Multi-criteria decision-making for selecting occupational health and safety training methods: An Entropy AHP

Abstract

Background

Occupational Health and Safety (OHS) training plays a central role in promoting safety culture and risk awareness among university students. While face-to-face education has traditionally been preferred, online and hybrid models have gained prominence. However, there is limited research that systematically compares these delivery methods using structured decision-making models.

Objective

This study aims to develop a quantitative decision-support framework to evaluate and rank face-to-face, online, and hybrid OHS basic training methods using an integrated Entropy AHP–VIKOR multi-criteria decision-making approach.

Methods

Three training alternatives were evaluated based on eight pedagogical, technical, and administrative criteria. Criterion weights were determined objectively using the Entropy-based Analytic Hierarchy Process (AHP), and the VIKOR method was applied to identify the compromise solution among alternatives.

Results

The hybrid training model achieved the lowest S, R, and Q values in the VIKOR analysis, ranking first among alternatives. Pedagogical impact, participation, and retention emerged as the most influential criteria according to Entropy AHP weighting results.

Conclusions

The integrated Entropy AHP–VIKOR model provides a transparent and objective framework for selecting OHS training methods. The findings support prioritizing hybrid delivery for university-level OHS education while reserving face-to-face components for practice-oriented modules and online components for theoretical content.

Keywords

entropy AHP OHS VIKOR MCDM OHS training hybrid education

Introduction

Education is widely recognized as a central driver of social and economic development. It plays a key role in preparing new generations to cope with current and future responsibilities in working life. In the 2030 Agenda for Sustainable Development, the United Nations explicitly emphasized the importance of education and employment in achieving sustainable development.¹ Young people entering the labor market are particularly vulnerable: according to 2022 ILO data, the highest rates of occupational accidents, including fatal cases, are often observed in the early phases of working life.¹ For this reason, the European Union and other international bodies promote the integration of occupational health and safety (OHS) awareness into education systems at all levels and seek to foster a strong prevention culture before individuals enter the workforce.¹

Occupational health and safety is now regarded not only as a legal obligation but also as a cornerstone of sustainable working life.^2,3 The effectiveness of OHS practices in any organization depends on policies implemented at different levels, including administrative, educational, and technical measures.⁴ Among these, OHS training is a key component. Its main purpose is to equip individuals with the knowledge and skills needed to recognize, evaluate, and control workplace hazards. Training programs help participants understand risks, use safe working methods, and comply with legal requirements. In this way, OHS education contributes to reducing occupational diseases, work-related injuries, and associated material and non-material losses.⁵

Although OHS training frameworks differ across countries, understanding the structure of national systems is essential for interpreting institutional training practices. In this context, the Turkish OHS framework provides a clear example of a legally regulated and systematically implemented model.

Across Türkiye's higher education landscape, OHS training is implemented at multiple academic levels, including associate degree programs, bachelor's programs (especially engineering and technical faculties), postgraduate programs, and specialist training for future OHS experts. Distinguishing between these levels is important because training content, depth, and practical requirements may differ significantly across technician, intern, engineer, and expert profiles.

In Türkiye, OHS training is delivered by various institutions such as universities, training units of public organizations, professional bodies, and Joint Occupational Safety and Health Units authorized by the Ministry of Labor and Social Security, within the legal framework of Law No. 6331.⁶ Higher education institutions that prepare future occupational safety experts have a particularly important role in this process. They are expected to help students understand the philosophy of OHS and transform legal regulations into internalized habits and components of workplace culture.^7,8 For this transformation to occur, students need a solid understanding of OHS principles and the related legal framework.

Early exposure to OHS training has been shown to improve university students’ knowledge of general, technical, and health-related occupational safety and health, strengthen risk awareness, and support the development of safe behavior.^9,10 This applies not only to associate degree students but also to bachelor's and engineering students who are required to complete laboratory courses and mandatory internships.

Fundamental OHS training in higher education institutions therefore aims to increase awareness of potential hazards, encourage safe practices, and support the formation of a safety culture among students.^9,10

According to Law No. 6331, employees in low-hazard workplaces must receive at least eight hours of OHS training every three years.⁶ However, this periodic structure may not fully meet the needs of university students, particularly those who participate in compulsory or voluntary internships. Additionally, many engineering and technical faculties in Türkiye require compulsory OHS courses, which often span one or two semesters, to prepare students for laboratory work, field work, and industrial internships.

For these students, receiving timely and practice-oriented OHS training before their first workplace experience is essential for both safety and educational relevance. Although universities are generally classified as low-hazard environments, students often complete internships in workplaces with higher risk levels. For this reason, many institutions prefer to provide more comprehensive basic OHS training (e.g., 16 h) aligned with “very hazardous” categories, in order to achieve broader preparedness.

Traditionally, OHS training has been delivered face-to-face. This method offers high levels of interaction, immediate feedback, and the opportunity for hands-on practice. With the rapid expansion of digital technologies, especially after the COVID-19 pandemic, online training has become a widely used alternative.¹¹ Online methods provide flexibility and wider access but may also suffer from limitations such as reduced attention, lower engagement, and unequal technical infrastructure.^12,13 Hybrid models, which combine face-to-face and online components, have therefore attracted increasing interest as a way to balance flexibility with interaction.

Choosing the most appropriate training method is a complex decision that involves multiple factors. These include the content and learning objectives of the course, student characteristics, institutional infrastructure, cost, pedagogical quality, participation, attention, and learning retention.^14,15 Multi-criteria decision-making (MCDM) approaches are well suited to handle such complexity because they allow different qualitative and quantitative criteria to be evaluated simultaneously and transparently.^16,17 Although MCDM methods are frequently used in areas such as equipment selection, risk assessment, and performance evaluation, there is limited research that systematically compares OHS training methods using structured decision-making tools.

In Türkiye, academic interest in OHS education has increased, particularly after the enactment of Law No. 6331.⁶ Studies have examined the effectiveness of OHS courses in universities and vocational schools, the impact of different instructional methods on knowledge and behavior, and the role of training in shaping safety culture.^2,3,18–21 Content analyses of associate degree programs show that curricula are largely shaped by legal requirements but often lack sufficient practical orientation.¹⁸ Basic OHS training typically covers several topic groups such as legislation, personal protective equipment, fire safety, ergonomics, occupational diseases, and exposure to chemical, physical, and mechanical risks.^10,18,21

In this study, the basic education subjects presented in Table 1 refer to the education contents provided to associate and undergraduate students participating in laboratory practices and workplace internships, especially in higher education institutions.

Table 1.

Main topics of basic occupational health and safety training.

Topic group	Subtopics
1) General Topics	a) Information about labor legislation b) Legal rights and responsibilities of employees c) Workplace cleanliness and organization d) Legal consequences of occupational accidents and diseases
2) Health Topics	a) Causes of occupational diseases b) Principles and techniques for disease prevention c) Biological and psychosocial risk factors d) First aid
3) Technical Topics	a) Chemical, physical, and ergonomic risk factors b) Manual lifting and carrying c) Prevention of explosions, fires, and flammable materials d) Safe use of work equipment e) Working with display screen equipment f) Electricity: hazards, risks, and precautions g) Causes of occupational accidents and prevention principles h) Occupational health and safety signs i) Use of personal protective equipment j) General OHS rules and safety culture k) Evacuation and rescue

The main topic groups of basic OHS training are summarized in Table 1.

Recent studies have also focused on the impact of online and hybrid education on student participation, satisfaction, and learning outcomes. Findings indicate that while digital tools can support access and flexibility, they may be less effective when participation is passive or content is not interactive.^22–24 Research on OHS courses in higher education underlines the importance of course quality, instructor competence, and alignment between training content and students’ future professional contexts.^25–30 There is also evidence that OHS training benefits not only students but also workers and teachers by improving awareness and contributing to safer behaviors.^31–34 However, meta-analytic findings suggest that while training generally increases knowledge, behavioral change is more difficult to achieve and requires more comprehensive approaches.³⁵

Frameworks such as ISO 21001:2018 and ISO/IEC 40180:2017 provide quality guidelines for educational organizations and e-learning environments, emphasizing learner-centered design, accessibility, and continuous improvement.^36,37 Integrating these perspectives into OHS training supports a more systematic evaluation of training quality and outcomes.

Despite the growing body of literature on OHS education in Türkiye and internationally, most studies remain descriptive. They often examine knowledge levels, attitudes, or safety culture but do not provide a formal decision model for selecting among alternative training methods.^18,22–24^,35 At this point, the integration of MCDM approaches such as Entropy-based Analytic Hierarchy Process (Entropy AHP) and VIKOR into OHS training offers the potential to address an important gap in the literature.

Accordingly, the present study focuses on developing a structured analytical framework rather than presenting procedural details. The objective is to position the research within the broader academic context and highlight the methodological contribution.

This study addresses that gap by proposing a comprehensive, quantitative framework for evaluating three basic OHS training methods, face-to-face, online, and hybrid, based on multiple pedagogical, technical, and administrative criteria. The Entropy AHP method is used to determine objective criterion weights, and the VIKOR method is applied to identify the most appropriate compromise solution among the alternatives. Through this integrated MCDM framework, the study aims to (i) present a structured decision-support model for evaluating OHS training methods; (ii) provide a comparative assessment of training methods across criteria such as cost, accessibility, pedagogical impact, participation, and retention; and (iii) support data-driven policy and curriculum development to improve the quality and effectiveness of OHS training at the institutional level.

Materials and methods

In this study, the implementation methods of occupational health and safety (OHS) basic training at the university level, namely face-to-face, online, and hybrid approaches, are evaluated using a multi-criteria decision-making (MCDM) approach. In this context, the term ‘university level’ refers specifically to associate and bachelor's degree students who are required to complete basic OHS training prior to laboratory work or workplace internships. The main objective of the study is to identify the most appropriate training method based on specific pedagogical, technical, and administrative criteria. For this purpose, the criterion weights were objectively determined using the Entropy AHP method, and subsequently, the alternatives were ranked according to these weights using the VIKOR method.

The overall methodological workflow applied in this study is summarized in Figure 1, illustrating the sequential steps followed throughout the Entropy AHP–VIKOR evaluation process.

Figure 1.

Workflow of the Entropy AHP-VIKOR evaluation process.

The VIKOR (VlseKriterijumska Optimizacija I Kompromisno Resenje) method, developed by Opricovic and Tzeng (2004), is a multi-criteria decision-making technique designed to determine a compromise solution among conflicting criteria. Unlike traditional ranking methods that aim to identify an absolute best alternative, VIKOR focuses on finding a solution that represents the closest approximation to the ideal while considering both group utility and individual regret.³⁸

The method begins with the construction of a normalized decision matrix and identification of the best (f*) and worst (f−) values for each criterion. Then, for each alternative, two measures are computed:

the utility measure (S_i), which represents the overall group satisfaction based on the weighted sum of normalized distances from the ideal solution; and

the regret measure (R_i), which reflects the maximum individual regret observed for any criterion.

After calculating S_i and R_i, the VIKOR index (Q_i) is computed to balance these two measures according to the following compromise formula:

Q_{i} = v \frac{S_{i} - S *}{S^{-} - S *} + (1 - v) \frac{R_{i} - R *}{R^{-} - R *}

(1)

S_i: Utility measure (group utility) for alternative i.

S*: Minimum (best) utility value among all alternatives.

S⁻: Maximum (worst) utility value among all alternatives.

R_i: Regret measure for alternative i.

R*: Minimum (best) regret value among all alternatives.

R⁻: Maximum (worst) regret value among all alternatives.

v: Weight of group utility (commonly set to 0.5).

The VIKOR method is particularly advantageous in educational and managerial decision-making contexts where trade-offs exist between multiple qualitative and quantitative factors. In this study, it allows for a balanced assessment of OHS training methods by jointly considering pedagogical, technical, and cost-related dimensions, thereby identifying the most appropriate alternative under realistic compromise conditions.

Definition of the decision problem

The effective planning of occupational health and safety (OHS) training necessitates a strategic decision-making process, encompassing not only the content of the training but also the delivery format (face-to-face, online, or hybrid).^39,40 Particularly in OHS basic training at the university level, factors such as student profiles, technical infrastructure, instructional resources, and time management directly influence this process.^41–43 The choice of training method should be adapted to students’ knowledge levels, learning styles, and attention spans, while also aligning with institutional resources, cost, and accessibility.^41,44,45 In this context, selecting the most appropriate method among face-to-face, online, and hybrid models in OHS training should be approached as a multidimensional decision-making problem.^46–48 The primary aim of this study is to determine the most suitable training method among these alternatives based on specific criteria and to provide decision-makers with a systematic roadmap.

In defining the decision problem, the evaluation of training methods was based on a typical university-level student profile with access to basic digital infrastructure (internet connection and personal device), average self-regulation skills, and prior exposure to introductory OHS concepts. These assumptions were considered to ensure consistency across alternatives, as the performance of face-to-face, online, and hybrid methods may vary depending on learners’ technical resources, autonomy, and educational level.

This study focuses specifically on basic OHS training delivered to associate and bachelor's degree students in higher education institutions. Although the evaluation framework was developed based on university-level training scenarios, the multi-criteria structure of the model allows adaptation to other educational contexts such as vocational schools, technical certification programs, and workplace-based OHS training. However, applicability to these contexts may require adjusting the criteria weights according to learner characteristics and institutional conditions.

It is important to note that this study does not involve empirical measurement of student performance. Instead, the evaluation was carried out through expert judgment within a scenario-based MCDM framework. The decision matrix reflects expert assessments of how each training method would be expected to perform under typical institutional conditions, rather than data collected from actual learners.

Alternatives

In this study, the alternatives evaluated in the decision-making process consist of the three most commonly applied basic OHS training methods at universities in Türkiye:

A1: Face-to-Face Training – A traditional classroom-based method delivered directly by the instructor. It provides a high level of interaction but is limited by time and location constraints.

A2: Online Training – A digital training method conducted remotely using synchronous (live sessions) or asynchronous (videos, presentations, e-books) tools. While it offers accessibility and flexibility, it may also involve challenges such as distraction and low engagement.

A3: Hybrid Training – A blended model that incorporates both face-to-face and online components. It aims to combine the advantages of both approaches but requires greater technical coordination and planning.

A report published in 2021 indicated that online occupational health and safety training, implemented subsequent to the emergence of the novel Coronavirus, will contend with face-to-face training in the long term.⁴⁹ Digital training applications are also recommended for occupational health and safety training.⁵⁰

These three alternatives were analyzed and compared using multi-criteria decision-making (MCDM) models based on the selected evaluation criteria.

Post-training assessment mechanisms differ substantially across the three training modalities and may influence learner engagement and learning outcomes. In face-to-face training, instructors can directly verify participation, observe attention, and conduct immediate performance-based assessments. In contrast, online training, especially asynchronous delivery, relies heavily on system-logged interactions, self-reported participation, or automated quizzes, which may not fully capture actual learner engagement or comprehension. Hybrid models typically combine both types of evaluation mechanisms. Although the present study does not empirically measure post-training outcomes, these differences in assessment and verification structures were taken into conceptual consideration when defining the criteria related to participation, attention, and measurability. Acknowledging this variation is essential for ensuring a fair comparison of training methods within the MCDM framework.

Evaluation criteria

The criteria used for evaluating the alternatives encompass multidimensional measures that are critical for the pedagogical quality, applicability, and sustainability of occupational health and safety (OHS) training programs. To ensure international comparability and quality consistency, these criteria were aligned with the ISO/IEC 40180:2017 e-learning Quality Reference Framework (QRF), which integrates pedagogical, technical, and organizational dimensions for instructional design and evaluation.³⁷ Based on a comprehensive literature review and insights from practitioner experiences, eight evaluation criteria were identified and are summarized below:

C1 – Accessibility: The ability of the training method to reach a larger number of students regardless of geographical or technical limitations.

Digital and blended (hybrid) teaching in higher education enables access to a wider student population by providing scalability and flexibility. Numerous studies have reported that online/hybrid systems produce outcomes at least equivalent to face-to-face teaching (and in most cases better), particularly in clinical skills training, and can also be cost-effective.^51,52

In accordance with international accessibility frameworks, this criterion is consistent with ISO/IEC 24751, which defines individualised adaptability and accessibility in e-learning environments. This promotes inclusive, flexible, and personalised learning for diverse user needs.^53,54

C2 – Pedagogical Impact: The influence of the training on learning outcomes, including comprehension, retention, and behavioral transformation.

It has been asserted that the integration of hybrid and digital teaching methodologies has yielded favorable outcomes in terms of learning outcomes, skill acquisition, and student feedback within the short and medium term.^45,51

C3 – Participation and Attention: The level of active student participation, attention span, and engagement during the training.

Online environments have been shown to increase flexibility and student satisfaction. However, they can also cause fluctuations in participation due to distractions and the need for self-regulation. Therefore, design quality, interaction, and Self-Regulated Learning (SRL) support are highlighted as decisive factors in participation/attention criteria.^51,52,55

C4 – Technological Requirements: The necessary technical infrastructure such as hardware, software, and internet connectivity for implementing the training method.

It is imperative to acknowledge that hybrid/HyFlex and large-scale online applications are prone to suboptimal functionality in the absence of a robust infrastructure, encompassing hardware, software, and connectivity. The efficacy of these applications is further contingent on the presence of educator/student digital literacy and institutional support. According to the extant literature, the primary obstacles are infrastructure deficiencies and inadequate teaching design and teacher training.^42,56

C5 – Cost: The direct and indirect expenditures associated with implementing the training, including personnel needs and time cost.

In the context of occupational health and safety training, e-learning platforms have been demonstrated to be a cost-effective and efficient method of knowledge and skill acquisition, facilitating progress tracking at the corporate level.⁵⁷

C6 – Retention: The training's effectiveness in supporting long-term learning and memory recall.

The efficacy of digital and hybrid teaching methods in enhancing not only examination outcomes but also knowledge retention has been a subject of considerable interest.⁵²

C7 – Measurability: The ease with which training outcomes can be tracked, assessed, and quantified.

E-learning platforms in the domain of Occupational Health and Safety (OHS) offer direct administrative and measurable outputs, including completion and progress tracking.⁵⁷

C8 – Motivation: The degree to which students show interest, willingness, and voluntary engagement in the training.

Flipped/blended learning applications have been observed to enhance participation and motivate students in higher education contexts. However, the available research has demonstrated a notable correlation between learning motivation, emotional engagement, and academic performance in blended learning environments.^58,59

Based on these criteria, the training methods were compared, and the weighting process was carried out using the Entropy-based Analytic Hierarchy Process (AHP) method.

Determining criteria weights with Entropy AHP method

In multi-criteria decision-making (MCDM) problems, determining the relative importance of the criteria is a critical step that directly affects the outputs of the decision model. Although the traditional Analytic Hierarchy Process (AHP) method is widely used for this purpose, it relies on subjective expert judgments, making it prone to biases and inconsistencies.^16,60,61 Therefore, in this study, the Entropy-based AHP method fully data-driven approach independent of decision-maker opinions, was preferred.

Entropy, a concept from information theory, measures the degree of disorder or uncertainty in a system and allows criteria to be weighted based on their discriminating power within the decision matrix.^62,63 In other words, a criterion that exhibits high variation across decision alternatives carries more informational value for the decision process and is thus assigned a higher weight. Conversely, a criterion with low variance is assumed to contribute less to the decision process and receives a lower weight.

In the context of this study, the application of the Entropy-AHP method was carried out through the following mathematical steps:

Step 1: Construction of the Decision Matrix (X)

Initially, the alternatives (face-to-face, online, and hybrid training methods) are evaluated based on the predefined criteria (C1–C8). This evaluation is based on scores obtained from the literature, pedagogical analyses, and expert opinions. As a result, a decision matrix of size m × n is constructed,^64,65 where m represents the number of alternatives and n denotes the number of criteria.

X = [x_{i j}] (i = 1, . . ., m; j = 1, . . ., n)

(2)

Step 2: Construction of the Normalized Decision Matrix (P)

Before performing entropy calculations, the data in the decision matrix must be normalized. The normalized matrix P is obtained by dividing each cell value by the sum of the values in the same column,^38,43 as follows:

P_{i j} = \frac{x_{i j}}{\sum_{i = 1}^{m} x_{i j}}

(3)

This step provides a comparable distribution for each criterion.

Step 3: Calculation of the Entropy Value ( $E_{j})$

For each criterion, the entropy value $E_{j}$ is calculated based on information theory principles using the following formula:^62–64

E_{j} = - k . \sum_{i = 1}^{m} P_{i j} . \ln (P_{i j}) w h e r e k = \frac{1}{l n (m)}

(4)

In this formula, for cases where $P_{i j} = 0$ , it is assumed that $P_{i j}$ . ln( $P_{i j})$ = 0.

The entropy value is inversely proportional to the information uncertainty of the criterion. That is, as the value of $E_{j}$ approaches 1, the discriminatory power of the criterion decreases; as it approaches zero, the criterion becomes more distinctive.

Step 4: Calculation of the Information Utility Value ( $d_{j})$

Based on the entropy value, the information utility value for each criterion is calculated as follows:

d_{j} = 1 - E_{j}

(5)

This value represents the contribution of each criterion to the decision-making process. In other words, the larger the value of $d_{j}$ , the greater the impact of that criterion on the decision.

Step 5: Calculation of the Criterion Weights ( $w_{j})$

In the final step, the obtained information utility values are normalized to determine the criterion weights:

w_{j} = \frac{d_{j}}{\sum_{j = 1}^{n} d_{j}}

(6)

As a result of this process, the relative weights of the criteria are normalized to fall between 0 and 1, and their total equals 1. These weights will be used in the subsequent VIKOR analysis to rank the alternatives.^60,61,66

In this respect, the Entropy AHP method offers a criterion weighting technique based entirely on numerical data, independent of the subjective judgments of decision-makers, making it a robust alternative especially for desk-based theoretical studies. In this study, the weights obtained through Entropy AHP have enabled the objective comparison of the performance of different training methods.

Decision matrix and Entropy AHP criteration

Table 2 presents the theoretical scores assigned to the three OHS training methods (A1–A3) based on the eight evaluation criteria (C1–C8). The scores range from 1 (low suitability) to 9 (high suitability) and reflect expert group assessments under typical university-level implementation conditions. Following the construction of the decision matrix, the values were normalized and the Entropy AHP procedure was applied to calculate the entropy values, divergence measures, and the corresponding criterion weights. The resulting criterion weights are presented in Table 3.

Table 2.

Decision matrix for training methods (1–9 scale).

Criteria	Face-to-Face (A1)	Online (A2)	Hybrid (A3)
C1 – Accessibility	4	9	7
C2 – Pedagogical Effectiveness	8	5	9
C3 – Participation/Attention	7	4	8
C4 – Technological Requirements	3	6	6
C5 – Cost	4	7	6
C6 – Retention	7	5	8
C7 – Measurability	6	8	7
C8 – Motivation	7	5	8

Table 3.

Criterion weights calculated by the entropy AHP method.

Criteria	Weight (wj)
C1 – Accessibility	0.116
C2 – Pedagogical Effectiveness	0.162
C3 – Participation/Attention	0.145
C4 – Technological Requirements	0.082
C5 – Cost	0.090
C6 – Retention	0.143
C7 – Measurability	0.131
C8 – Motivation	0.131

Results

According to the Entropy AHP results (Table 3), the criterion with the highest weight is “Pedagogical Impact” (C2), making it the most decisive factor in the decision-making process. It is followed by “Participation and Attention” (C3), “Retention” (C6), and “Motivation” (C8), respectively. This section presents the numerical findings of the integrated Entropy AHP–VIKOR evaluation. The criterion weights obtained from Entropy AHP were used as inputs in the VIKOR method to rank the training alternatives.

Ranking of alternatives using the VIKOR method

The criterion weights obtained from the Entropy AHP method (Table 3) were directly incorporated into the VIKOR analysis. For each training alternative, the relative deviation from the ideal value was first calculated for every criterion. These deviations were then multiplied by the corresponding Entropy-based weights, and the resulting weighted values were summed to obtain the S_i (group utility) score. The R_i (individual regret) score was determined by selecting the maximum weighted deviation for each alternative, representing its weakest performance area. Finally, the Q_i compromise index was computed by normalizing and combining the S_i and R_i values. Through this integrated procedure, the ranking presented in Table 4 fully reflects and operationalizes the objective criterion weights reported in Table 3, ensuring a transparent and methodologically consistent connection between the Entropy AHP and VIKOR approaches.

Table 4.

VIKOR method results.

Alternatives	Si	Ri	Qi	Ranking
A1 – Face-to-Face	0.431	0.163	0.585	2
A2 – Online	0.586	0.249	1.000	3
A3 – Hybrid	0.000	0.000	0.000	1

In the VIKOR calculations, the compromise parameter was set to v = 0.5, reflecting equal importance for group utility and individual regret. The resulting S_i, R_i, and Q_i values obtained through this procedure are presented in Table 4.

Discussion

The present study proposes a structured, multi-criteria, and consensus-based framework to evaluate alternative delivery methods for Occupational Health and Safety (OHS) basic training by integrating the Entropy–AHP and VIKOR methods. This approach provides an objective and transparent mechanism through which pedagogical, technical, and administrative criteria can be jointly assessed.

Interpretation of findings

The findings of the VIKOR analysis indicate that the hybrid (A3) training model represents the most balanced alternative, achieving the lowest S, R, and Q values. This result reflects the hybrid model's capacity to combine the pedagogical strengths of face-to-face instruction with the accessibility advantages of online delivery. Although face-to-face education (A1) performed well in pedagogical dimensions, it ranked second due to disadvantages in criteria such as accessibility and cost. Conversely, online education (A2), despite high accessibility, scored lowest because it yielded weaker performance in attention, pedagogical impact, and retention-related criteria.

These results demonstrate that achieving optimal outcomes in OHS basic training, a practice-oriented field that targets awareness, behavioral competence, and risk perception, is difficult through a single instructional method. The hybrid model addresses this challenge by enabling theoretical components to be delivered flexibly through online modules, while practical applications, peer interaction, and motivation-enhancing activities are effectively carried out in face-to-face environments.

The hybrid method also offers structural advantages across multiple criteria. It supports pedagogical depth through in-person engagement (C2), maintains student attentiveness through structured face-to-face activities (C3), and provides stable technological conditions for tasks requiring digital infrastructure (C4). These combined strengths explain why the hybrid model consistently outperformed the other alternatives across the eight evaluation criteria.

In practical components of OHS basic training, certain modules inherently require face-to-face delivery due to the need for hands-on practice, instructor supervision, and real-time feedback. These include (i) fire safety and extinguisher use, (ii) basic first-aid applications, (iii) laboratory safety procedures, (iv) hazard identification and on-site risk assessment, and (v) ergonomics training involving posture and manual handling demonstrations. Because these modules contain psychomotor and behavioral learning outcomes, online-only delivery is insufficient. Accordingly, the hybrid structure, where theoretical content is provided online and skill-based modules are conducted face-to-face, offers a pedagogically coherent approach.

Comparison with literature

The present findings align closely with prior research applying multi-criteria decision-making models to educational and OHS-related problems. For example, Demir et al. (2025) employed FWENSLO–FARTASI to identify the most effective safety interventions, while Kalem & Akpınar (2022) used Entropy–MABAC to reduce subjectivity in personnel evaluations.^61,66 Similar to these studies, the integrated Entropy AHP–VIKOR approach in this research enabled the systematic evaluation of alternatives across multiple dimensions.

Our results indicating that the hybrid model provides superior accessibility (C1) reflect observations by Wang et al. (2024) and McGee et al. (2024), who reported that hybrid tools enhance learner accessibility and scheduling flexibility.^43,52 From a pedagogical standpoint, the strong performance of hybrid education in C2 is consistent with findings by Berga et al. (2021) and others who noted enhanced pedagogical engagement, interaction, and flexibility in hybrid settings.^39,67

The hybrid model's advantage in participation and attention (C3) is supported by Ben Khalifa et al. (2025) and Wang & Raman (2025), who found that hybrid structures strengthen focus and engagement compared to fully online instruction.^68,69 Technological requirements (C4) were also better addressed through hybrid models, aligning with studies noting that blended environments mitigate digital divide issues and support infrastructure-dependent tasks.^50,70

Cost-related findings (C5) likewise reflect the literature, which shows that hybrid education can reduce operational and student-related expenditures while maintaining quality.^39,52 The hybrid model's superiority in retention (C6) is consistent with research suggesting that integrating online theory with in-person practice enhances long-term memory of procedural and safety-related tasks.⁵²

In terms of measurability (C7), our findings agree with Rehman et al. (2025) and Hazmidar et al. (2025), who demonstrated that hybrid systems allow improved tracking of learning behaviors, video engagement metrics, and safe behavior observations.^71,72 Similarly, the strong performance of the hybrid model in motivation (C8) parallels studies reporting positive effects of hybrid environments on intrinsic motivation and learner satisfaction.^71,73

These convergences indicate that the hybrid education model not only fits within theoretical expectations but also shows practical relevance across diverse learning contexts.

Practical implications

The results of this study provide actionable guidance for universities, OHS coordinators, and public institutions responsible for designing OHS training programs. The multidimensional evaluation highlights which instructional components benefit most from online, face-to-face, or hybrid delivery. Institutions can use the model to align training strategies with learner profiles, infrastructure conditions, and module requirements. Integrating this decision-support framework into routine OHS planning may enhance training effectiveness, optimize resource allocation, and promote sustainable learning outcomes.

Implications for future research

This study is based on a desk-based, simulation-oriented evaluation model rather than empirical measurements of student learning outcomes. Although the hybrid model appears optimal within the constructed decision matrix, real-world validation involving test scores, behavioral observations, and longitudinal retention data is necessary to confirm the model's predictive robustness. Future research should incorporate experimental designs, student performance analytics, and field studies to refine the weighting structure and enhance the generalizability of the findings.

Limitations

Although this study provides a robust methodological framework for selecting OHS training methods, it has certain limitations. First, the decision matrix is based on expert group evaluations supported by pedagogical literature rather than empirical measurements collected directly from students or trainees. While the Entropy-based AHP method minimizes subjective bias in weighting criteria, the initial scoring of alternatives still reflects expert judgment. Second, the study specifically examines basic OHS training at the university level in the Turkish higher education context; therefore, the findings may require adaptation for different sectors (e.g., construction, mining) or cultural settings with varying institutional and technological capacities. Finally, due to the desk-based nature of the research, the evaluation remains theoretical. Future studies should incorporate longitudinal empirical data to validate the model by assessing actual learning outcomes, behavioral change, and incident-reduction impacts across different training methods.

Conclusion

The present study employed an integrated Entropy AHP–VIKOR approach to present a structured, data-driven evaluation of three fundamental OHS training delivery methods, face-to-face, online, and hybrid, by examining the relative merits of each method. The findings indicate that hybrid training provides the most balanced and effective solution across pedagogical, technical, and administrative dimensions. While face-to-face training has been shown to promote interaction and behavioral engagement, online training offers accessibility and flexibility. However, the hybrid model effectively combines the strengths of both formats.

The Entropy AHP method enabled objective weighting of criteria such as pedagogical impact, participation, cost, and measurability, while the VIKOR analysis identified the hybrid model as the closest alternative to the ideal solution. The findings of this study indicate that decision-makers should prioritize the implementation of hybrid training models, particularly within the context of university-level OHS education, where the simultaneous consideration of flexibility, engagement, and retention is imperative.

Moreover, the methodological framework proposed in this study provides a replicable decision-support tool for education planners and OHS professionals. The model can be adapted beyond the current study's scope to inform training strategies in vocational education, public institutions, and corporate environments. The integration of MCDM techniques, such as Entropy AHP and VIKOR, contributes to the development of OHS training programs that are more evidence-based, efficient, and context-sensitive.

Footnotes

Acknowledgments

The authors would like to thank all individuals and institutions who contributed to this study. An AI-assisted tool was used to support English language editing and the generation of a representative workflow figure. All scientific content, analyses, and interpretations are solely the responsibility of the authors.

Ethical approval and informed consent

This study does not involve human participants, personal data, or experimental procedures. Therefore, ethical approval and informed consent were not required.

Author contributions

Conceptualization (Samet TOSUN, Ömer Faruk ALACAHAN), Data curation (Samet TOSUN), Formal Analysis (Samet TOSUN, Ömer Faruk ALACAHAN), Funding Acquisition (Samet TOSUN), Investigation (Samet TOSUN, Ömer Faruk ALACAHAN), Methodology (Samet TOSUN), Validation (Samet TOSUN, Ömer Faruk ALACAHAN), Visualization (Samet TOSUN, Ömer Faruk ALACAHAN), Writing – original draft (Samet TOSUN, Ömer Faruk ALACAHAN), Writing – review & editing (Samet TOSUN, Ömer Faruk ALACAHAN)

Funding

The authors received no financial support for the research, authorship, and/or publication of this article.

Declaration of conflicting interests

The authors declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Data availability

Not applicable.

ORCID iDs

Samet Tosun

Ömer Faruk Alacahan

References

Bikfalvi

Villar Hoz

Giménez Leal

, et al. Riding the wave of sustainability: integrating OSH into education. Saf Sci 2024; 170: 106370.

Ceylan

. Türkiye’deki İş Sağliği ve Güvenliği Eğitimi Sorunlar ve Çözüm Önerileri. Electron J Vocat Coll 2012; 2(2): 94–104.

Topgül

Alan

. Öğrencilerin İş Güvenliği ve İş Güvenliği Eğitimi Algısının Değerlendirilmesi. Süleyman Demirel Üniversitesi İktisadi ve İdari Bilimler Fakültesi Dergisi 2017; 22: 587–598.

Çağlarer

. A current overview of multi-criteria decision-making methods used in occupational health and safety. Rev Manag Econ Eng 2024; 23: 320–337.

Alqahtani

Algarni

, et al. Development of an educational program for occupational health and safety in Saudi Arabia. J Eng Res 2025; 13: 551–560.

Ministry of Labor and Social Security of the Republic of Türkiye. Occupational Health and Safety Law, https://www.mevzuat.gov.tr/mevzuat?MevzuatNo=6331&MevzuatTur=1&MevzuatTertip=5 (2012, accessed 20 October 2025).

Ozen

Yilmaz

Kartal

. Evaluation of occupational safety courses given in undergraduate departments in Turkey and investigation of the effect on working life. Sustainability 2023; 15: 12140.

Yusuf

Suarbawa

Dewi

, et al. Understanding occupational health and safety regulations and the influence on students’ behavior in practical workshops. KESMAS 20. Epub ahead of print 30 May 2025. DOI: 10.7454/kesmas.v20i2.1628.

Taşkın

Şahin

Deniz

, et al. Evaluation of students’ awareness of occupational health and safety education. Sürekli Tıp Eğitimi Dergisi 2025; 34(2): 140–149.

10.

Liu

Pei

Han

, et al. Critical factors in the implementation of risk awareness education in universities in China. Front Psychol 2022; 13: 803360.

11.

Teki̇n

Suvaydan

, et al. İlkokul Öğrencilerinin İş Sağlığı ve Güvenliği Eğitimi Algısının Değerlendirilmesi. Soma Meslek Yüksekokulu Teknik Bilimler Dergisi 2021; II: 9–18.

12.

Dalyan

Öztürk

ÖF

, et al. İş Sağlığı ve Güvenliğinde Yüz Yüze ve Uzaktan Eğitim Sistemlerinin Karşılaştırılması. Karaelmas İş Sağlığı ve Güvenliği Dergisi 2021; 5: 219–228.

13.

Doğan

Yalçınkaya

Balcı

. Türkiye’de Mühendislik Fakültelerinde İş Sağlığı Ve Güvenliği Eğitimi. Mühendis ve Makina 2017; 58: 1–15.

14.

Olcay

Temur

Sakalli

. A research on the knowledge level and safety culture of students taking occupational health and safety course. CJES 2021; 16: 187–200.

15.

Özgüler

Koca

. MESLEK YÜKSEKOKULLARINDA İŞ SAĞLIĞI VE GÜVENLİĞİ EĞİTİMİNİN GEREKLİLİĞİ. Electron J Vocational Colleges 2013; 3: 15–20.

16.

Saaty

. How to make a decision: the analytic hierarchy process. Eur J Oper Res 1990; 48(1): 9–26.

17.

Yeni̇sarı

Mestav

Öztürk

ÖF

. Üniversite Çalışanlarının İş Sağlığı ve Güvenliği Eğitimi Konusundaki Bilinç Düzeylerinin Araştırılması. Çanakkale Onsekiz Mart Üniversitesi Fen Bilimleri Enstitüsü Dergisi 2019; 5: 339–355.

18.

Tuna

Karaçam

. Ön Lisans İş Sağlığı ve Güvenliği Eğitiminin Mevcut Durumunun Analizine Yönelik Bir Araştırma. Gümüşhane Üniversitesi Sağlık Bilimleri Dergisi 2021; 10: 724–739.

19.

Akboğa Kale

Kayahan

Baş

, et al. 2025 İSG dersinin mühendislik fakültesi öğrencilerinin İSG farkındalığı üzerindeki etkisi Devlet üniversitesi örneği. International Journal of Engineering Approaches 2025; 2: 7–16.

20.

Aydoğan

. Meslek yüksekokulu Öğrencilerinin İş Sağlığı ve Güvenliği Eğitimleri ve Bu Eğitimin Çevre Sağlığı Farkındalığına Etkisi. IAUSBD 2021; 13: 261–284.

21.

Sarıkaya

Güllü

Seyman

. Meslek Yüksek Okullarında İş Sağlığı ve Güvenliği Eğitimi Verilmesinin Önemi (Kırıkkale Meslek Yüksek Okulu Örneği). Türk Bilim Araştırma Vakfı 2009; 2: 327–332.

22.

Nam

Tutar

. İş Sağlığı ve Güvenliği Eğitiminde E-Öğrenmenin Etkinliği. Int J Entrepreneur Manage Inquir 2019; 3: 71–93.

23.

Çeti̇n

Çolak

. İş sağliği ve güvenliği eğitiminin pandemi sürecinde Öğrenciler bakımından analizi. Ergonomi 2023; 6: 178–192.

24.

Durak

Şık

. İş sSağlığı ve Güvenliği Eğitimi Süreçlerinin Eğitim Teknolojisinin Temel Ögeleriyle Betimlenmesi. OHS ACADEMY 2022; 5: 198–207.

25.

Şahmaran

Kar

Arısal

. İş Sağlığı ve Güvenliği Ön Lisans Programında Verilen Eğitim ve Öğretimin İş Sağlığı ve Güvenliği Algısı Üzerine Etkisi. OPUS Uluslararası Toplum Araştırmaları Dergisi. Epub ahead of print 30 June 2019. DOI: 10.26466/opus.565166.

26.

Kar

Şahmaran

. Üniversitede Verilen İş Sağliği ve Güvenliği Dersinin Öğrenciler Üzerindeki Etkisinin Araştırılması. Ejons Int J Math Eng Nat Sci 2021; 5: 587–588.

27.

Şensöğüt

. 2018 İş Güvenliği Kültürü ve Üniversiteler. MCBÜ Soma Meslek Yüksekokulu Teknik Bilimler Dergisi 2018; 1: 9–15.

28.

Yeşil

Sönmez Çalış

. İş Sağlığı ve Güvenliği Eğitiminde Önlisans Programları: türkiye’deki Uygulama ve ABD’deki Uygulama ile Karşılaştırılması. Süleyman Demirel Üniversitesi İktisadi ve İdari Bilimler Fakültesi Dergisi 2016; 21: 1173–1189.

29.

Gür

Yavuz

. Üniversitelerin İş Sağlığı ve Güvenliği Programında Görev Yapan Akademisyenler Üzerine Bir İnceleme. ispecijssh 2020; 4: 328–342.

30.

Kayabaşı

. Özdemir H, Cündübeyoğlu İ. Kamu Üniversitelerinde İş Sağlığı ve Güvenliği Uygulama Rehberi İçerik Analizi. Kayseri Üniversitesi Sosyal Bilimler Dergisi 2021; 3: 29–45.

31.

Ateş

. Çalışanların İş Sağlığı ve Güvenliği Açısından Eğitilmeleri. Selcuk Universitesi Hukuk Fakultesi Dergisi. Epub ahead of print 12 June 2020. DOI: 10.15337/suhfd.738406.

32.

Savcı

Şerbetçi

Kılıç

. Status of health discipline students to receive training on occupational health and safety and exposure to occupational accident. SHYD. Epub ahead of print 2018. DOI: 10.5222/SHYD.2018.036.

33.

Şener

Fandaklı

Yağımlı

, et al. Eğitim Sektöründe İş Sağlığı ve Güvenliği. J Int Scientif Res 2023; 8: 83–94.

34.

Bağcı

Azimli

Uluçay

, et al. Eğitim Öğretim Kurumlarında İş Sağlığı ve Güvenliği Bilincinin İncelenmesi. Int J Soc Human Sci Res 2024; 11: 545–555.

35.

Arıcak

Çağlarer

. Meta analysis method overview of models used in occupational safety trainings. PEARSON 2021; 6: 482–490.

36.

Abu Bakar

Ismail

Mohamad Darwi

. ISO 21001:2018 In academia: a review of literature. IJEMP 2024; 7: 68–82.

37.

Bağcı

Azimli

Uluçay

Keskin

. Eğitim öğretim kurumlarında iş sağlığı ve güvenliği bilincinin incelenmesi. Int J Soc Humanit Sci Res. 2024;11(104):545–555.

38.

Opricovic

Tzeng

G-H

. Compromise solution by MCDM methods: a comparative analysis of VIKOR and TOPSIS. Eur J Oper Res 2004; 156: 445–455.

39.

Berga

K-A

Vadnais

Nelson

, et al. Blended learning versus face-to-face learning in an undergraduate nursing health assessment course: a quasi-experimental study. Nurse Educ Today 2021; 96: 104622.

40.

Schmid

Borokhovski

Bernard

, et al. A meta-analysis of online learning, blended learning, the flipped classroom and classroom instruction for pre-service and in-service teachers. Comput Educ Open 2023; 5: 100142.

41.

Mahdavi Ardestani

Adibi

Golshan

, et al. Factors influencing the effectiveness of E-learning in healthcare: a fuzzy ANP study. Healthcare 2023; 11: 2035.

42.

Sareen

Mandal

. Challenges of blended learning in higher education across global north-south: a systematic and integrative literature review. Soc Sci Human Open 2024; 10: 101011.

43.

Wang

Liu

Jia

, et al. Hybrid teaching after COVID-19: advantages, challenges and optimization strategies. BMC Med Educ 2024; 24: 753.

44.

Kulju

Jarva

Oikarinen

, et al. Educational interventions and their effects on healthcare professionals’ digital competence development: a systematic review. Int J Med Inf 2024; 185: 105396.

45.

Dai

, et al. Does the instructional approach really matter? A comparative study of the impact of online and in-person instruction on learner engagement. Acta Psychol (Amst) 2025; 253: 104772.

46.

Maral

. Examining the research performance of universities with multi-criteria decision-making methods. Sage Open 2024; 14(4): 21582440241300542.

47.

Maral

Özdemir

. A systematic review on multi-criteria decision-making methods in educational research. Br Educ Res J 2025: berj.70002.

48.

Zhiqi

. VIKOR Method for evaluating the management quality of teacher education with probabilistic hesitant fuzzy information. Discrete Dyn Nat Soc 2021; 2021: 1–8.

49.

Sarpy

SAC

Stachowski

Gustafson

, et al. The use of distance learning in occupational health and safety training: Assessing effectiveness and sustainability in the context of the COVID-19 pandemic. USA: CPWR-The Center for Construction Research and Training, https://www.cpwr.com/ (2021).

50.

Vukićević

Mačužić

Djapan

, et al. Digital training and advanced learning in occupational safety and health based on modern and affordable technologies. Sustainability 2021; 13: 13641.

51.

Man

Wen

BCL

. Are virtual reality applications effective for construction safety training and education? A systematic review and meta-analysis. J Saf Res 2024; 88: 230–243.

52.

McGee

Wark

Mwangi

, et al. Digital learning of clinical skills and its impact on medical students’ academic performance: a systematic review. BMC Med Educ 2024; 24: 1477.

53.

Obidat

. Bibliometric analysis of global scientific literature on the accessibility of an integrated E-learning model for students with disabilities. Cont Ed Technology 2022; 14: ep374.

54.

Ingavélez-Guerra

Otón-Tortosa

Hilera-González

, et al. The use of accessibility metadata in e-learning environments: a systematic literature review. Univ Access Inf Soc 2023; 22: 445–461.

55.

Miao

. Students’ online interaction, self-regulation, and learning engagement in higher education: the importance of social presence to online learning. Front Psychol 2022; 13: 815220.

56.

Ashraf

Yang

Zhang

, et al. A systematic review of systematic reviews on blended learning: trends, gaps and future directions. PRBM 2021; 14: 1525–1541.

57.

Barati Jozan

Ghorbani

Khalid

, et al. Impact assessment of e-trainings in occupational safety and health: a literature review. BMC Public Health 2023; 23: 1187.

58.

Baig

Yadegaridehkordi

. Flipped classroom in higher education: a systematic literature review and research challenges. Int J Educ Technol High Educ 2023; 20: 61.

59.

Liu

Chen

. The impacts of learning motivation, emotional engagement and psychological capital on academic performance in a blended learning university course. Front Psychol 2024; 15: 1357936.

60.

Al-Aomar

. Combined ahp-entropy method for deriving subjective and objective criteria weights. Int J Industr Eng 2010; 17: 12–24.

61.

Kalem

Akpinar

. Personnel performance assessment using entropy based MABAC method: An application in the food sector. Equinox J Econ Bus Polit Stud. Epub ahead of print 18 March 2022. DOI: 10.48064/equinox.1063776.

62.

Shannon

. A mathematical theory of communication. Bell Syst Tech J 1948; 27: 379–423.

63.

Zeleny

(ed). Multiple criteria decision making Kyoto 1975. Berlin, Heidelberg: Springer Berlin Heidelberg. Epub ahead of print 1976. DOI: 10.1007/978-3-642-45486-8.

64.

Dong

Yang

. Application of the entropy-DEMATEL-VIKOR multicriteria decision-making method in public charging infrastructure. PLoS ONE 2021; 16: e0258209.

65.

Badeel

Subramaniam

Muhammed

, et al. A multicriteria decision-making framework for access point selection in hybrid LiFi/WiFi networks using integrated AHP–VIKOR technique. Sensors 2023; 23: 1312.

66.

Demir

Bouraima

Badi

, et al. Identification of industrial occupational safety risks and selection of optimum intervention strategies. Fuzzy MCDM Approach. Mathematics 2025; 13: 301.

67.

Seraji

. What differences? Thematic analyses of blended learning researches in Iran. Open Learn: J Open Distance e-Learn 2022; 37: 312–329.

68.

Ben Khalifa

Rehab Tharwat Abd El Ghani

Hoda Abdel Hameed

, et al. The impact of blended learning on academic achievement in smart universities: a contemporary analysis of digital transformation in Saudi higher education. Acad J Interdiscip Stud 2025; 14: 38.

69.

Wang

Raman

. Systematic literature review on the effects of blended learning in nursing education. Nurse Educ Pract 2025; 82: 104238.

70.

Adedoyin

Soykan

. COVID-19 pandemic and online learning: the challenges and opportunities. Inter Learn Environ 2023; 31: 863–875.

71.

Rehman

Hassan

Zubair

, et al. A framework for effective construction workers safety training using flipped learning. JCEM 2025; 31: 206–223.

72.

Hazmidar

Wulandari

Prayogo

, et al. Increasing the effectiveness of KPLH SAKTI (smart, awareness, knowledge, training, and improvement) training for the Q2 period (April-June 2025) as an effort to strengthen a resilient safety culture at PT Antareja Mahada Makmur Jobsite Mifa Bersaudara. Int J Soc Serv Res 5. Epub ahead of print 2025. 10.46799/ijssr.v5i10.1328

73.

Kara

Kayacan

. The effect of flipped learning model on Pre-service science Teachers’ laboratory practices. J Educ Sci Environ Health 2023; 9: 178–193.

Multi-criteria decision-making for selecting occupational health and safety training methods: An Entropy AHP–VIKOR approach

Abstract

Background

Objective

Methods

Results

Conclusions

Keywords

Introduction

Materials and methods

Definition of the decision problem

Alternatives

Evaluation criteria

Determining criteria weights with Entropy AHP method

Decision matrix and Entropy AHP criteration

Results

Ranking of alternatives using the VIKOR method

Discussion

Interpretation of findings

Comparison with literature

Practical implications

Implications for future research

Limitations

Conclusion

Footnotes

Acknowledgments

Ethical approval and informed consent

Author contributions

Funding

Declaration of conflicting interests

Data availability

ORCID iDs

References