This combination enhances the robustness and reliability of multi-criteria analysis results and increases decision confidence. Additionally, comparing different MCDM tools in specific contexts helps identify the most suitable methodology for each case, such as selecting industrial robots [37].

AHP and TOPSIS were the most frequently used MCDM methods in the reviewed literature. Specifically, AHP was identified in 27 of the 80 analyzed articles and TOPSIS in 22. This high frequency of use suggests that both approaches are widely recognized as valuable multi-criteria tools in the academic field of technology selection.

AHP is used in different contexts and with different approaches in the reviewed studies. It is used directly for technology selection or prioritization, facilitating decision-making among multiple alternatives. For instance, AHP is employed to select communication technologies for substations across urban, suburban, and rural environments, evaluating criteria such as technical standards, economic factors, and infrastructure maturity [36].

Beyond its direct application in technology selection, AHP is also used to assign weights to evaluation criteria, allowing for the quantification of their relative importance in the decision-making process. This capability makes it a key tool in hybrid approaches, where it is combined with other multicriteria methods such as TOPSIS, VIKOR, or ANP to enhance the final selection. One methodological proposal within this approach uses AHP to weight criteria and TOPSIS to select technologies in hospital dispensing processes, enabling structured and precise decision-making [61].

These techniques are also widely applied in technology selection, especially in contexts of uncertainty and imprecision in expert judgments. To address these challenges, MCDM methods are employed in fuzzy environments.

A representative study analyzes Industry 4.0 technology selection in manufacturing companies. The authors identify big data, cloud computing, cyber-physical systems, Internet of Things, computer simulations, blockchain, autonomous/industrial robots, and additive manufacturing as high-impact technologies. The study proposes a methodology combining Fuzzy AHP and Fuzzy TOPSIS, using a set of indicators designed to capture various dimensions of the companies, applying this method to a case study. The results rank cyber-physical systems, big data, and autonomous/industrial robots as the most relevant [32]. Similarly, other studies combine Fuzzy AHP and Fuzzy TOPSIS for Industry 4.0 technology selection [33,68].

TOPSIS has also been employed individually in technology-selection studies. Its main advantage is its simplicity and ranking accuracy compared to other MCDM algorithms [121]. Additionally, it effectively adapts to both quantitative and qualitative criteria, making it a versatile tool [122].

The flexibility of TOPSIS has facilitated its application across various sectors. In a case study conducted in Iran, it was utilized to evaluate photovoltaic solar energy, concentrated solar power, wind turbines, and geothermal systems, analyzing six key criteria to determine the most suitable option [39]. Likewise, this method has been applied in vehicular networks to optimize the selection of radio access technologies, a challenge of increasing relevance due to the development of cooperative intelligent transport systems [38].

Integrating AHP with Fuzzy VIKOR is another common multicriteria technique, used to facilitate Industry 4.0 technology implementation and accelerate customized product and process introduction [72]. AHP has also been combined with a QFD-based analysis framework to select technologies and generate usage scenarios, helping SMEs identify and prioritize suitable solutions for their needs [79]. In renewable energy, AHP has been used to compare hybrid solar-wind, solar mini-grid, and wind mini-grid technologies. These studies consider economic, technical, environmental, and sociopolitical criteria, using AHP for criteria weighting and CODAS for alternative ranking [76].

Similarly, a methodology for selecting strategic projects in a company is presented and can be adapted for technology initiatives. This proposal is based on the use of the balanced scorecard and combines DEMATEL with MCDM to establish bidirectional relationships between objectives and strategic projects. This integration expands information availability and improves project prioritization through a classification system [82].

Following the decision-making tools approach, another study uses the fuzzy Delphi method to gather expert opinions, ensuring undistorted original ideas. This technique handles judgment imprecision. Subsequently, DEMATEL identifies relationships among criteria and constructs an NRM. Finally, based on the NRM, the ANP method is implemented to account for the interdependence and complex relationships among the decision criteria [46].

Another study presents a model based on BWM and TODIM techniques with IVIF, aimed at identifying and prioritizing strategies for the implementation of Industry 4.0 technologies. This proposal seeks to facilitate companies’ adaptation to digital transformation, illustrating the methodology with a case study [43]. Additionally, an analysis of the adoption of Industry 4.0 technologies in SMEs in India employs the fuzzy complex proportional assessment multicriteria method. This approach allows for determining the most suitable technology based on the specific characteristics of each company [90].

Category (2) of the bibliographic portfolio includes technology selection methods based on optimization and integer programming. An optimization model is proposed for Industry 4.0 project portfolio selection. This model addresses the challenge of accurately estimating savings as stochastic values and incorporates multiple constraints and project interdependencies. It aims to maximize economic success through effective digital transformation, evaluated via a case study on an automotive manufacturer, demonstrating its effectiveness and adaptability [104].

Complementarily, a Bayesian-network-based tool analyzes the impact of Industry 4.0 technologies on manufacturers’ operational performance. This approach selects technologies like 3D printing, Automatic Guided Vehicle, and robotics to optimize performance. The Bayesian network models technology–performance relationships, using machine learning to identify hidden connections and considering indicators like quality, time, flexibility, efficiency, servitization, and sustainability [107].

The studies by [105,106] also address technology selection in Industry 4.0. The first combines the AHP and QFD methods with mixed-integer programming, while the second develops an optimization model for managing technology project portfolios, considering resources, costs, and probabilities of completion and success at different stages.

Similarly, a nonlinear 0–1 optimization model is formulated for electric company R&D portfolio selection. Its multicriteria approach incorporates project evaluations and synergies, using a consistency parameter β to adjust contributions and enhance coherence. This approach overcomes previous model limitations by integrating multiple criteria, project interrelationships, and managerial experience [103].

An innovative technology-selection model uses a multicriteria common weight model within the DEA framework. This approach, with mixed-integer linear programming and dichotomy, ensures that only one technology is evaluated as most efficient. By maximizing minimum efficiency among alternatives, this model also reduces required linear programs, making the process independent of the number of candidates [108].

A significant finding in the portfolio review of category (4) is the creation of strategic procedures tailored to the specific needs of each company. In this context, a model for selecting technologies for big data use is presented, focusing on SMEs. This model, called SSF, is based on a causal diagram for which the acronym stands for strategy, time, analytics, data, and technology. SSF provides a structured layered framework that classifies technologies into groups based on their common characteristics and functionalities, facilitating their selection and application [114].

Furthermore, a qualitative analysis-based strategic approach is proposed for technology selection, providing decision-makers evaluation tools. The model consists of five steps: technological strategy, process analysis, technological alternatives, classification/investment, and implementation [115].

4. Analysis of Articles and Discussions

To analyze keyword clusters and their recurrence in the selection articles, a network map was constructed using VOSviewer software. This bibliometric analysis software identified relationships among prominent keywords and graphically represented connections and emerging thematic groupings in the reviewed literature.

To ensure the network map’s relevance, only keywords that appeared more than five times in the articles were considered, thus reflecting recurring terms within the study field (Figure 4). Visualizing the map revealed distinct keyword clusters, which represent the main research areas and common technology selection approaches. This analysis offers a clearer perspective on the areas that have garnered significant research attention.

Figure 4 reveals a thematic structure organized into four main clusters: yellow, blue, green, and red. These clusters represent interrelated conceptual subdomains within the literature on technology selection in Industry 4.0 contexts.

The green cluster positions MCDM as a central axis of research on technology selection. This approach enables the integration of economic, technical, environmental, and social criteria into robust evaluation models and systems. The other key terms reflect a focus on modeling, technology selection management, process optimization (efficiency, energy), and life cycle assessment, with a strong analytical component. Tools such as Data Envelopment Analysis stand out for their usefulness in benchmarking technical and operational efficiency.

The red cluster groups the most commonly used specific methodologies within the MCDM approach. It primarily includes techniques such as AHP, TOPSIS, and, to a lesser extent, MULTIMOORA. Additionally, relevant terms such as extension and project selection appear, highlighting the practical application of these tools in evaluation and prioritization. The presence of fuzzy logic in this cluster indicates both the capability and the necessity of these methodologies to manage uncertainty or imprecision in decision data. The use of aggregation operators reflects the need to combine various measures or preferences. The terms criteria and performance emphasize the importance of defining and assessing performance metrics in the selection process.

The blue cluster focuses on the procedural, operational, and strategic aspects of the decision-making process in technology selection. This group underscores a systematic approach in which technology selection is directly linked to investment decisions and implementation strategies. In this context, methodological planning and the rigorous justification of decisions are emphasized as critical elements. Within this cluster, ANP emerges as a specific methodology that can complement or serve as an alternative to AHP, particularly in situations involving complex interdependencies. The terms alternatives, selection, and ranking reflect the fundamental stages and key components of the decision-making process for technology selection.

The yellow cluster identifies a thematic line focused on the incorporation of sustainability and renewable energy criteria into technology decision-making processes. The prominent presence of the VIKOR method, along with the term’s sustainability, renewable energy, and ranking, highlights an approach aimed at balancing conflicting environmental, economic, and operational criteria when selecting technologies in advanced production environments.

The system node serves as a semantic bridge between clusters, along with model and ranking, indicating their cross-cutting role in the articulation of methodological frameworks, specific applications, and evaluation criteria.

The analysis reveals a strong connection between MCDM methodologies and the technology-selection process. This link is evident in the broad spectrum of criteria considered, from technical performance to strategic and sustainability-related aspects. This diversity highlights the need for more integrative methodologies. These methods must not only handle multiple factors but also adapt to the dynamic and complex conditions of Industry 4.0. Consequently, developing up-to-date, interdisciplinary, and practice-oriented approaches is crucial to facilitate effective technology adoption.

All keywords exhibit at least ten connections, indicating the coherence and relevance of the reviewed articles. This high level of connectivity reinforces the robustness of the analyzed material and the usefulness of the proposed methodology. Although sustainability is present, it remains peripheral to the core decision-making process, which limits its strategic impact.

Figure 4 shows no direct links between selection methodologies and specific Industry 4.0 technologies like the Internet of Things, artificial intelligence, or blockchain, among others. This lack of explicit co-occurrence implies that proposed solutions are largely generic and insufficiently tailored to the challenges of this advanced industrial environment.

This conceptual disconnect presents significant challenges for the practical applicability of existing models, as they do not always provide sufficiently concrete guidance for real-world decision-making. They often lack specific directives for selecting an appropriate Internet of Things system or identifying the most suitable artificial intelligence platform for a company operating in a particular production context.

Figure 5 represents the density map, where each node indicates the frequency of its occurrences within the set of analyzed studies.

Figure 5 shows the frequency of the analyzed keywords. More intense colors indicate higher density, reflecting the term frequency in reviewed articles. MCDM, TOPSIS, AHP, model, and technology selection are most prominent, suggesting their fundamental role and reaffirming their widespread use in technology selection.

An analysis of the full reading of the articles suggests that technology-selection methodologies based solely on qualitative criteria may not be the most suitable for application across different contexts. These methods rely solely on decision-makers’ experience and perspective, particularly in assigning weights to selection criteria. However, numerical data is essential for objective technology ranking and prioritization. Subjective criteria can lead to debates and discrepancies among decision-makers, especially in purely judgment-based processes. Even a well-structured qualitative model may be limited by the absence of quantitative support.

In this context, the literature has yet to establish a universally dominant model for technology selection and implementation, both in general applications and specifically within Industry 4.0 [110]. Although the analyzed bibliographic portfolio reveals certain trends, particularly concerning the type of MCDM methodologies utilized, selection strategies continue to exhibit significant diversity.

While these methodologies have been enhanced over time to overcome the limitations in previous studies, none has yet proven to be definitively superior or widely implemented across diverse studies and settings. Moreover, the requisites for technology selection and implementation diverge among organizations, resulting in the adoption of varied methods tailored to individual company needs [111]. Consequently, the selection of an appropriate MCDM method is contingent upon both the intended decision-making objective and the rational preferences of decision-makers.

An important gap in the literature is the lack of a strategic approach in most studies on technology selection for Industry 4.0. Most methodologies focus on technical, financial, or operational efficiency aspects, often overlooking their integration with the company’s strategic objectives. As a result, technology decisions are often rendered without due regard for their congruence with business strategy. This misalignment can impede the reinforcement of corporate identity, the consolidation of value propositions, the attainment of growth objectives, and the enhancement of competitive positioning. Consequently, companies may struggle to effectively respond to environmental challenges and opportunities.

Acknowledging this constraint, it becomes imperative to formulate methodologies that incorporate strategic considerations into the decision-making process. Nonetheless, the development of a universally applicable model is rendered impractical by the inherent heterogeneity of organizational contexts and requirements. Therefore, this study advocates for the design of flexible and adaptive methodological frameworks applicable to specific contexts, allowing for the customization of criteria and evaluation mechanisms. Due to the inherent complexity and high cost of transitioning to Industry 4.0, each methodological framework must be meticulously tailored to the specific requirements of individual enterprises, ensuring an optimized implementation aligned with strategic objectives [17].

In this regard, integrating elements from different theoretical and methodological approaches provides a more holistic view, addressing the limitations of single models. A practical solution is to combine relevant aspects from various sources, creating a more adaptable tool, rather than developing entirely new ones. However, this raises challenges like methodological compatibility and validation across scenarios, requiring a thorough analysis before implementation

AHP and TOPSIS, known for structuring complex decisions, are often combined in hybrid MCDM approaches. Notably, the AHP-TOPSIS integration is the most common, appearing in eight analyzed documents. This frequency indicates a strong preference for this combination, as other articles utilize diverse techniques without a defined trend.

It is important to emphasize that this hybrid approach is applied in both fuzzy and non-fuzzy environments. This includes Fuzzy AHP combined with Fuzzy TOPSIS, as well as standard AHP with TOPSIS, providing increased flexibility and accuracy, especially when dealing with data uncertainty. This adaptability addresses the need for methodologies that can adjust to diverse contexts and business goals.

In this context, hybrid MCDM methods have garnered significant attention due to their capacity to integrate complementary strengths. However, substantial disparities persist in terms of approach, complexity, and outcomes. The review identified that AHP-VIKOR hybrid methodologies prioritize criteria using AHP and subsequently apply VIKOR to derive a compromised solution based on the distance to the ideal alternative. This approach was effective for balancing conflicting alternatives, although its sensitivity to normalization may undermine the robustness of the results.

Similarly, AHP-CODAS prioritizes criteria through AHP but employs CODAS to assess the deviation between alternatives and the negative ideal solution. This combination generates more discriminative results, although it relies heavily on the consistency of the initial judgments.

TOPSIS-BWM, in contrast, combines the efficiency of TOPSIS with the simplicity of the BWM method for weight assignment. This integration enhances consistency compared to traditional AHP but necessitates well-structured and reliable data, potentially limiting its applicability in contexts characterized by high uncertainty.

The TOPSIS-AHP combination, with or without fuzzy logic, is valued for its applicability and ease of implementation. However, it exhibits sensitivity to data scaling and normalization, and the subjective weighting in AHP can significantly influence the final decision.

Despite their increasing adoption, the literature currently lacks systematic analyses that directly compare these hybrid methods in problems with similar design requirements. This gap makes it challenging to determine whether they lead to convergent or divergent decisions. Such a lack of a critical comparison represents a significant deficiency, as methodological differences may impact the quality, interpretability, and feasibility of technology-related decisions.

The years 2023 (15 publications), 2022 (9), 2021 (7), and 2020 (6) saw the most contributions, indicating a growing, though still limited, interest in methodological approaches for technology-selection decisions, especially within Industry 4.0.

Regarding the quality of the publications based on their quartile distribution, 39% originate from high-impact (Q1) journals, with the remaining 19% spread across Q2 (11%), Q3 (6%), and Q4 (2%). Further, 29% are conference papers, 4% are book chapters, 4% utilize a city score indicator, 4% are referenced studies without defined quality indicators, and 1% are technical reports. While a relevant proportion of the studies comes from high-impact journals, a considerable part of the reviewed literature consists of publications not indexed in databases with impact factors. This may indicate that the topic is continuously evolving and that the most recent findings have not yet been consolidated in high-impact journals. Alternatively, it could suggest that the academic community values different types of publications beyond traditional quartile classifications, although this is less likely.

The distribution of study types reveals that 66% are research articles, 23% are case studies, 6% are literature reviews, 4% are books chapters, and 1% are surveys. This distribution demonstrates a strong preference for empirical and applied research, with a notable emphasis on validating methodologies through case studies. The relatively low representation of literature reviews and surveys suggests an opportunity for further studies that consolidate existing knowledge and delve into the qualitative perspectives of the phenomenon under study.

The main databases contributing to the collection of studies were ScienceDirect with 22 studies, IEEE with 16, Springer with 9, and MDPI with 8, among others. Geographically, Turkey had the highest representation of first author research institutions (12 publications), followed by Iran and China (8 each), and India (7). Expert Systems with Applications stood out as the journal with the largest number of methodological contributions (3 studies).

The design of manufacturing systems within the context of Industry 4.0 exhibits significant differences in various aspects compared to traditional product-design approaches, particularly regarding the methodologies employed for technology selection. The systematic analysis conducted reveals that these differences, especially in areas such as trade-off analysis, economic evaluation, and sensitivity analysis, do not stem from the authors’ assumptions but rather reflect concrete gaps in the existing specialized literature.

Most of the studies reviewed focus on the final stages of the selection process, such as the evaluation of alternatives (80% of the reviewed studies) and their prioritization (70% of the reviewed studies), primarily through multi-criteria decision-making methods such as AHP, TOPSIS, or VIKOR. However, only 47.5% of the studies justify weight assignments empirically, and merely 37.5% include a sensitivity analysis. This trend contrasts with more comprehensive approaches in product design, where such analyses are fundamental for informed and robust decision-making.

The omission of a sensitivity analysis may lead to an underestimation of the risks associated with variability in implementation and operational costs, potentially resulting in unrealistic economic expectations and less efficient technology adoption. To address these gaps, it is recommended to adopt more comprehensive methodologies that explicitly integrate technical and economic criteria. A viable alternative to address these gaps is the development of decision-making frameworks that systematically incorporate trade-off and sensitivity analyses.

5. Conclusions

This study presents a literature review following the ProKnow-C methodology for the collection, classification, and analysis of publications on the decision-making process related to technology selection and implementation within organizations, with a particular focus on Industry 4.0.

In accordance with the methodological process and applying predefined exclusion criteria (keywords, database filters, timeframes, language, and search equations), an initial set of 3400 articles was identified. Applying inclusion criteria resulted in a final bibliographic portfolio of 80 studies, published between 2005 and 2024.

The review reveals that most studies concentrate on the final groups of the technology-selection process, particularly on the evaluation of alternatives (80%) and prioritization and selection of the best option (70%). This reflects a methodologically sound approach, focused on the use of multi-criteria tools to structure decision-making. However, significant weaknesses are identified in other groups of the process. Fewer than half of the studies rigorously address the assignment of weights to the criteria (47.5%), which undermines the justification of the decisions made. Additionally, only 37.5% conduct sensitivity analyses, which limits the validation of recommendations in the face of parameter changes or contextual shifts.

There is also insufficient attention paid to the initial groups, such as problem definition (65%) and the identification of candidate technologies (60%), which may reduce the relevance and effectiveness of the proposed solutions. Taken together, these findings suggest that, while a strong technical foundation exists for evaluating and selecting technologies, significant gaps remain in terms of contextual integration, the comprehensive exploration of alternatives, and robustness assessments of the decisions.

Most studies recommend the use of multicriteria methods, spanning traditional approaches, combined, modified, and advanced forms. In general, these studies identify a set of technologies tailored to each company’s needs, applying MCDM methodologies, optimization techniques, or approaches designed for specific cases. These methods enable the establishment of a ranking primarily based on quantitative indicators, facilitating the selection of the best option according to defined criteria, such as weights, priorities, or the most suitable alternative for the company.

An analysis of emerging trends reveals a strong emphasis on hybrid MCDM methods, integrating diverse methodologies. Examples include the combination of AHP and VIKOR [51], AHP y CODAS [95], TOPSIS y BMW [42], and Fuzzy AHP with Fuzzy ANP [47], among others. However, the most prevalent combination is AHP and TOPSIS, in both standard and fuzzy forms, which accounts for 51.25% of the reviewed literature, either individually or as a hybrid.

This study represents one of the first efforts to compare MCDM methodologies applied to technology selection in Industry 4.0 environments. In this domain, systematic analyses of hybrid approaches remain scarce, particularly in the context of manufacturing system design. Although some studies propose conceptual frameworks, advancements, applications, and future directions [123,124], most available methodological comparisons focus on other domains. These include forecasting [125], project classification [126], supply chain segments [127], educational models [128], or specific industries such as mining [129] and construction [130], among others.

The thematic dispersion observed in the literature makes it difficult to identify optimal MCDM methodological combinations based on the nature of the problem or the specific requirements of the decision-making environment. While numerous applications across various sectors have been documented, there is a lack of a conceptual framework for systematically comparing the performance of these methods under similar design conditions. This gap limits the ability to analyze how a given methodological approach performs when applied to comparable problems. The literature shows that different method combinations can produce significantly divergent results, even when using similar datasets. This highlights the need to establish homogeneous comparison parameters to critically assess how methodological choices affect the robustness of results, their sensitivity to input variations, and their alignment with the strategic objectives of the production system.

Future Research Directions

To guide future research on technology selection, especially in the Industry 4.0 context, the following study areas are proposed:

• Alignment with the company’s overall strategy: Emphasize a strategic approach, focusing on emerging Industry 4.0 technologies and their alignment with business development, innovation, and value creation.
• Long-term integration: Align technology selection with strategic objectives and operational enhancement, ensuring seamless integration with production planning, supply chain management, and product/service development
• Sustainable development: Integrate sustainability criteria into technology selection, including the lifecycle analysis, environmental impact, resource efficiency, and carbon footprint. Promote a circular economy to optimize material use and minimize waste, ensuring responsible technology investments. This balances economic, social, and environmental factors for sustainable technology adoption.
• Flexible methodologies: Develop adaptable and more robust methodologies that may or may not rely on MCDM, considering organizational characteristics such as the size, sector, culture, and resources, while potentially integrating a scenario analysis, simulation, and artificial intelligence.
• Structured comparison: Design a homogeneous conceptual framework to evaluate the performance of hybrid MCDM methods in comparable contexts, considering their robustness, sensitivity, and alignment with strategic manufacturing objectives.

Finally, there is a clear commitment within the scientific and technological community to advancing technology selection, particularly for Industry 4.0. The development of new approaches should focus on evolving multicriteria methodologies, exploring hybrid models, and creating flexible strategies aligned with strategic objectives. These efforts will enhance technology selection and foster sustainable growth in an increasingly competitive and digitalized environment.