Supply Chain Sustainability Drivers: Identification and Multi-Criteria Assessment

Abstract

Background: Supply chains operate under the changing influences of multiple external and internal factors. Sustainable supply chain development requires an assessment of these factors, as well as drivers and barriers. Various sustainability assessment criteria, methods, and models based on the consideration of the influence of different factors are used depending on the type and structure of the supply chain. Methods: The combination of DEMATEL and CRADIS multi-criteria methods is applied to rank the efficiency of drivers for achieving sustainable development goals, both for the supply chain as a whole and for each of its structural elements. Results: This study proposes a system of drivers for sustainable supply chain development. The peculiarity of the used supply chain model is the universality of its structure, which ensures the realization by the structural elements of the chain of all known functional areas of logistics. A framework for sustainable supply chain drivers’ multi-criteria assessment based on the use of the original two-level system of drivers’ assessment criteria is developed. Conclusions: The results of the supply chain sustainability drivers’ ranking and the framework developed by the authors are intended to justify decisions on the green logistics methods and instrument selection.

1. Introduction

The variability and uncertainty of many supply chain performance factors complicate decision making. Globalization, shorter product life cycles, government regulation, and the development of innovations lead to the emergence of new products and services, as well as changes in the goals and behavior of stakeholders in supply chains [1]. However, each supply chain participant assesses the impact of environmental factors differently. Such assessments do not always coincide with the objective intensity and strength of the impact of factors that have a different impact on the elements of supply chains. This complicates the coordinated functioning of the supply chain links and reduces their efficiency. Therefore, solving the problem of prioritizing factors and determining their importance and the degree of their influence on supply chain elements is necessary for effective decision making.

At present, the efficiency of supply chain performance is increasingly being assessed by the sustainability criterion—a complex criterion for achieving economic, social, and environmental goals. This is the peculiarity of the supply chain structure: elements of the same chain can be in regions with different natural, climatic, and political conditions, as well as different socio-economic levels. The only factor that has a similar impact on different elements of supply chains is the need to address a set of global environmental problems related to greenhouse gas emissions, environmental pollution, exhaustion of natural resources, and climate change.

According to UN Secretary General Antonio Guterres, “Every country, city, financial institution and company must adopt zero emission plans and act now to get on the right track towards this goal, which means reducing global emissions by 45% by 2030 compared to 2010 levels” [2]. Following this trend, many of the world’s leading organizations are planning to achieve carbon neutrality in their operations and supply chains [3]. Most countries have adopted Sustainable Development Goals (SDGs) and are implementing the concept of sustainable development. Growing pressure from stakeholders and governmental and non-governmental organizations forces various industries to implement sustainable supply chain management initiatives—Sustainable Supply Chain Management (SSCM) [4]. Integrating the concept of sustainable development into supply chain operations allows companies to create a competitive advantage [5]. Therefore, sustainability and SDGs are becoming increasingly relevant for inclusion in business logistics and supply chain management [6].

The implementation of sustainability initiatives depends on various internal and external drivers and barriers [7]. Recently, the composition of these drivers and barriers of sustainable and green supply chains has been actively debated in the scientific community [5,8,9]. Answers are sought as to what drivers influence sustainability [10,11,12], how to measure the performance of sustainable supply chains [9,13], and what models and methods to use for evaluation [14]. Many researchers have investigated individual drivers of SSCM. Despite the acknowledged importance of addressing this issue, the tasks of systematizing the known drivers of SSCM and determining their significance and the degree of their influence on supply chain elements to make effective decisions remain unsolved.

The main goal of our study is to systematize the factors, drivers, and barriers of supply chain sustainability to design a universal sustainability assessment framework using multi-criteria analysis. This study makes both theoretical and practical contributions to the field of sustainable supply chain management. First, based on the literature review and the structural–functional approach, we performed the identification and developed a universal system of supply chain sustainability drivers. The main idea of the systematization is to identify and group various factors that affect the performance of supply chain elements in their logistics functions. Second, we developed a methodology for ranking supply chain sustainability drivers using multi-criteria decision-making (MCDM) techniques. The methodology will help stakeholders to make effective decisions on logistics flow management using a justified sustainability assessment of supply chain elements as well as the factors impacting the performance of their logistics functions.

The paper is organized as follows. The second part reviews the existing research in the field of assessing the drivers and barriers of supply chain sustainability and identifies the main shortcomings of the existing approaches to sustainability assessment. The third part contains the description of the original methodology for assessing the drivers of supply chain sustainability. The methodology is based on the proposed universal system of sustainable supply chain drivers, which considers the structure and functions of supply chain elements, as well as a combined multi-criteria model of driver ranking. The fourth part presents a case study of sustainable supply chain driver ranking using the developed MCDM model. The conclusion presents the main results and the prospects for future research.

2. Literature Review

There are many articles related to supply chain management in the scientific literature of the world, collected in the Scopus database. Using the [Article title, Abstract, Keywords] filter and a combination of the keywords “Supply Chain” and “Drivers”, we found that there have been 3763 publications (including 2960 Articles and 803 Conference papers) indexed in Scopus in the past 20 years (from 2005 to 2024) (Figure 1). To identify keywords, we narrowed the list of articles using the filter [Article title] (“Green Supply Chain”) OR [Article title] “Sustainable Supply Chain” AND [Article title] (“Drivers”). As a result, we were able to identify the main research areas and keywords used in articles in relation to sustainable or green supply chain drivers. The ten most common keywords were Supply Chain Management, Green Supply Chain Management, Sustainability, Sustainable Development, Drivers, Environmental Management, Sustainable Supply Chains, Sustainable Supply Chain Management, Green Supply Chain, and Barriers. The most common terms used by the authors in relation to the causes of supply chain sustainability were “drivers”, “drivers and barriers”, “factors”, “driving factors”, “internal drivers”, “critical factors”, “critical success factors”, “enablers”, “pressures”, and others. We used these keywords to search for open-access publications in the scientific databases Elsilver, Google Scholar, MDPI, and ResearchGate. In total, we managed to find 89 articles devoted to the problems of supply chain sustainability assessment.

We attempted to address two main objectives in the process of analyzing publications: a comprehensive analysis of existing research on identifying and evaluating supply chain sustainability drivers and the identification of possible gaps in existing research.

The paper analyzes 101 articles on the problems of supply chain sustainability assessment (Table 1). The review of the publications allowed us to identify the following main research areas:

• Factors, drivers, and barriers of supply chain sustainability, as well as indicators of their assessment.
• Models and methods for assessing the performance and sustainability of different types of supply chains.
• Case studies of supply chains of different types and structures.
• The generalized results of the analysis are presented in Figure 2.

The distribution of articles by type of research indicates that case studies predominate, with a share of 80%. Conceptual studies and review articles account for 11% and 9%, respectively. The results for each group of articles are analyzed below.

2.1. Review Papers

Paper [9] emphasizes supply chain structure, inventory control policy, information sharing, customer demand, forecasting method, lead time, and review period length as the main factors of supply chain efficiency. According to the authors [9], an optimal set of parameters of these factors increases the efficiency of the supply chain. By analyzing 1559 sustainable supply chain drivers (from 217 papers) [8], the authors compiled a list of 40 unique SSCM drivers. All SSCM drivers were grouped into two groups—internal and external drivers. The group of internal drivers included corporate strategy, organizational culture, organizational resources, organizational characteristics. The group of external drivers included regulatory pressures, societal pressures, and market pressures. In [11], the results of the study of critical factors for the realization of sustainable supply chain innovations are presented. Fourteen main categories of critical factors are identified, the most significant of which are cooperation, strategic orientation, culture, practice, and political context. Paper [15] investigated the realization of sustainable supply chains. The authors researched 115 papers and identified 69 attributes of supply chain visibility, which were grouped into 15 groups of factors. In [10], based on a review of 47 articles, an ABCDE framework is proposed which includes five groups of supply chain visibility factors: antecedents (A), barriers and challenges (BC), drivers (D), and effects (E), and provides a holistic view of supply chain visibility. In [16], based on the analysis of 59 research papers, 37 drivers and 36 barriers are systematized as the main factors affecting the performance of reverse logistics. In [17], based on a literature review, a grouping of 10 drivers and 15 barriers of green supply chains is proposed considering internal and external factors. A review of 188 papers [18] identified 15 success factors to implement sustainable supply chain management. Based on a systematic review of 362 papers [5], the authors found 22 drivers and 19 barriers of sustainable supply chains. The authors note the prevalence of case studies, which help to analyze the latest advances in SSCM for a specific supply chain.

2.2. Conceptual Studies

The authors in [19] concluded that empirical studies on green supply chain management factors show different results. They investigated six factors and showed that green manufacturing was the most influential and the green logistics factor was influenced by all other factors. The results of ranking 13 green supply chain factors in [20] found a strong influence of lack of leadership activity. This is due to the view of the management that implementing a green supply chain will bring negative effects on finance. The authors of [21] investigated synchromodal logistics critical success factors. Six groups of factors were identified: network, collaboration, and trust; sophisticated planning; ICT (Information and Communication Technologies) and ITS (Intelligent Transportation Systems); physical infrastructure; legal and political framework; awareness and mental shift; and pricing/cost/service. Furthermore, the technologies that influence these critical factors are substantiated.

Paper [12] investigated the internal environment factors of supply chains and identified eight groups of factors, the most significant of which were top management commitment, reverse logistic management, materials store and management. Out of the eight green supply chain factors investigated in [22], stakeholder cooperation was the most important factor of eco design. By analyzing the factors underlying successful sustainable supply chain management, [23] found that signaling, information provision, and the adoption of standards are crucial preconditions for strategy commitment, mutual learning, and the establishment of ecological cycles, as well as, hence, the overall success of SSCM. In [24], the factors influencing the development of supply chains are considered, and a model for the development of the regional logistics industry is proposed. The authors in [25] suggested a model of lean and sustainable supply chain management considering eight factors with 29 influencing determinants. The model in [26] of green-lean supply chain management integration considers 15 factors. Paper [7] identifies exogenous drivers and endogenous drivers of supply chains and recommends initiatives to move towards a sustainable state.

2.3. Case Studies

The Green Supply Chain and Sustainable Supply Chain are the most frequent objects of study in case studies. They account for 58% and 22% of all case studies, respectively. Other case studies investigate the drivers [27] and barriers [28] of the Close-Loop Supply Chain, Resilience Supply Chain factors [29], and critical success factors of reverse logistics [30]. The distribution of case studies by country shows the predominance of India (29% of articles) and China (14% of articles). Bangladesh, Thailand, the UK, and Turkey each account for 4–6% of the analyzed articles, while the EU, Germany, Iran, Malaysia, Pakistan, Taiwan, and Vietnam each account for 3%. Examples of supply chain research in Brazilia, Egypt, Indonesia, Jordan, Korea, Lithuania, Middle East, New Zeland, Nigeria, Spain, and the USA are each represented by one article in the analyzed sample.

Factors (43% of the total number of case study articles), drivers (22% of the articles), and barriers (18% of the articles) are the most frequent subjects of study. Simultaneous studies of drivers and barriers are found in 8% of the case study articles. Several studies focus on the study of drivers, drivers, and barriers of individual supply chain elements: input element [31,32,33,34], processing element [35,36,37,38,39], cumulative element [40], transport element [41,42], and output element [43,44,45,46]. However, most studies (about 50% of all studies) focus on the comprehensive assessment of the impact of factors on the entire supply chain.

The most used methods for assessing supply chain sustainability factors are: Interpretive Structural Modeling (ISM); Matrice d’Impacts Cruoses Multiplication Applique a un Classement (MICMAC), and factor analysis. These methods account for 20% and 13%, respectively. The most frequent MCDM methods are DEcision-Making Trial and Evaluation Laboratory, DEMATEL (10% of articles); Fuzzy DEMATEL (7% of articles); and Analytic Hierarchy Process, AHP (6% of articles).

We analyze and evaluate factors, drivers, and barriers using two types of grouping: one-level (53% of articles) and two-level (40%). The number of factors (drivers) can vary and depends on the type of supply chain. In a one-level grouping, the number of factors (drivers) ranges from 3 to 41. In a two-level grouping, the first level contains from 2 to 14; the second level contains from 6 to 73.

Table 1. Supply chain factor assessment studies.

№	Supply Chain Type	Object of Assessment	Number of Factors, Drivers, and Barriers	Methods and Models **	Reference
1	Supply Chain	Other	7 factors	—	[9]
2	Green Supply Chain	Supply Chain	7 factors	DEMATEL	[19]
3	Green Supply Chain	Other	13 factors	ISM	[20]
4	Sustainable Supply Chain	Supply Chain	7 drivers/40 sub-drivers	SLR	[8]
5	Green Supply Chain	Supply Chain	8 factors	ISM-MICMAC	[12]
6	Green Supply Chain	Other	8 factors	FQFD	[22]
7	Sustainable Supply Chain	Supply Chain	8 factors	SLR/Survey	[23]
8	Green Supply Chain	Supply Chain	16 factors	DEMATEL	[4]
9	Green Supply Chain	Supply Chain	5 factors/24 sub-factors	AHP-DEMATEL	[47]
10	Other Supply Chain	Other	6 factors	ANOVA	[48]
11	Green Supply Chain	Supply Chain	11 drivers	ISM-MICMAC	[49]
12	Green Supply Chain	Supply Chain	4 critical factors/20 sub-factors	Factor Analysis	[50]
13	Other Supply Chain	Other	3 factors/20 sub-factors	ANOVA	[51]
14	Reverse Supply Chain	Supply Chain	5 factors/25 sub-factors	AHP-DEMATEL	[30]
15	Close-Loop Supply Chain	Other	3 drivers	—	[52]
16	Supply Chain	Supply Chain	3 drivers	AHP	[53]
17	Green Supply Chain	Processing element *	12 drivers	Fuzzy AHP	[35]
18	Sustainable Supply Chain	Supply Chain	13 factors	Hesitant Fuzzy DEMATEL	[54]
19	Sustainable Supply Chain	Supply Chain	14 critical factors/62 sub-factors	SLR	[11]
20	Sustainable Supply Chain	Supply Chain	4 critical factors/20 sub-factors	Factor Analysis	[55]
21	Resilience Supply Chain	Supply Chain	15 factors	Data Analysis	[29]
22	Sustainable Supply Chain	Supply Chain	15 factors	SLR	[15]
23	Supply Chain	Supply Chain	4 factors/43 sub-factors	SLR	[10]
24	Supply Chain	Supply Chain	4 barriers	Cross-case Analysis	[3]
25	Green Supply Chain	Supply Chain	5 barriers/22 sub-barriers	AHP	[56]
26	Reverse Supply Chain	Supply Chain	7 drivers/37 sub-drivers 7 barriers/36 sub-barriers	SLR	[16]
27	Green Supply Chain	Supply Chain	2 drivers/10 sub-drivers 2 barriers/15 sub-barriers	SLR	[17]
28	Green Supply Chain	Supply Chain	19 factors	ISM-MICMAC	[57]
29	Green Supply Chain	Supply Chain	26 critical factors	ISM-MICMAC	[58]
30	Sustainable Supply Chain	Supply Chain	7 critical factors/32 sub-factors	ISM-MICMAC	[59]
31	Green Supply Chain	Input element *	7 drivers/26 sub-drivers	Structural Equation Modelling	[31]
32	Sustainable Supply Chain	Input element *	7 drivers/17 sub-drivers	Hierarchical Linear Modelling	[32]
33	Green Supply Chain	Supply Chain	8 drivers	ISM-MICMAC	[60]
34	Green Supply Chain	Input element *	5 factors/15 sub-factors	Fuzzy DEMATEL	[61]
35	Green Supply Chain	Processing element *	13 drivers	ISM-MICMAC	[36]
36	Other Supply Chain	Output element *	8 barriers	Grey DEMATEL	[43]
37	Sustainable Supply Chain	Supply Chain	7 factors	GRA	[62]
38	Green Supply Chain	Supply Chain	18 barriers	ISM	[63]
39	Green Supply Chain	Output element *	13 factors	DEMATEL	[44]
40	Green Supply Chain	Supply Chain	5 factors/28 sub-factors	Factor Analysis	[64]
41	Green Supply Chain	Supply Chain	15 barriers	Hierarchical Clustering Analysis	[65]
42	Green Supply Chain	Supply Chain	4 critical factors/25 sub-factors	DEMATEL	[66]
43	Green Supply Chain	Supply Chain	10 factors	Fuzzy DEMATEL	[67]
44	Green Supply Chain	Transport element *	10 factors	Statistical Analysis	[41]
45	Green Supply Chain	Supply Chain	5 drivers/18 sub-drivers	ANOVA	[68]
46	Green Supply Chain	Transport element *	8 drivers	Structural Equation Modelling	[42]
47	Green Supply Chain	Supply Chain	3 drivers	Survey	[69]
48	Sustainable Supply Chain	Input element *	17 drivers and 16 barriers	Delphi method	[33]
49	Supply Chain	Supply Chain	32 drivers	Structural Equation Modelling	[70]
50	Green Supply Chain	Supply Chain	26 barriers	ISM-MICMAC	[71]
51	Green Supply Chain	Supply Chain	14 barriers	ISM-MICMAC	[72]
52	Green Supply Chain	Input element *	10 barriers	ISM-MICMAC	[45]
53	Sustainable Supply Chain	Supply Chain	13 barriers	ISM-MICMAC	[73]
54	Supply Chain	Processing element *	10 drivers and 4 barriers	—	[37]
55	Green Supply Chain	Supply Chain	7 drivers and 10 barriers	STI	[74]
56	Sustainable Supply Chain	Supply Chain	11 drivers	Fuzzy TISM-MICMAC	[75]
57	Sustainable Supply Chain	Supply Chain	3 drivers/11 sub-drivers 2 barriers/11 sub-barriers	—	[76]
58	Green Supply Chain	Supply Chain	5 barriers/22 sub-barriers	DEMATEL-Fuzzy EDAS-Fuzzy COPRAS	[77]
59	Supply Chain	Supply Chain	14 barriers	Fuzzy TISM-MICMAC	[78]
60	Close-Loop Supply Chain	Supply Chain	4 drivers/21 sub-drivers	Grey DEMATEL	[27]
61	Close-Loop Supply Chain	Supply Chain	6 barriers/35 sub-barriers	Pythagorean Fuzzy AHP-DEMATEL	[28]
62	Green Supply Chain	Supply Chain	9 critical factors	Fuzzy DEMATEL	[79]
63	Green Supply Chain	Output element *	25 critical factors	Factor Analysis	[46]
64	Other Supply Chain	Supply Chain	6 factors	GRA	[80]
65	Supply Chain	Supply Chain	6 factors/13 sub-factors	AHP	[81]
66	Supply Chain	Supply Chain	2 factors/10 sub-factors	—	[24]
67	Other Supply Chain	Transport element *	7 critical factors	—	[21]
68	Sustainable Supply Chain	Supply Chain	15 factors	ISM-MICMAC	[18]
69	Sustainable Supply Chain	Supply Chain	7 drivers and 6 barriers	AHP-TOPSIS, AHP-COPRAS	[82]
70	Green Supply Chain	Input element *	11 factors	SWARA-TOPSIS	[83]
71	Green Supply Chain	Supply Chain	3 critical factors/12 sub-factors	DEMATEL	[84]
72	Green Supply Chain	Supply Chain	12 factors	ISM-MICMAC	[85]
73	Green Supply Chain	Input element *	5 factors/17 sub-factors	ANP-TOPSIS	[86]
74	Green Supply Chain	Supply Chain	5 drivers	SEM	[87]
75	Other Supply Chain	Supply Chain	3 factors/8 sub-factors	ANP-AHP-BOCR	[25]
76	Green Supply Chain	Processing element *	5 factors/21 sub-factors	Confirmatory factor analysis	[88]
77	Sustainable Supply Chain	Processing element *	4 factors/14 sub-factors	Factor Analysis	[38]
78	Supply Chain	Processing element *	3 barriers/15 sub-barriers	Grey DEMATEL	[39]
79	Supply Chain	Cumulative element *	15 factors	ISM-MICMAC	[40]
80	Other Supply Chain	Transport element *	8 factors	—	[89]
81	Green Supply Chain	Supply Chain	5 drivers/17 sub-drivers	Fuzzy DEMATEL-Fuzzy ANP-Fuzzy TOPSIS	[90]
82	Green Supply Chain	Supply Chain	3 barriers/13 sub-barriers	AHP-ELECTRE I	[91]
83	Sustainable Supply Chain	Input element *	3 factors/10 sub-factors	DEMATEL	[92]
84	Green Supply Chain	Input element *	5 factors/21 sub-factors	AHP	[34]
85	Green Supply Chain	Input element *	15 factors	Fuzzy DEMATEL	[26]
86	Supply Chain	Supply Chain	10 factors	Grey system theory	[1]
87	Sustainable Supply Chain	Supply Chain	22 drivers and 19 barriers	SLR	[5]
88	Green Supply Chain	Supply Chain	2 drivers/6 sub-drivers	—	[7]
89	Green Supply Chain	Supply Chain	7 factors/47 sub-factors	Factor Analysis	[93]
90	Green Supply Chain	Supply Chain	5 drivers and 5 barriers	STI	[94]
91	Sustainable Supply Chain	Supply Chain	4 factors/12 sub-factors	Factor Analysis	[95]
92	Green Supply Chain	Supply Chain	15 drivers	ISM-MICMAC-DEMATEL	[96]
93	Green Supply Chain	Supply Chain	20 drivers and 16 pressures	—	[97]
94	Green Supply Chain	Supply Chain	4 factors/12 sub-factors	Statistical Analysis	[98]
95	Green Supply Chain	Supply Chain	3 factors/12 sub-factors	Statistical Analysis	[99]
96	Sustainable Supply Chain	Processing element *	4 factors/8 sub-factors	SEM-ANN	[100]
97	Sustainable Supply Chain	Supply Chain	10 factors	ISM	[101]
98	Green Supply Chain	Supply Chain	6 factors/30 sub-factors	Interval Type-2 Fuzzy AHP	[102]
99	Sustainable Supply Chain	Supply Chain	11 factors	TISM-MICMAC	[103]
100	Green Supply Chain	Supply Chain	22 factors	Duo-theme DEMATEL	[104]
101	Sustainable Supply Chain	Supply Chain	20 drivers	FDM-FISM-ANP-TOPSIS	[105]

* The authors identified the elements of the supply chain based on a structural–functional approach. A detailed description of the structural elements of the supply chain and their functions is presented in [106] and Section 3.1 of this paper. ** AHP—Analytic Hierarchy Process; ANN—Artificial neural networks; ANOVA—Analysis of Variance; ANP—Analytic Network Process; BOCR—Benefit, Opportunity, Cost, and Risk; COPRAS—Complex Proportional Assessment; DEMATEL—DEcision MAking Trial and Evaluation Laboratory; EDAS—Evaluation based on Distance from Average Solution; ELECTRE—ELimination and Choice Expressing the Reality; FDM—Fuzzy Delphi method; FISM—Fuzzy Interpretative Structural Modeling; FQFD—Fuzzy Quality Function Deployment; GRA—Grey Relation Analysis; ISM—Interpretive Structural Modeling; MICMAC—Matriced Impacts Cruoses Multiplication Applique a un Classement techniques; SEM—Structural Equation Modeling; SWARA—Stepwise Weight Assessment Ratio Analysis; SLR—systematic literature review; STI—semi-structured interview; TISM—Total Interpretive Structural Modeling; TOPSIS—Technique for the Order of Preference by Similarity to Ideal Solution; SEM—Structural Equation Modeling.

Table 1. Supply chain factor assessment studies.

№	Supply Chain Type	Object of Assessment	Number of Factors, Drivers, and Barriers	Methods and Models **	Reference
1	Supply Chain	Other	7 factors	—	[9]
2	Green Supply Chain	Supply Chain	7 factors	DEMATEL	[19]
3	Green Supply Chain	Other	13 factors	ISM	[20]
4	Sustainable Supply Chain	Supply Chain	7 drivers/40 sub-drivers	SLR	[8]
5	Green Supply Chain	Supply Chain	8 factors	ISM-MICMAC	[12]
6	Green Supply Chain	Other	8 factors	FQFD	[22]
7	Sustainable Supply Chain	Supply Chain	8 factors	SLR/Survey	[23]
8	Green Supply Chain	Supply Chain	16 factors	DEMATEL	[4]
9	Green Supply Chain	Supply Chain	5 factors/24 sub-factors	AHP-DEMATEL	[47]
10	Other Supply Chain	Other	6 factors	ANOVA	[48]
11	Green Supply Chain	Supply Chain	11 drivers	ISM-MICMAC	[49]
12	Green Supply Chain	Supply Chain	4 critical factors/20 sub-factors	Factor Analysis	[50]
13	Other Supply Chain	Other	3 factors/20 sub-factors	ANOVA	[51]
14	Reverse Supply Chain	Supply Chain	5 factors/25 sub-factors	AHP-DEMATEL	[30]
15	Close-Loop Supply Chain	Other	3 drivers	—	[52]
16	Supply Chain	Supply Chain	3 drivers	AHP	[53]
17	Green Supply Chain	Processing element *	12 drivers	Fuzzy AHP	[35]
18	Sustainable Supply Chain	Supply Chain	13 factors	Hesitant Fuzzy DEMATEL	[54]
19	Sustainable Supply Chain	Supply Chain	14 critical factors/62 sub-factors	SLR	[11]
20	Sustainable Supply Chain	Supply Chain	4 critical factors/20 sub-factors	Factor Analysis	[55]
21	Resilience Supply Chain	Supply Chain	15 factors	Data Analysis	[29]
22	Sustainable Supply Chain	Supply Chain	15 factors	SLR	[15]
23	Supply Chain	Supply Chain	4 factors/43 sub-factors	SLR	[10]
24	Supply Chain	Supply Chain	4 barriers	Cross-case Analysis	[3]
25	Green Supply Chain	Supply Chain	5 barriers/22 sub-barriers	AHP	[56]
26	Reverse Supply Chain	Supply Chain	7 drivers/37 sub-drivers 7 barriers/36 sub-barriers	SLR	[16]
27	Green Supply Chain	Supply Chain	2 drivers/10 sub-drivers 2 barriers/15 sub-barriers	SLR	[17]
28	Green Supply Chain	Supply Chain	19 factors	ISM-MICMAC	[57]
29	Green Supply Chain	Supply Chain	26 critical factors	ISM-MICMAC	[58]
30	Sustainable Supply Chain	Supply Chain	7 critical factors/32 sub-factors	ISM-MICMAC	[59]
31	Green Supply Chain	Input element *	7 drivers/26 sub-drivers	Structural Equation Modelling	[31]
32	Sustainable Supply Chain	Input element *	7 drivers/17 sub-drivers	Hierarchical Linear Modelling	[32]
33	Green Supply Chain	Supply Chain	8 drivers	ISM-MICMAC	[60]
34	Green Supply Chain	Input element *	5 factors/15 sub-factors	Fuzzy DEMATEL	[61]
35	Green Supply Chain	Processing element *	13 drivers	ISM-MICMAC	[36]
36	Other Supply Chain	Output element *	8 barriers	Grey DEMATEL	[43]
37	Sustainable Supply Chain	Supply Chain	7 factors	GRA	[62]
38	Green Supply Chain	Supply Chain	18 barriers	ISM	[63]
39	Green Supply Chain	Output element *	13 factors	DEMATEL	[44]
40	Green Supply Chain	Supply Chain	5 factors/28 sub-factors	Factor Analysis	[64]
41	Green Supply Chain	Supply Chain	15 barriers	Hierarchical Clustering Analysis	[65]
42	Green Supply Chain	Supply Chain	4 critical factors/25 sub-factors	DEMATEL	[66]
43	Green Supply Chain	Supply Chain	10 factors	Fuzzy DEMATEL	[67]
44	Green Supply Chain	Transport element *	10 factors	Statistical Analysis	[41]
45	Green Supply Chain	Supply Chain	5 drivers/18 sub-drivers	ANOVA	[68]
46	Green Supply Chain	Transport element *	8 drivers	Structural Equation Modelling	[42]
47	Green Supply Chain	Supply Chain	3 drivers	Survey	[69]
48	Sustainable Supply Chain	Input element *	17 drivers and 16 barriers	Delphi method	[33]
49	Supply Chain	Supply Chain	32 drivers	Structural Equation Modelling	[70]
50	Green Supply Chain	Supply Chain	26 barriers	ISM-MICMAC	[71]
51	Green Supply Chain	Supply Chain	14 barriers	ISM-MICMAC	[72]
52	Green Supply Chain	Input element *	10 barriers	ISM-MICMAC	[45]
53	Sustainable Supply Chain	Supply Chain	13 barriers	ISM-MICMAC	[73]
54	Supply Chain	Processing element *	10 drivers and 4 barriers	—	[37]
55	Green Supply Chain	Supply Chain	7 drivers and 10 barriers	STI	[74]
56	Sustainable Supply Chain	Supply Chain	11 drivers	Fuzzy TISM-MICMAC	[75]
57	Sustainable Supply Chain	Supply Chain	3 drivers/11 sub-drivers 2 barriers/11 sub-barriers	—	[76]
58	Green Supply Chain	Supply Chain	5 barriers/22 sub-barriers	DEMATEL-Fuzzy EDAS-Fuzzy COPRAS	[77]
59	Supply Chain	Supply Chain	14 barriers	Fuzzy TISM-MICMAC	[78]
60	Close-Loop Supply Chain	Supply Chain	4 drivers/21 sub-drivers	Grey DEMATEL	[27]
61	Close-Loop Supply Chain	Supply Chain	6 barriers/35 sub-barriers	Pythagorean Fuzzy AHP-DEMATEL	[28]
62	Green Supply Chain	Supply Chain	9 critical factors	Fuzzy DEMATEL	[79]
63	Green Supply Chain	Output element *	25 critical factors	Factor Analysis	[46]
64	Other Supply Chain	Supply Chain	6 factors	GRA	[80]
65	Supply Chain	Supply Chain	6 factors/13 sub-factors	AHP	[81]
66	Supply Chain	Supply Chain	2 factors/10 sub-factors	—	[24]
67	Other Supply Chain	Transport element *	7 critical factors	—	[21]
68	Sustainable Supply Chain	Supply Chain	15 factors	ISM-MICMAC	[18]
69	Sustainable Supply Chain	Supply Chain	7 drivers and 6 barriers	AHP-TOPSIS, AHP-COPRAS	[82]
70	Green Supply Chain	Input element *	11 factors	SWARA-TOPSIS	[83]
71	Green Supply Chain	Supply Chain	3 critical factors/12 sub-factors	DEMATEL	[84]
72	Green Supply Chain	Supply Chain	12 factors	ISM-MICMAC	[85]
73	Green Supply Chain	Input element *	5 factors/17 sub-factors	ANP-TOPSIS	[86]
74	Green Supply Chain	Supply Chain	5 drivers	SEM	[87]
75	Other Supply Chain	Supply Chain	3 factors/8 sub-factors	ANP-AHP-BOCR	[25]
76	Green Supply Chain	Processing element *	5 factors/21 sub-factors	Confirmatory factor analysis	[88]
77	Sustainable Supply Chain	Processing element *	4 factors/14 sub-factors	Factor Analysis	[38]
78	Supply Chain	Processing element *	3 barriers/15 sub-barriers	Grey DEMATEL	[39]
79	Supply Chain	Cumulative element *	15 factors	ISM-MICMAC	[40]
80	Other Supply Chain	Transport element *	8 factors	—	[89]
81	Green Supply Chain	Supply Chain	5 drivers/17 sub-drivers	Fuzzy DEMATEL-Fuzzy ANP-Fuzzy TOPSIS	[90]
82	Green Supply Chain	Supply Chain	3 barriers/13 sub-barriers	AHP-ELECTRE I	[91]
83	Sustainable Supply Chain	Input element *	3 factors/10 sub-factors	DEMATEL	[92]
84	Green Supply Chain	Input element *	5 factors/21 sub-factors	AHP	[34]
85	Green Supply Chain	Input element *	15 factors	Fuzzy DEMATEL	[26]
86	Supply Chain	Supply Chain	10 factors	Grey system theory	[1]
87	Sustainable Supply Chain	Supply Chain	22 drivers and 19 barriers	SLR	[5]
88	Green Supply Chain	Supply Chain	2 drivers/6 sub-drivers	—	[7]
89	Green Supply Chain	Supply Chain	7 factors/47 sub-factors	Factor Analysis	[93]
90	Green Supply Chain	Supply Chain	5 drivers and 5 barriers	STI	[94]
91	Sustainable Supply Chain	Supply Chain	4 factors/12 sub-factors	Factor Analysis	[95]
92	Green Supply Chain	Supply Chain	15 drivers	ISM-MICMAC-DEMATEL	[96]
93	Green Supply Chain	Supply Chain	20 drivers and 16 pressures	—	[97]
94	Green Supply Chain	Supply Chain	4 factors/12 sub-factors	Statistical Analysis	[98]
95	Green Supply Chain	Supply Chain	3 factors/12 sub-factors	Statistical Analysis	[99]
96	Sustainable Supply Chain	Processing element *	4 factors/8 sub-factors	SEM-ANN	[100]
97	Sustainable Supply Chain	Supply Chain	10 factors	ISM	[101]
98	Green Supply Chain	Supply Chain	6 factors/30 sub-factors	Interval Type-2 Fuzzy AHP	[102]
99	Sustainable Supply Chain	Supply Chain	11 factors	TISM-MICMAC	[103]
100	Green Supply Chain	Supply Chain	22 factors	Duo-theme DEMATEL	[104]
101	Sustainable Supply Chain	Supply Chain	20 drivers	FDM-FISM-ANP-TOPSIS	[105]

* The authors identified the elements of the supply chain based on a structural–functional approach. A detailed description of the structural elements of the supply chain and their functions is presented in [106] and Section 3.1 of this paper. ** AHP—Analytic Hierarchy Process; ANN—Artificial neural networks; ANOVA—Analysis of Variance; ANP—Analytic Network Process; BOCR—Benefit, Opportunity, Cost, and Risk; COPRAS—Complex Proportional Assessment; DEMATEL—DEcision MAking Trial and Evaluation Laboratory; EDAS—Evaluation based on Distance from Average Solution; ELECTRE—ELimination and Choice Expressing the Reality; FDM—Fuzzy Delphi method; FISM—Fuzzy Interpretative Structural Modeling; FQFD—Fuzzy Quality Function Deployment; GRA—Grey Relation Analysis; ISM—Interpretive Structural Modeling; MICMAC—Matriced Impacts Cruoses Multiplication Applique a un Classement techniques; SEM—Structural Equation Modeling; SWARA—Stepwise Weight Assessment Ratio Analysis; SLR—systematic literature review; STI—semi-structured interview; TISM—Total Interpretive Structural Modeling; TOPSIS—Technique for the Order of Preference by Similarity to Ideal Solution; SEM—Structural Equation Modeling.

Thus, the analysis of scientific works in the field of supply chain sustainability assessment allows us to draw the following general conclusions.

1. There is an increase in the number of scientific publications devoted to sustainable and green supply chain management, including works related to the identification of factors, drivers, and barriers of SSCM. At the same time, there is a discrepancy in the terminology used. The authors use different concepts—“factors”, “critical factors”, “critical success factors”, and “drivers and barriers”—to identify the causes that affect the sustainability of supply chains. This variety of concepts makes it difficult to understand and systematize the causes of supply chain sustainability. Moreover, there is frequent duplication of the same reasons with different names, which, as a consequence, complicates the assessment of supply chain sustainability.

2. We were unable to find generally accepted systems of factors and indicators for assessing the impact of these factors on SSCM. Researchers use different factors and indicators for assessing these factors, depending on the type of supply chain, the elements under study, or the tasks to be solved. In most cases, two grouping types of sustainability factors are used. The first one-level type contains only factors. In the second type, a second additional level of sub-factors is introduced. In a one-level grouping, the number of factors ranges from 3 to 41 (average 13). The number of factor groups ranges from 2 to 14 (average 5) in the two-level grouping, and the number of sub-factors in the groups ranges from 6 to 73 (average 24).

3. We have identified a wide variety of supply chain sustainability factors. Many authors note the complexity of collecting input data for their assessment. This explains the researchers’ wide use of multi-criteria analysis methods and expert methods, including those based on the provisions of fuzzy set theory, gray systems theory, and similar. The most frequently used methods are the Interpretive Structural Modeling and MICMAC approach (20% of the total number of methods used), factor analysis (about 13%), DEMATEL and Fuzzy DEMATEL (respectively, 10% and 7%), AHP (about 6%), and others (see Table 1). The use of Crisp prevails in the studies (64% of the total number of papers analyzed). The statements of Fuzzy Set Theory and Gray System Theory are found in 12% and 6% of the analyzed papers, respectively.

4. The disadvantage of most of the approaches used by researchers is the lack of a systematic approach to assessing the sustainability of all elements and processes in supply chains in interconnection. The most frequent objects of assessment are individual elements of the supply chain and their functions—“sustainable supply (input element)” and “sustainable production (processing element)” (9% and 7%, respectively). The lowest number of studies concern “sustainable warehousing (cumulative element)”.

3. Methodology for Assessing Sustainable Supply Chain Drivers

The proposed methodology for the assessment of sustainable supply chain (SSC) drivers includes methods for describing the structure and functions of SSC, a system of SSC drivers, and a substantiated system of SSC assessment criteria. This methodology formed the basis of the driver assessment and ranking framework.

3.1. Sustainable Supply Chain Structure and Functions

The structural elements of the supply chain and their functions are identified in accordance with the concept proposed in [106]. The supply chain is considered as a fragment of the global logistics system and is linearly ordered along the logistics flow, a set of logistics elements. The specification of logistics elements is based on the idea of functional isolation of an element and the limitation of its decomposition within a certain management task. In addition, the specificity of the element is determined by a set of certain functions, the fulfillment of which ensures the achievement of both its local goal and the goal for the entire logistics system. In [106], the following supply chain elements are identified: input, processing, cumulative, transport, output, and control elements.

Each element of the supply chain (SC) performs the following basic functions to influence logistics flows (Figure 3):

• Input element (E1): the entry of material flows into the system, i.e., the purchase of necessary raw materials, supplies, or services.
• Processing element (E2): changes the qualitative properties of material flows and is involved in their transformation from raw materials into finished products.
• Cumulative element (E3): regulation of the speed of material flows as a result of their inhibition, accumulation, and storage.
• Transport element (E4): acceleration and movement of material flows.
• Output element (E5): withdrawal of material flow from the system, marketing, and distribution of finished products and services.
• Control element (E6): provides information and financial connection between other LS elements, controlling their functions and operations and regulating the promotion of information and financial flows in the SC.

Figure 3. Supply chain structure and functions.

Figure 3. Supply chain structure and functions.

The basis functions are logistic elements’ generalized functions. The realization of basis functions is achieved by performing specific supporting functions accordingly [107,108,109].

The presented structural–functional approach to describing the logistics system is fundamentally different from the common way of identifying functional areas of logistics, such as transportation, sales, production, supply, and warehousing logistics. The disadvantage of this traditional functional approach is the “binding” of logistics functions and operations to infrastructural elements of logistics chains—warehouses, industrial enterprises, supply and sales departments, and transportation. As a result, the same method of logistics flow management can be realized in different functional areas of logistics [110]. For example, the same “green” solutions are implemented using different methodological frameworks regulated by different, often contradictory, regulatory and legal rules [106].

SC sustainability is achieved as a result of orientation of each of its elements to the overall objectives of the logistics system—economic, technological, social, and environmental. Effective coordination of SC elements requires systematization of and research into SC sustainability factors (drivers). Assessment of the impact of these factors (drivers) on SC elements is necessary for making decisions to achieve the goals of the concept of sustainable development in relation to supply chains.

3.2. Sustainable Supply Chain Driver System

Supply chain sustainability drivers are a causes, resources, and actions that have a meaningful impact on accelerating the process of supply chain sustainability.

Table 2. Sustainable supply chain driver system.

Supply Chain Elements	Sustainable Supply Chain Drivers	References
Input element	D1.1—Environmentally friendly raw materials (at supplier)	[19,31,43,50,55,56,59,60,61,64,66,67,68,69,72,77,96,99,101,102]
	D1.2—Raw materials able to reuse or recycle	[11,16,19,27,31,33,44,46,49,56,64,97,104]
	D1.3—Raw materials procurement system	[30,31,36,50,52,53,57,59,60,64,66,67,81,105]
	D1.4—Eco-friendly suppliers	[4,10,12,17,22,28,32,33,42,47,48,49,50,55,57,58,60,61,63,64,66,67,68,72,75,76,77,94,96,97,99,101,102,103,105]
	D1.5—Delivery distance of raw materials	[31,32,64]
	D1.6—Type of packaging for raw materials	[12,16,36,46,52,53,60,61,64,68,73,94]
	D1.7—Raw material eco-labeling	[16,46,59]
	D1.8—E-commerce with supplier	[62,75]
Processing element	D2.1—Eco-friendly raw materials from the manufacturer	[11,12,19,43,51,56,59,60,64,66,68,69,72,77,96,99]
	D2.2—Reusable or recyclable raw materials	[16,27,44,46,56,64,77,104]
	D2.3—Eco-friendly equipment	[8,19,33,44,51,59,64,69,97,102]
	D2.4—Energy and resource saving technologies	[4,22,33,35,36,44,47,49,69,72,94,102,104,105]
	D2.5—Eco-friendly production technologies	[4,8,19,22,33,35,36,44,47,51,56,57,58,61,63,64,65,69,70,71,72,73,77,78,81,95,96,97,102,103,104]
	D2.6—Environmental protection systems	[20,32,35,51,61,64]
	D2.7—Industrial waste	[11,12,16,20,28,30,31,36,41,44,50,52,58,64,66,67,72,73,77,97,104,105]
	D2.8—Labor conditions	[51,55,58,59]
	D2.9—Eco-learning	[4,8,11,16,19,23,28,30,31,32,36,47,48,54,56,57,58,59,62,64,65,72,73,74,77,78,79,81,95,96,102,103,104]
Transport element	D3.1—Transport type	[19,51,61,69,81]
	D3.2—Transport link type	[51]
	D3.3—Route of transportation	[64]
	D3.4—Cargo flow structure	[81]
	D3.5—Frequency and size of shipments	[31,36,50,52,53,57,59,60,81]
	D3.6—Compliance of transportation vehicles with legal regulations	[19,35,44,51,61,64,69,94,102,104]
	D3.7—Fuel type	[51,64]
	D3.8—Technical condition of vehicle fleet	[64]
	D3.9—Vehicle type and model	[61,64]
	D3.10—Vehicles loading degree	[64]
	D3.11—Equipment of rolling stock with navigation and telecommunication systems	[10]
	D3.12—Eco-driving	[41,89]
Cumulative element	D4.1—Eco-raw materials and materials for warehouse construction	[72]
	D4.2—Warehouse type	[40,83,86]
	D4.3—Spatial organization of warehouse facilities	[44]
	D4.4—Energy-saving technologies	[4,22,33,36,44,47,49,69,72,73,104,105]
	D4.5—Environment protection systems	[20,35,61,64]
	D4.6—Eco-friendly loading and unloading equipment	[19,35,44,64]
	D4.7—Mechanization and automation of loading and unloading operations	[15,29,44,102,105]
	D4.8—Inventory management system	[9,12,32]
	D4.9—Placement and storage of products and waste	[11,12,16,20,36,44,47,50,52,53,58,66,72]
	D4.10—Type of packaging for products	[36,46,52,53,61,64,68,73,94,102]
	D4.11—Labor conditions	[19,59]
Output element	D5.1—Eco-marketing	[4,16,20,22,27,28,30,33,35,36,41,43,46,47,48,49,58,59,63,64,65,67,68,69,72,75,76,77,81,96,99,101,105]
	D5.2—Eco-friendly sales channels	[10,11,16,17,22,27,33,46,48,50,51,55,57,58,59,64,72,75,76,94,96,97,102]
	D5.3—Tare and packaging return system	[12,16,27,36,50,60,61,64,99]
	D5.4—Type of packaging for products	[12,16,36,46,52,53,61,64,68,73]
	D5.5—Eco-labeling of products	[15,29,46,58,59,60]
	D5.6—E-commerce with consumer	[62,75]
Control element	D6.1—Environment strategy	[4,8,10,11,16,19,20,22,23,27,28,30,31,32,33,35,36,41,42,46,47,48,49,50,52,53,55,56,57,58,61,63,65,66,68,70,71,73,74,75,76,77,78,79,80,81,98,101,102]
	D6.2—Environmental audit	[16,20,22,36,48,50,59,66,95,98,102,103]
	D6.3—Corporate Information Systems	[8,10,11,15,16,17,23,29,31,33,43,47,50,54,56,59,62,63,65,66,70,71,74,75,76,79,80,81,94,95,100,102,103]
	D6.4—Information and communication technologies	[3,9,10,12,15,23,27,29,30,33,35,42,46,50,57,58,59,62,63,66,70,71,72,73,75,76,78,79,80,81,97,100,103]
	D6.5—Intelligent Transportation Systems	[21,83,89]
	D6.6—Corporate social responsibility	[3,4,8,11,15,17,19,20,22,23,28,29,32,35,36,41,46,47,50,51,52,53,54,55,56,58,59,63,64,65,66,72,74,76,78,79,95,96,97,98,100,101,102,103,105]
	D6.7—Return and reverse flow management systems	[4,12,16,27,28,36,46,47,49,52,53,56,60,61,65,73,77,94,96,99,101]
	D6.8—Recycling processes for waste, packaging, finished products	[11,12,16,20,22,27,28,30,31,41,44,46,47,50,52,53,58,61,64,66,67,68,69,72,77,96,97,99,105]

presents the results of the systematization of SC sustainability drivers. As the main feature of systematization, we used the supporting functions of the SC elements.

3.3. Sustainable Supply Chain Drivers’ Assessment Criteria System

It is proposed to evaluate SSC drivers using a two-level system of criteria developed by the authors [111]. The system of criteria is based on the systematization of research and practice of using different indicators for assessing logistics flows and individual elements of supply chains to achieve compliance with the goals and principles of sustainable development. The main distinguishing feature of the proposed system is the comprehensiveness of the assessment of all logistics flows and SSC elements for compliance with the concept of sustainable development. The system includes five groups of criteria of the first level and fifteen sub-criteria of the second level (

Table 3. Sustainable supply chain driver assessment criteria system [111].

Criteria	Characteristic	Sub-Criteria	Characteristic
Economic criteria (C1)	The efficiency of using all types of resources in the SSC	Profit (C1.1)	Difference between total revenue and operating costs
		Operating expenses (C1.2)	The sum of all costs associated with converting investments into profits
		Fixed investment (C1.3)	Cash flow for the formation of fixed assets
Energy–ecological criteria (C2)	The efficiency of energy use during the movement of logistics flows and their impact on the environment	The energy intensity (C2.1)	The amount of energy spent on the movement of the logistics flow
		Greenhouse gas emissions of CO2 (C2.2)	The total volume of greenhouse gas emissions from all sources involved in the movement of the logistics flow
Quality criteria (C3)	The safety and timeliness of movement and processing of logistics flows, as well as the quality of their management	Safety of cargo transportation (C3.1)	Comprehensive indicator of the material flow movement without damage, pollution, or loss
		Timeliness of cargo transportation (C3.2)	Comprehensive indicator of the material flow movement by the appointed date, regularly, or at the required speed
		The coefficient of flow controllability (C3.3)	The ratio of the number of information messages on compliance with the indicators of safety and timeliness of transportation to the total number of management decisions
Statistical criteria (C4)	The patterns of change in the controlled sub-criteria of logistics flows	The coefficient of flow irregularity (C4.1)	Deviation of the logistics flows’ physical parameters of from their average values
		The coefficient of complexity structure of flow (C4.2)	The number of streams within the logistic flow
		The coefficient of flow discreteness (C.4.3)	The number of elements of the logistic flow in the stream
		The coefficient of differentiability of flow (C4.4)	Changing the structure of the logistics flow (number of streams) in the process of movement
Flow’s physical criteria (C5)	The intensity of logistics flows and their spatio-temporal changes	The mass (quantity) of flow (C5.1)	The total number of elements in the logistics flow
		The speed of flow (C5.2)	The speed of movement of the logistics flow elements
		The length of the route (C5.3)	Distance traveled by a logistic flow element while moving along a route

).

3.4. A Framework for Sustainable Supply Chain Drivers’ Multi-Criteria Assessment

The two-level criteria system presented in the previous section is proposed to be used to evaluate SSC drivers using multi-criteria methods. The flow chart of the developed assessment framework is presented in Figure 4. This framework is universal and can be applied both for evaluating the set of drivers identified by the authors and for any other combination of drivers. The combination of drivers depends on the structure of a particular SSC as well as the specific elements that comprise it.

Stage I. A specific supply chain is formalized as a model consisting of certain logistic elements. The boundaries of the supply chain are defined. Management objectives for achieving sustainability goals are selected. The supply chain structure [106] and management objectives determine the composition of indicators (criteria) [111] for assessing and selecting SSC drivers (alternatives). Scientific literature and best practices of implementing green principles and technologies are sources of information for the justification and systematization of SSC drivers and criteria for their evaluation. Supporting functions of SC elements [110] are recommended to be used as the main feature of systematization. The system of driver evaluation criteria should meet the requirements of comprehensiveness of logistics flow assessment and consistency of logistics flows [111].

We recommend using the criteria and drivers’ systems presented in Section 3.2 and Section 3.3, as they are considered the most comprehensive and universal systems. Systems with a smaller number of criteria and drivers can be used for supply chains with fewer logistics elements or sustainability objectives.

Stage II. Expert scoring of the identified criteria, sub-criteria, and drivers. Expert assessment is performed using a standardized methodology. The methodology of expert scoring includes the formation of an expert group of at least five people [54], development of questionnaires, filling out questionnaires, and verification of the data obtained. The mean and standard deviation of the responses as well as Cronbach’s alpha coefficient can be calculated to check the reliability and consistency of the results [26].

Stage III. Ranking of criteria, sub-criteria, and drivers. The weight of criteria and sub-criteria is calculated using the DEcision MAking Trial and Evaluation Laboratory (DEMATEL) method. This method is a decision support tool developed in the 1970s by the Science and Humanitarian Affairs Program of the Battelle Memorial Institute of Geneva [112]. This method was chosen because it allows for the investigation of the causal relationships between the criteria and sub-criteria of a sustainable supply chain driver assessment system. This allows for the identification of key criteria that have the greatest impact and are influenced by other criteria. This is difficult to achieve using other methods. Furthermore, DEMATEL allows the use of subjective data, i.e., expert opinions. This is appropriate to use in the chosen field where there may not be clear quantitative data in the evaluation process due to the heterogeneous nature of the criteria. This is confirmed by the analysis that is presented in Section 2 (see Figure 2). Recently, DE-MATEL has been applied to assess the factors, barriers, and drivers of green supply chain sustainability in the forms of DEMATEL [4,19,44,66,84,92,104], Fuzzy DE-MATEL [26,54,61,67,79], and Grey DEMATEL [27,39,43], as well as in combination with other methods such as AHP [30,47], Fuzzy EDAS-Fuzzy COPRAS [77], Fuzzy ANP-Fuzzy TOPSIS [90], and ISM-MICMAC [96].

Our research has shown that the accuracy of the ranking results depends on the composition and size of the expert group. Therefore, we recommend using different scales to assess the importance of criteria and sub-criteria and then comparing the obtained results. Crisp, Fuzzy, and Grey number scales are used in the present study. The correspondence of these scales to the experts’ linguistic scores is presented in

Table 4. Scales for scoring criteria and alternatives.

Linguistic Variables	Scale Numbers
	Crisp	Fuzzy	Grey
No influence (N)	0	[0,0,0]	[0,0]
Low influence (L)	1	[0,1,2]	[0,1]
Medium influence (M)	2	[1,2,3]	[1,2]
High influence (H)	3	[2,3,4]	[2,3]
Very high influence (VH)	4	[3,4,4]	[3,4]

[77,113,114,115].

The scores obtained using the three scales are used to rank the criteria and sub-criteria using the C-DEMATEL, F-DEMATEL, and G-DEMATEL methods, respectively.

The combination of DEMATEL with fuzzy set theory and gray systems’ theory provides the opportunity for decision making considering uncertain, ambiguous, or incomplete information and, thus, provides more reliable analysis and evaluation results. The main steps of DEMATEL using Crisp, Fuzzy, and Grey are presented in

Table 5. DEMATEL’s basic steps.

Steps	C-DEMATEL	F-DEMATEL	G-DEMATEL
III.1. Construction of the initial matrix of direct links between the criteria	$C=\left[\begin{array}{ccccc}{a}_{11}& \cdots & a_{1j}& \cdots & a_{1n}\\ ⋮& & ⋮& & ⋮\\ a_{i1}& \cdots & a_{ij}& \cdots & a_{in}\\ ⋮& & ⋮& & ⋮\\ a_{n1}& \cdots & a_{nj}& \cdots & a_{nn}\end{array}\right],$ where C—initial matrix of direct relations, aij—degree of influence of i-th criterion on j-th criterion	$\tilde{F}=\left[\begin{array}{ccccc}\left[0,0,0\right]& \dots & {\tilde{f}}_{0i}& \dots & {\tilde{f}}_{1n}\\ ⋮& \left[0,0,0\right]& ⋮& \dots & ⋮\\ {\tilde{f}}_{i1}& \dots & \left[0,0,0\right]& \dots & {\tilde{f}}_{in}\\ ⋮& \dots & ⋮& \left[0,0,0\right]& ⋮\\ {\tilde{f}}_{n1}& \dots & {\tilde{f}}_{nj}& \dots & \left[0,0,0\right]\end{array}\right],$ where $\tilde{F}$—initial fuzzy matrix of direct relations, ${\tilde{f}}_{ij}=(f_{ij1},f_{ij2},f_{ij3})$—degree of influence of i-th criterion on j-th criterion, represented by triangular fuzzy numbers	$G=\left[\begin{array}{ccccc}\left[0,0\right]& \dots & \otimes g_{0i}& \dots & \otimes g_{1n}\\ ⋮& \left[0,0\right]& ⋮& \dots & ⋮\\ \otimes g_{i1}& \dots & \left[0,0\right]& \dots & \otimes g_{in}\\ ⋮& \dots & ⋮& \left[0,0\right]& ⋮\\ \otimes g_{n1}& \dots & \otimes g_{nj}& \dots & \left[0,0\right]\end{array}\right],$ where G—the initial gray matrix of direct relations, ⊗gij—the gray number showing the degree of influence of the i-th criterion on the j-th criterion. If $\overline{g_{ij}}$ is the upper boundary of the gray number and $\underset{¯}{g_{ij}}$ is its lower boundary, then $\otimes g_{ij}=\left[\underset{¯}{g_{ij}},\overline{g_{ij}}\right]$
III.2. Normalization of the direct relations matrix	$X_c=C\lambda$, $\lambda =\min \left[{\displaystyle \frac{1}{\underset{1\le i\le n}{\max }{\displaystyle \sum _{j=1}^n{a}_{ij}}}},{\displaystyle \frac{1}{\underset{1\le j\le n}{\max }{\displaystyle \sum _{i=1}^n{a}_{ij}}}}\right],$ where $X_c$—normalized matrix of direct relations	${\tilde{X}}_f={\displaystyle \frac{\tilde{F}}{r}},$ ${\tilde{X}}_f=\left[\begin{array}{ccccc}\left[0,0,0\right]& \dots & {\tilde{x}}_{0i}& \dots & {\tilde{x}}_{1n}\\ ⋮& \left[0,0,0\right]& ⋮& \dots & ⋮\\ {\tilde{x}}_{i1}& \dots & \left[0,0,0\right]& \dots & {\tilde{x}}_{in}\\ ⋮& \dots & ⋮& \left[0,0,0\right]& ⋮\\ {\tilde{x}}_{n1}& \dots & {\tilde{x}}_{nj}& \dots & \left[0,0,0\right]\end{array}\right],$ where ${\tilde{X}}_f$—normalized fuzzy matrix of direct relations, $r=\underset{1\le i\le n}{\max }{\displaystyle \sum _{j=1}^n{f}_{ij3}},$ It is considered that there is at least one value i such that ${\displaystyle \sum _{j=1}^n{f}_{ij3}}>r,$ where 3—number of values defining the fuzzy number	$X_g=\otimes s\cdot G$$\otimes s=\left[\underset{¯}{s},\overline{s}\right]={\displaystyle \frac{1}{\underset{1\le i\le m}{\max }{\displaystyle \sum _{j=1}^n\otimes g_{ij}}}},i,j=1,2,\dots ,n,$ where $X_g$—normalized matrix of direct relations; $\underset{¯}{s},\overline{s}$—, the lower and upper limits of the gray number, respectively; n—criteria number
III.3. Calculation of the total matrix of direct relations	$T=\underset{k\to \infty }{\lim }(X+X^2+\dots +X^k)=X{(1-X)}^{-1},$ where $T$—total matrix of direct relations between criteria	${\tilde{T}}_f=\underset{k\to \infty }{\lim }\left({\tilde{X}}_f^1+{\tilde{X}}_f^2+\dots +{\tilde{X}}_f^k\right)={\tilde{X}}_f{(1-{\tilde{X}}_f)}^{-1},$ then ${\tilde{T}}_f=\left[\begin{array}{ccccc}\left[0,0,0\right]& \dots & {\tilde{t}}_{0i}& \dots & {\tilde{t}}_{1n}\\ ⋮& \left[0,0,0\right]& ⋮& \dots & ⋮\\ {\tilde{t}}_{i1}& \dots & \left[0,0,0\right]& \dots & {\tilde{t}}_{in}\\ ⋮& \dots & ⋮& \left[0,0,0\right]& ⋮\\ {\tilde{t}}_{n1}& \dots & {\tilde{t}}_{nj}& \dots & \left[0,0,0\right]\end{array}\right],$ where ${\tilde{T}}_f$—total fuzzy matrix of direct relations between criteria; ${\tilde{t}}_{ij}=(t_{ij1},t_{ij2},t_{ij3})$—fuzzy numbers of the total relational matrix, $T_{f1}={\left[t_{ij1}\right]}_{n\times n}=X_{f1}{\left(I-X_{f1}\right)}^{-1}$,$T_{f2}={\left[t_{ij2}\right]}_{n\times n}=X_{f2}{\left(1-X_{f2}\right)}^{-1},$$T_{f3}={\left[t_{ij3}\right]}_{n\times n}=X_{f3}{\left(1-X_{f3}\right)}^{-1}$	$T_g=X_g{\left(I-X_g\right)}^{-1},$ where $T_g$—total matrix of direct relations between criteria, I—single matrix
III.4. Calculation of the number of relationships (Di+Ri) and forces of influence (Di–Ri) between criteria	The sum of rows and the sum of columns are denoted, respectively, as vectors D and R in the total relation matrix T and are calculated by the following formulas $T={\left[t_{ij}\right]}_{n\times n},{\displaystyle } i,j=1,2,\dots ,n,$$D={\left[{\displaystyle \sum _{j=1}^n{t}_{ij}}\right]}_{n\times 1}={\left[t_i\right]}_{n\times 1},$$R={\left[{\displaystyle \sum _{i=1}^n{t}_{ij}}\right]}_{1\times n}={\left[t_j\right]}_{1\times n}$	Compute $\left({\tilde{D}}_i+{\tilde{R}}_j\right)$ and $\left({\tilde{D}}_i-{\tilde{R}}_j\right)$, where ${\tilde{D}}_i$ and ${\tilde{R}}_j$ are the sum of the rows and the sum of the columns in the overall fuzzy relations matrix ${\tilde{T}}_f$. Then the fuzzy numbers are converted to absolute values [116,117].	The sums of rows ⊗Ri and the sums of columns ⊗Di in the total relation matrix Tg are calculated by the following formulas: $\otimes R_i={\left[\otimes R_i\right]}_{n\times 1}={\left[{\displaystyle \sum _{j=1}^n\otimes t_{ij}}\right]}_{n\times 1},$$\otimes D_j={\left[\otimes D_j\right]}_{1\times n}={\left[{\displaystyle \sum _{j=1}^n\otimes t_{ij}}\right]}_{1\times n}$
III.5. Building a network relationship map of the criteria. Ranking of criteria according to the values of weighting coefficients calculated based on the results of the assessment of the number of interrelationships (Di + Ri) and the strength of influence (Di − Ri) between the criteria

.

The three versions of the criteria and sub-criteria scores are further used to rank the drivers in combination with the CRADIS method—Compromise Ranking of Alternatives from Distance to Ideal Solution. CRADIS is a rather new method, proposed in 2021 in [118]. The CRADIS method determines the best alternatives in a more comprehensive and simple way, utilizing the advantages of ARAS (Additive Ratio ASsessment), MARCOS (Measurement of Alternatives and Ranking according to Compromise Solution), and TOPSIS (Technique for the Order of Preference by Similarity to Ideal Solution) methods, while reducing their disadvantages [119,120]. At the same time, CRADIS is different from TOPSIS, MARCOS, and ARAS. Unlike solutions based on individual criteria, CRADIS defines a single ideal and anti-ideal solution for all alternatives and criteria, preserving the values of the criteria weights [120]. The CRADIS method has been applied to the evaluation of sustainable suppliers [121], green contractors [122], selection of blockchain technology in logistics companies [123], locations of distribution centers [124], risk assessment in supply chains [120], and more.

The calculations using CRADIS are performed in the following steps.

Step III.6. Formation of an initial decision-making matrix consisting of a set of n criteria and m alternatives:

$$X=\left[\begin{array}{cccc}{x}_{11}& x_{12}& \cdots & x_{1n}\\ x_{21}& x_{22}& \dots & x_{2n}\\ ⋮& ⋮& \ddots & ⋮\\ x_{m1}& x_{m2}& \dots & x_{mn}\end{array}\right],$$

(1)

where x_mn—assessment of i-th alternative; i = 1, 2, …, m according to the j-th criterion; and j = 1,2, …, n.

Step III.7. Normalization of the initial decision matrix X using the following formulas:

$$N={\left[n_{ij}\right]}_{m\times n},i=1,2,\dots ,m;j=1,2,\dots ,n,$$

(2)

$$n_{ij}={\displaystyle \frac{x_{ij}}{x_{j\max }}},ifj\in C$$

(3)

$$n_{ij}={\displaystyle \frac{x_{j\min }}{x_{ij}}},ifj\in B$$

(4)

where C—cost criteria and B—benefit criteria.

Step III.8. Calculation of the weighted matrix V by multiplying the values of the normalized matrix elements n_ij by the corresponding weight coefficients w_j of the criteria:

$$V={\left[v_{ij}\right]}_{m\times n},i=1,2,3\dots ,m; j=1,2,3\dots ,n,$$

(5)

$$V=n_{ij}\times w_j,$$

(6)

$${\displaystyle \sum _{i=1}^m{w}_j}=1.$$

(7)

Step III.9. Calculation of ideal and anti-ideal solutions. Ideal solutions t_i are the largest values of v_ij in the weighted matrix of solutions, and anti-ideal solutions tai, respectively, are the smallest values.

$$t_i=\max v_{ij},$$

(8)

$$t_{ai}=\min v_{ij},$$

(9)

Step III.10. Calculation of deviations from ideal and anti-ideal solutions:

$$d^{+}=\max t_i-v_{ij},$$

(10)

$$d^{-}=v_{ij}-\min t_{ai}.$$

(11)

Step III.11. Calculation of the degree of deviation of alternatives from ideal and anti-ideal solutions:

$$s_i^{+}={\displaystyle \sum _{j=1}^n{d}^{+}},$$

(12)

$$s_i^{-}={\displaystyle \sum _{j=1}^n{d}^{-}}.$$

(13)

Step III.11. Calculation of the utility functions for each alternative regarding deviations from the ideal alternatives:

$$K_i^{+}={\displaystyle \frac{s_0^{+}}{s_i^{+}}},$$

(14)

$$K_i^{-}={\displaystyle \frac{s_i^{-}}{s_0^{-}}},$$

(15)

where $s_0^{+}$—ideal alternative with the smallest distance from the ideal solution, i.e., $s_0^{+}=\min (s_i^{+}),\forall m$; $s_0^{-}$—ideal alternative, as far as possible from the anti-ideal solution, i.e., $s_0^{-}=\max (s_i^{-}),\forall m$.

Step III.12. Calculation of average values of utility functions for alternatives Q_i and ranking of alternatives according to their values:

$$Q_i={\displaystyle \frac{K_i^{+}+K_i^{-}}{2}}.$$

(16)

The best alternative is the alternative with the maximum value of the utility function Q_i.

Stage IV. Quality assessment of ranking results and development of management decisions. The value of the Mean Relative Error of the Ranking results (MRER) of drivers obtained by different methods is proposed to be generally calculated by the following formula:

$$MRER={\displaystyle \frac{{\displaystyle \sum _{i=1}^M{\displaystyle \sum _{l=1}^{L-1}\left(\left|R_{il}-R_{il+1}\right|+\left|R_{iL}-R_{i1}\right|\right)}}}{M\times L}},$$

(17)

where R_il—rank of the i-th alternative, calculated by the l-th method; M—number of alternatives; and L—number of ranking methods, L ≥ 3.

Finally, the sensitivity of the obtained results is assessed by ranking the alternatives by different MCDMs and then calculating the Spearman coefficient. If the MRER or Spearman coefficient values exceed the acceptable values, it is necessary to increase the size of the expert group or change its composition.

The resulting SSC driver ranks are used to develop and justify decisions on the implementation of green logistics methods and instruments [108].

4. Numerical Example

4.1. Input Data

An example of driver ranking for all elements of the supply chain is presented in this section.

The studied supply chain model includes six elements: input, processing, cumulative, transportation, output, and control. The characterization of the elements is presented in Section 3.1. The SSC driver system includes 54 drivers: 8 drivers of the input element (D1.1—D1.8), 9 drivers of the processing element (D2.1—D2.9), 11 drivers of the cumulative element (D3. 1—D3.11), 12 transport element drivers (D4.1—D1.12), 6 output element drivers (D5.1—D1.6), and 8 control element drivers (D6.1—D6.8). The characterization of the drivers is presented in Section 3.2 (See Table 2). The drivers were ranked using a criteria system containing 15 attributes (Section 3.3, See Table 3).

In the second stage of the framework for ranking the identified drivers (Figure 3), we formed a panel of five experts (

Table 6. Information about experts.

№	Academic Degree	Number of Experts	Expert Science Interests	Work Experience in the Field of Research
1	Professor, doctor (Technical Science)	1	Supply chain management, transport systems	41
		1	Supply chain management, logistics	34
		1	Transport systems, logistics	18
2	Assistant Professor (PhD)	1	Supply chain management	17
		1	Transport systems, warehouse systems	17

).

Initially, the experts performed an assessment of the importance of criteria and sub-criteria of drivers in SSC. The expert questionnaire is provided in Appendix A. The linguistic scale presented in Table 4 was used for the assessment. The results of the experts’ assessment are presented in

Table 7. Results of experts’ assessment of the importance of criteria and sub-criteria *.

Criteria		C1					C2					C3					C4					C5
Experts		Ex1	Ex2	Ex3	Ex4	Ex5	Ex1	Ex2	Ex3	Ex4	Ex5	Ex1	Ex2	Ex3	Ex4	Ex5	Ex1	Ex2	Ex3	Ex4	Ex5	Ex1	Ex2	Ex3	Ex4	Ex5
Criteria	C1	N	N	N	N	N	H	H	VH	H	H	H	H	VH	VH	VH	L	VH	M	M	M	VH	H	H	M	H
	C2	VH	H	L	VH	M	N	N	N	N	N	M	M	L	H	N	L	M	N	H	N	M	H	M	M	N
	C3	VH	VH	VH	VH	H	H	VH	M	M	M	N	N	N	N	N	H	H	L	M	M	H	VH	H	L	M
	C4	L	N	L	M	M	N	N	N	M	L	L	N	N	M	VH	N	N	N	N	N	L	N	N	M	L
	C5	VH	VH	H	H	H	H	VH	M	H	VH	H	VH	H	M	L	H	VH	L	M	L	N	N	N	N	N
Sub-criteria		C1.1					C1.2					C1.3					C2.1					C2.2
Experts		Ex1	Ex2	Ex3	Ex4	Ex5	Ex1	Ex2	Ex3	Ex4	Ex5	Ex1	Ex2	Ex3	Ex4	Ex5	Ex1	Ex2	Ex3	Ex4	Ex5	Ex1	Ex2	Ex3	Ex4	Ex5
Sub-criteria	C1.1	N	N	N	N	N	M	M	VH	H	N	VH	VH	VH	VH	L	M	M	VH	L	N	L	H	VH	L	N
	C1.2	VH	VH	VH	H	VH	N	N	N	N	N	L	H	VH	H	N	M	H	VH	H	L	L	M	M	L	L
	C1.3	H	H	VH	VH	VH	H	M	VH	H	H	N	N	N	N	N	L	VH	VH	H	H	L	H	VH	H	H
	C2.1	H	VH	VH	H	L	VH	VH	VH	H	M	M	L	H	M	L	N	N	N	N	N	VH	VH	H	H	M
	C2.2	M	H	L	L	N	H	H	L	L	H	M	L	H	M	H	H	VH	N	L	L	N	N	N	N	N
	C3.1	VH	VH	H	H	VH	M	VH	M	H	H	L	L	N	L	M	L	H	VH	L	N	L	M	N	N	N
	C3.2	H	H	H	H	VH	H	H	M	H	H	L	L	N	M	L	H	VH	L	M	N	M	M	N	L	N
	C3.3	M	VH	L	L	M	H	VH	L	M	H	L	L	N	L	H	L	VH	M	L	L	M	M	N	L	M
	C4.1	M	VH	L	L	M	H	VH	L	M	H	L	L	N	L	L	L	VH	M	M	M	L	VH	N	L	N
	C4.2	L	VH	N	M	L	M	VH	M	H	M	L	L	N	L	M	M	H	M	H	L	M	H	N	M	N
	C4.3	L	VH	L	L	M	M	VH	L	M	H	L	L	N	L	M	M	VH	L	M	N	H	H	N	L	N
	C4.4	L	M	N	L	M	M	M	M	M	H	L	L	N	L	M	M	H	L	M	N	M	H	N	L	N
	C5.1	H	VH	VH	VH	VH	H	VH	VH	VH	VH	H	H	N	M	M	VH	VH	VH	H	M	VH	VH	L	VH	H
	C5.2	M	VH	VH	H	M	H	VH	L	VH	VH	M	M	N	M	VH	H	VH	M	VH	VH	M	VH	N	M	M
	C5.3	M	L	M	H	L	M	VH	M	VH	L	L	H	N	L	L	H	VH	H	VH	L	M	VH	N	H	M
Sub-criteria		C3.1					C3.2					C3.3					C4.1					C4.2
Experts		Ex1	Ex2	Ex3	Ex4	Ex5	Ex1	Ex2	Ex3	Ex4	Ex5	Ex1	Ex2	Ex3	Ex4	Ex5	Ex1	Ex2	Ex3	Ex4	Ex5	Ex1	Ex2	Ex3	Ex4	Ex5
Sub-criteria	C1.1	M	H	VH	M	N	M	H	M	M	N	M	H	M	M	N	L	H	L	M	N	L	H	N	L	N
	C1.2	H	H	M	M	M	H	H	L	H	M	L	M	L	L	L	L	L	N	M	L	L	L	N	M	N
	C1.3	M	M	H	M	VH	M	M	L	M	H	M	H	L	M	M	L	L	M	L	M	L	L	M	M	M
	C2.1	L	L	M	L	N	L	L	L	H	N	L	H	L	M	N	L	VH	N	M	N	L	H	N	H	N
	C2.2	L	N	N	L	N	L	N	N	L	N	L	N	N	L	N	L	N	N	L	N	L	N	N	L	N
	C3.1	N	N	N	N	N	M	H	N	VH	L	M	H	N	L	L	L	L	L	M	L	M	N	N	H	N
	C3.2	L	H	N	M	N	N	N	N	N	N	M	H	L	M	L	M	VH	L	H	M	L	VH	N	H	N
	C3.3	M	H	H	L	M	VH	VH	H	M	H	N	N	N	N	N	L	VH	VH	M	M	L	VH	H	H	M
	C4.1	L	H	L	M	L	H	VH	H	H	H	H	VH	L	H	L	N	N	N	N	N	M	VH	N	H	N
	C4.2	H	H	H	M	M	M	VH	H	H	H	M	VH	M	H	VH	M	VH	M	M	H	N	N	N	N	N
	C4.3	M	H	N	M	N	M	VH	L	H	N	M	VH	L	H	M	M	VH	VH	H	M	M	VH	L	H	H
	C4.4	M	M	H	L	N	M	VH	H	M	N	M	VH	M	M	M	M	VH	M	M	L	H	VH	H	M	VH
	C5.1	H	H	L	H	L	M	VH	H	H	L	H	H	L	H	L	M	VH	L	M	M	L	VH	N	VH	M
	C5.2	M	VH	N	M	N	VH	VH	VH	VH	N	H	H	M	H	M	M	H	L	M	VH	L	M	N	H	L
	C5.3	M	H	N	H	M	H	VH	VH	H	H	M	VH	L	H	N	M	VH	L	M	L	L	H	N	M	L
Sub-criteria		C4.3					C4.4					C5.1					C5.2					C5.3
Experts		Ex1	Ex2	Ex3	Ex4	Ex5	Ex1	Ex2	Ex3	Ex4	Ex5	Ex1	Ex2	Ex3	Ex4	Ex5	Ex1	Ex2	Ex3	Ex4	Ex5	Ex1	Ex2	Ex3	Ex4	Ex5
Sub-criteria	C1.1	L	H	N	L	N	L	H	N	L	N	M	VH	H	M	M	L	H	L	M	M	L	L	N	L	N
	C1.2	L	L	N	L	L	L	L	N	L	L	L	L	H	H	N	L	L	L	H	L	L	L	N	L	N
	C1.3	M	M	L	L	M	L	L	N	L	M	L	L	H	M	H	M	M	M	H	H	L	L	L	M	L
	C2.1	M	VH	N	H	N	M	H	L	H	N	H	L	VH	M	N	M	L	H	M	N	M	L	L	L	N
	C2.2	L	N	N	L	N	L	N	N	L	N	L	N	N	L	N	L	N	N	L	N	L	N	N	L	N
	C3.1	L	L	L	M	L	L	L	L	M	L	M	M	L	L	N	L	H	L	M	N	L	H	N	M	N
	C3.2	L	VH	N	H	N	L	VH	N	H	N	L	L	L	L	N	H	M	VH	L	N	M	M	VH	L	N
	C3.3	L	VH	VH	M	L	L	VH	H	M	L	L	L	M	L	N	M	H	VH	L	H	M	L	M	L	N
	C4.1	M	VH	N	M	N	L	H	L	H	L	L	L	H	L	N	M	L	M	H	L	L	N	N	L	N
	C4.2	L	H	L	H	M	M	VH	L	H	M	L	H	N	M	L	L	M	L	H	M	L	L	L	M	N
	C4.3	N	N	N	N	N	L	M	L	H	L	L	L	H	H	N	H	M	L	H	M	M	M	N	L	N
	C4.4	L	M	M	M	M	N	N	N	N	N	L	L	L	M	L	L	H	M	M	L	M	M	L	L	N
	C5.1	H	VH	L	M	M	M	H	L	H	M	N	N	N	N	N	VH	H	M	H	VH	L	N	N	L	H
	C5.2	H	M	M	M	L	L	H	M	H	L	L	N	N	H	N	N	N	N	N	N	L	N	N	L	N
	C5.3	M	M	N	M	N	M	M	N	M	N	L	N	N	M	M	VH	M	M	H	L	N	N	N	N	N

* N—No influence; L—Low influence; M—Medium influence; H—High influence; VH—Very high influence.

. The evaluation data were used as input data for the DEMATEL method. Then, the experts assessed the impact of drivers on criteria and sub-criteria on a five-point scale: 1—very low, 2—low, 3—medium, 4—high, and 5—very high. The final evaluation results for all five experts were calculated as a geometric mean and are presented in

Table 8. Averaged results of SSC driver’s expert assessment.

Supply Chain Elements	Drivers	Sub-Criteria
		C1.1	C1.2	C1.3	C2.1	C2.2	C3.1	C3.2	C3.3	C4.1	C4.2	C4.3	C4.4	C5.1	C5.2	C5.3
Input element	D1.1	3.10	4.00	1.78	2.70	3.76	1.00	1.00	1.15	1.15	2.70	2.05	2.64	2.17	1.89	2.76
	D1.2	3.81	4.32	2.05	3.59	4.32	1.52	1.52	1.43	1.32	2.70	2.17	2.49	3.29	2.55	2.72
	D1.3	4.13	4.37	2.41	3.29	3.10	3.90	4.32	4.32	4.13	3.37	3.57	3.57	2.86	3.10	2.86
	D1.4	2.55	2.70	1.74	2.22	3.64	2.35	2.40	2.09	1.89	2.22	2.05	2.17	1.64	2.17	4.18
	D1.5	3.90	4.32	1.74	3.52	3.13	3.57	4.32	4.18	3.81	2.86	3.37	3.00	2.17	2.76	4.78
	D1.6	3.10	4.00	2.35	2.70	4.13	1.64	1.32	1.64	1.52	2.70	2.49	2.49	2.55	2.35	2.14
	D1.7	2.05	2.55	1.89	2.00	1.82	1.89	1.43	1.52	1.43	1.93	1.32	1.32	1.32	1.43	1.55
	D1.8	2.70	3.10	2.70	2.35	2.05	2.93	3.29	2.95	2.40	2.17	1.89	2.05	1.52	2.14	1.43
Processing element	D2.1	3.39	4.57	2.61	3.47	3.31	1.38	1.15	1.15	1.52	2.35	1.93	2.05	2.05	2.17	1.89
	D2.2	3.13	4.32	2.35	3.39	4.32	1.15	1.32	1.32	1.64	2.22	2.05	2.05	2.76	2.05	2.27
	D2.3	3.10	3.57	3.98	3.59	4.78	1.74	1.74	1.78	1.89	1.78	1.78	1.64	2.35	2.22	2.05
	D2.4	3.29	3.95	3.98	3.98	4.32	1.52	1.52	1.43	1.64	1.52	1.43	1.64	2.49	2.17	2.05
	D2.5	2.86	3.18	3.98	3.59	5.00	1.15	1.32	1.32	1.32	1.43	1.32	1.32	1.52	1.89	1.89
	D2.6	2.55	3.18	3.64	3.39	4.51	1.52	1.15	1.64	1.52	1.32	1.15	1.52	1.64	1.74	1.74
	D2.7	3.68	3.90	3.18	3.44	4.08	1.15	1.32	2.55	2.22	2.70	2.22	2.70	3.29	2.49	2.76
	D2.8	2.35	3.95	2.83	3.10	2.86	1.52	1.52	1.78	1.78	1.74	1.64	1.64	1.52	1.52	1.52
	D2.9	2.70	3.57	1.32	2.83	3.13	1.32	1.32	2.00	1.74	1.52	1.52	1.64	1.52	1.64	1.32
Transport element	D3.1	1.74	3.10	4.37	2.72	2.40	1.15	1.15	1.15	1.15	1.15	1.15	1.15	1.32	1.15	1.15
	D3.2	2.49	3.95	4.57	3.81	3.39	2.27	2.05	2.09	2.00	1.89	1.78	1.64	2.35	2.35	1.78
	D3.3	1.74	2.70	4.08	2.46	2.83	1.43	1.74	1.74	1.52	1.52	1.52	1.52	2.17	2.49	1.82
	D3.4	3.25	3.68	4.13	4.37	4.32	1.52	1.32	1.43	1.15	1.15	1.15	1.15	1.43	1.32	1.15
	D3.5	2.00	2.83	3.95	3.37	3.95	1.74	1.43	1.74	1.32	1.15	1.15	1.15	1.15	1.15	1.15
	D3.6	2.35	3.06	3.95	3.73	4.78	1.58	1.52	1.52	1.43	1.52	1.32	1.32	1.43	1.43	1.32
	D3.7	3.59	4.08	4.37	3.18	3.10	3.13	3.25	3.13	2.99	2.22	2.35	2.05	1.74	3.59	1.82
	D3.8	4.32	3.90	2.40	4.13	3.13	3.44	3.98	3.98	3.98	3.81	3.81	3.44	3.73	3.98	2.40
	D3.9	2.49	2.86	2.35	2.49	2.40	3.17	2.61	2.27	2.61	2.61	2.46	2.61	2.35	2.30	2.17
	D3.10	2.09	2.67	1.78	2.70	3.64	1.52	1.64	1.74	1.32	1.32	1.32	1.32	1.32	1.32	1.15
	D3.11	2.49	2.83	3.68	3.95	3.39	3.68	1.32	1.32	1.52	1.15	1.15	1.15	1.32	1.32	1.15
	D3.12	3.39	3.68	3.39	4.32	4.51	2.55	2.70	2.86	2.49	2.49	2.30	2.17	1.89	2.55	2.72
Cumulate element	D4.1	2.99	2.99	2.61	3.17	3.44	2.64	2.55	2.83	2.22	2.49	2.35	2.17	1.52	2.93	2.67
	D4.2	3.25	3.52	1.89	3.52	3.73	2.67	3.90	3.44	3.44	2.27	2.61	2.37	1.78	4.13	3.98
	D4.3	2.99	3.64	2.41	3.29	2.86	2.22	3.44	4.08	3.31	3.47	3.31	3.47	3.31	3.47	3.10
	D4.4	2.17	3.59	2.05	3.57	3.10	1.89	3.73	3.10	3.59	3.25	3.17	2.70	3.57	3.81	2.00
	D4.5	2.35	2.70	3.02	3.31	3.90	1.64	1.64	1.32	1.32	1.43	1.32	1.32	1.15	1.89	1.32
	D4.6	3.31	4.08	2.27	3.39	3.59	1.58	1.58	1.82	1.58	1.52	1.52	1.52	1.32	1.82	1.58
	D4.7	2.86	3.25	3.31	4.08	4.57	2.17	2.35	1.89	2.00	1.74	2.00	1.89	1.74	2.35	1.52
	D4.8	2.35	2.70	3.31	3.59	4.51	1.64	1.78	1.74	1.74	1.89	1.74	1.52	1.52	2.55	1.64
	D4.9	3.29	3.44	2.51	3.90	4.32	1.97	1.97	1.64	2.30	2.27	2.17	1.78	2.40	2.14	1.64
	D4.10	2.35	2.99	1.64	3.18	3.10	1.74	2.70	3.17	2.35	2.00	1.52	1.89	1.52	3.31	1.58
	D4.11	2.40	3.10	2.05	3.44	3.52	2.00	2.49	3.03	2.17	2.14	1.82	1.82	1.58	3.44	1.55
Output element	D5.1	2.46	2.70	1.55	2.22	1.89	1.32	1.32	2.05	1.78	1.64	1.55	1.43	1.52	1.89	1.74
	D5.2	3.06	3.25	2.30	3.78	4.13	2.83	3.18	3.52	2.35	2.22	2.30	2.30	1.52	2.55	3.37
	D5.3	2.86	3.57	2.35	3.64	3.44	2.17	1.52	2.70	2.35	2.61	1.89	2.00	2.64	2.05	2.61
	D5.4	2.86	3.52	2.17	3.29	3.90	2.99	2.05	2.35	1.89	2.00	1.64	1.74	2.86	1.89	1.43
	D5.5	1.89	2.17	1.89	2.05	3.68	1.64	1.52	1.52	1.32	1.32	1.32	1.32	1.64	1.89	1.52
	D5.6	2.55	2.64	2.35	2.70	2.49	2.35	3.57	3.81	2.35	2.55	2.05	1.89	1.52	3.13	1.89
Control element	D6.1	2.35	2.30	2.14	3.03	3.81	1.64	1.52	1.93	1.32	1.25	1.15	1.32	1.78	1.55	1.89
	D6.2	2.05	2.77	1.64	2.64	3.59	1.64	1.52	1.89	1.32	1.15	1.15	1.32	1.64	1.43	1.64
	D6.3	3.52	3.25	3.73	3.10	2.49	2.83	3.52	3.98	3.44	2.70	2.55	2.70	2.70	3.37	1.89
	D6.4	3.44	3.73	4.08	3.29	3.29	3.52	3.73	3.76	3.59	3.44	3.03	3.03	2.55	3.90	2.86
	D6.5	3.73	3.68	4.13	3.73	3.29	3.73	4.13	4.32	3.59	3.13	3.17	3.17	2.35	4.32	3.73
	D6.6	2.35	2.35	1.32	2.17	2.93	1.89	1.74	1.64	1.52	1.43	1.43	1.43	1.15	1.15	1.15
	D6.7	3.52	3.73	2.64	3.81	4.13	1.52	1.32	2.22	2.27	2.40	1.97	2.09	2.99	1.78	2.27
	D6.8	3.10	3.95	2.30	3.48	3.90	1.64	1.32	1.78	2.05	2.40	2.17	2.00	2.61	2.05	2.35

. These data were used as input data for the CRADIS method.

4.2. Results of DEMATEL and CRADIS Methods

The initial matrix of direct relations of criteria and sub-criteria of SSC driver evaluation based on the results of the expert evaluation was constructed (Table 7). A fragment of such matrices for different DEMATEL models is presented in

Table 9. Fragment of the initial matrix of direct links of SSC driver evaluation criteria.

Criteria	C1	C2	C3	C4	C5
Crisp DEMATEL
C1	0	3.2	3.6	2	3
C2	2.8	0	1.6	1.2	1.8
C3	3.8	2.6	0	2.2	2.6
C4	1.2	0.6	1.4	0	0.8
C5	3.4	3.2	2.6	2.2	0
Fuzzy DEMATEL
C1	[0; 0; 0]	[2.2; 3.2; 4]	[2.6; 3.6; 4]	[1; 2; 3]	[2; 3; 3.8]
C2	[1.8; 2.8; 3.4]	[0; 0; 0]	[0.8; 1.6; 2.4]	[0.6; 1.2; 1.8]	[1; 1.8; 2.6]
C3	[2.8; 3.8; 4]	[1.6;2.6; 3.4]	[0; 0; 0]	[1.2; 2.2; 3.2]	[1.6; 2.6; 3.4]
C4	[0.4; 1.2; 2]	[0.2; 0.6; 1]	[0.8; 1.4; 2]	[0; 0; 0]	[0.2; 0.8; 1.4]
C5	[2.4; 3.4; 4]	[2.2; 3.2; 3.8]	[1.6; 2.6; 3.4]	[1.2; 2.2; 3]	[0; 0; 0]
Grey DEMATEL
C1	[0; 0]	[2.2; 3.2]	[2.6; 3.6]	[1; 2]	[2.0; 3.0]
C2	[1.8; 2.8]	[0; 0]	[0.8; 1.6]	[0.6; 1.6]	[1.0; 1.8]
C3	[2.8; 3.8]	[1.6; 2.6]	[0; 0]	[1.2; 2.2]	[1.6; 2.6]
C4	[0.4; 1.2]	[0.2; 0.6]	[0.8; 1.4]	[0; 0]	[0.2; 0.8]
C5	[2.4; 3.4]	[2.2; 3.2]	[1.6; 2.6]	[1.2; 2.2]	[0; 0]

.

The normalized direct linkage matrix of the SSC driver evaluation criteria is presented in

Table 10. Normalized matrix of direct links.

Criteria	C1	C2	C3	C4	C5
	Crisp DEMATEL (Xc)
C1	0	0.2712	0.3051	0.1695	0.2542
C2	0.2373	0	0.1356	0.1017	0.1525
C3	0.3220	0.2203	0	0.1864	0.2203
C4	0.1017	0.0508	0.1186	0	0.0678
C5	0.2881	0.2712	0.2203	0.1864	0
	Fuzzy DEMATEL (Xf)
C1	(0; 0; 0)	(0.1433; 0.2083; 0.2600)	(0.1700; 0.2350; 0.2600)	(0.0650; 0.1300; 0.1950)	(0.1292; 0.1942; 0.2467)
C2	(0.1158; 0.1808; 0.2200)	(0; 0; 0)	(0.0517; 0.1033; 0.1683)	(0.0392; 0.0775; 0.1425)	(0.0642; 0.1158; 0.1808)
C3	(0.1817; 0.2467; 0.2600)	(0.1025; 0.1675; 0.2200)	(0; 0; 0)	(0.0767; 0.1417; 0.2067)	(0.1025; 0.1675; 0.2200)
C4	(0.0267; 0.0792; 0.1442)	(0.0133; 0.0400; 0.0925)	(0.0533; 0.0925; 0.1442)	(0; 0; 0)	(0.0133; 0.0525; 0.1175)
C5	(0.1550; 0.2200; 0.2600)	(0.1425; 0.2075; 0.2467)	(0.1025; 0.1675; 0.2200)	(0.0758; 0.1408; 0.1933)	(0; 0; 0)
	Grey DEMATEL (Xg)
C1	(0; 0)	(0.186; 0.271)	(0.220; 0.305)	(0.085; 0.169)	(0.169; 0.254)
C2	(0.153; 0.237)	(0; 0)	(0.068; 0.136)	(0.051; 0.102)	(0.085; 0.153)
C3	(0.237; 0.322)	(0.136; 0.220)	(0; 0)	(0.102; 0.186)	(0.136; 0.220)
C4	(0.034; 0.102)	(0.017; 0.051)	(0.068; 0.119)	(0; 0)	(0.017; 0.068)
C5	(0.203; 0.288)	(0.186; 0.271)	(0.136; 0.220)	(0.102; 0.186)	(0; 0)

for each DEMATEL model.

The results of calculating the number of relationships and the strength of influence between the criteria are presented in

Table 11. Results of calculation of the number of interrelationships and strength of influence of criteria.

Criteria	Crisp DEMATEL
	D	R	D + R	D − R
C1	4.7272	4.4452	9.1724	0.2819
C2	3.2426	3.9660	7.2086	−0.7234
C3	4.5289	3.8167	8.3457	0.7122
C4	1.8302	3.1622	4.9924	−1.3321
C5	4.5466	3.4854	8.0321	1.0612
Fuzzy DEMATEL
	D	R	D + R	D − R
C1	(0.8106; 1.9120; 4.8102)	(0.7579; 1.7981; 4.4548)	(1.5686; 3.7101; 9.2650)	(−3.6442; 0.1140; 4.0523)
C2	(0.4593; 1.2799; 3.7697)	(0.6527; 1.5931; 4.2086)	(1.1120; 2.8729; 7.9783)	(−3.7493; −0.3132; 3.1169)
C3	(0.7495; 1.8194; 4.5742)	(0.6154; 1.5343; 4.0834)	(1.3649; 3.3537; 8.6576)	(−3.3339; 0.2851; 3.9587)
C4	(0.1845; 0.7311; 2.7448)	(0.4179; 1.2599; 3.8177)	(0.6024; 1.910; 6.5626)	(−3.6332; −0.5288; 2.3269)
C5	(0.7577; 1.8298; 4.6375)	(0.5176; 1.3868; 3.9718)	(1.2753; 3.2166; 8.6093)	(−3.2141; 0.4430; 4.1199)
Grey DEMATEL
	D	R	D + R	D − R
C1	(1.3040; 4.7272)	(1.2202; 4.4452)	(2.5242; 9.1724)	(−3.1417; 3.5070)
C2	(0.7567; 3.2426)	(1.0576; 3.9660)	(1.8143; 7.2086)	(−3.2093; 2.1849)
C3	(1.2189; 4.5289)	(0.9936; 3.8167)	(2.2125; 8.3457)	(−2.5978; 3.5353)
C4	(0.2961; 1.8302)	(0.6836; 3.1622)	(0.9797; 4.9924)	(−2.8661; 1.1466)
C5	(1.2288; 4.5466)	(0.8495; 3.4854)	(2.0783; 8.0320)	(−2.2566; 3.6971)

.

The conversion of the final fuzzy data into crisp values was performed using the CFCS method proposed in [116,117] for defuzzification.

Total matrices of links between criteria and network relationship maps obtained as a result of the DEMATEL method calculation are presented in Figure 5. The shaded cells in the tables show the presence of significant relationships between the criteria. The significant relationships were those whose values were higher than the threshold values set by the experts: α_Crisp = 0.755, α_Fuzzy = 0.373, and α_Grey = 0.345. The threshold value was calculated as the average of all estimates of the total direct linkage matrix T. Criteria C1, C3, and C5 were assigned to the “cause” group and C2 and C4 to the “effect” group in all models according to the values of (D_i − R_i).

Similar calculations were performed for the evaluation of sub-criteria. Maps of network relations of sub-criteria are presented in Figure 6.

The figures indicate that the uses of Crisp, Fuzzy, and Grey DEMATEL models yielded different results: the number of links and the values of the strength of influence of the criteria on each other changed.

Criteria and sub-criteria were ranked according to the values of weighting coefficients, which were calculated based on the results of assessing the number of relationships (Di + Ri) and the strength of influence (D_i − R_i) between criteria and sub-criteria (

Table 12. Results of the ranking of criteria and sub-criteria for the SSC drivers’ assessments.

Criteria	Sub-Criteria	Local Weight						Global Weight
		Crisp		Fuzzy		Grey		Crisp	Fuzzy	Grey
Economic (C1)	Profit (C1.1)	0.241	0.345	0.233	0.345	0.244	0.351	0.083	0.081	0.086
	Operating expenses (C1.2)		0.316		0.318		0.311	0.076	0.074	0.076
	Fixed investment (C1.3)		0.338		0.336		0.337	0.082	0.078	0.083
Energy-ecological (C2)	The energy intensity (C2.1)	0.190	0.521	0.192	0.500	0.189	0.500	0.099	0.096	0.095
	Greenhouse gas emissions of CO2 (C2.2)		0.478		0.500		0.500	0.091	0.096	0.095
Quality (C3)	Safety of cargo transportation (C3.1)	0.220	0.292	0.215	0.304	0.221	0.288	0.064	0.066	0.064
	Timeliness of cargo transportation (C3.2)		0.348		0.345		0.351	0.077	0.075	0.078
	The coefficient of flow controllability (C3.3)		0.358		0.350		0.360	0.079	0.076	0.080
Statistical (C4)	The coefficient of flow irregularity (C4.1)	0.135	0.247	0.147	0.246	0.130	0.247	0.034	0.036	0.032
	The coefficient of complexity structure of flow (C4.2)		0.266		0.263		0.267	0.036	0.039	0.035
	The coefficient of flow discreteness (C.4.3)		0.238		0.240		0.237	0.032	0.036	0.031
	The coefficient of differentiability of flow (C4.4)		0.247		0.249		0.246	0.034	0.037	0.032
Flow physical (C5)	The mass (quantity) of flow (C5.1)	0.212	0.329	0.211	0.326	0.213	0.330	0.070	0.069	0.071
	The speed of flow (C5.2)		0.397		0.385		0.404	0.085	0.081	0.086
	The length of the route (C5.3)		0.273		0.288		0.264	0.058	0.061	0.057

).

The obtained criterion and sub-criteria ranks were used to evaluate the SSC drivers using the CRADIS method. The results of SCC driver ranking using the three models—C-DEMATEL-CRADIS, F-DEMATEL-CRADIS, and G-DEMATEL-CRADIS—are presented in

Table 13. Results of driver ranking by GRADIS method.

Drivers	Method
	C-DEMATEL-CRADIS						F-DEMATEL-CRADIS						G-DEMATEL-CRADIS
	s+	s-	K+	K-	Q	Rank	s+	s-	K+	K-	Q	Rank	s+	s-	K+	K-	Q	Rank
D1.1	0.995	0.348	0.487	0.405	0.446	54	0.953	0.339	0.466	0.400	0.433	54	0.937	0.355	0.453	0.409	0.431	54
D1.2	0.951	0.392	0.510	0.456	0.483	49	0.911	0.381	0.487	0.449	0.468	49	0.891	0.401	0.476	0.462	0.469	47
D1.3	0.810	0.533	0.599	0.621	0.610	6	0.775	0.517	0.572	0.609	0.591	7	0.745	0.546	0.569	0.629	0.599	5
D1.4	0.925	0.419	0.525	0.488	0.506	37	0.885	0.408	0.502	0.480	0.491	38	0.866	0.426	0.490	0.491	0.490	37
D1.5	0.862	0.481	0.563	0.561	0.562	10	0.826	0.466	0.537	0.550	0.543	13	0.798	0.494	0.531	0.569	0.550	10
D1.6	0.951	0.392	0.510	0.457	0.483	48	0.911	0.381	0.487	0.449	0.468	48	0.892	0.400	0.476	0.461	0.468	50
D1.7	0.871	0.472	0.557	0.550	0.554	14	0.823	0.469	0.539	0.553	0.546	12	0.813	0.478	0.521	0.551	0.536	16
D1.8	0.815	0.529	0.595	0.616	0.606	7	0.773	0.520	0.574	0.613	0.594	6	0.754	0.538	0.563	0.620	0.591	7
D2.1	0.968	0.375	0.501	0.437	0.469	51	0.925	0.367	0.480	0.433	0.456	51	0.908	0.384	0.467	0.442	0.455	51
D2.2	0.980	0.363	0.495	0.423	0.459	53	0.939	0.353	0.473	0.417	0.445	53	0.921	0.371	0.461	0.428	0.444	53
D2.3	0.917	0.426	0.529	0.497	0.513	32	0.877	0.416	0.506	0.490	0.498	35	0.857	0.434	0.495	0.501	0.498	31
D2.4	0.920	0.423	0.527	0.493	0.510	35	0.878	0.415	0.506	0.489	0.497	36	0.860	0.431	0.493	0.497	0.495	34
D2.5	0.935	0.408	0.519	0.475	0.497	42	0.891	0.401	0.498	0.473	0.485	42	0.877	0.414	0.483	0.477	0.480	42
D2.6	0.930	0.413	0.521	0.481	0.501	41	0.885	0.407	0.501	0.480	0.490	40	0.872	0.419	0.486	0.483	0.485	41
D2.7	0.929	0.414	0.522	0.483	0.502	39	0.891	0.401	0.498	0.473	0.485	41	0.867	0.424	0.489	0.489	0.489	38
D2.8	0.947	0.396	0.512	0.461	0.487	46	0.902	0.390	0.492	0.460	0.476	46	0.888	0.403	0.477	0.465	0.471	46
D2.9	0.949	0.394	0.511	0.459	0.485	47	0.904	0.389	0.491	0.458	0.475	47	0.891	0.400	0.476	0.461	0.469	49
D3.1	0.877	0.467	0.553	0.544	0.548	17	0.827	0.465	0.537	0.548	0.542	14	0.820	0.471	0.517	0.543	0.530	17
D3.2	0.887	0.456	0.547	0.532	0.539	20	0.846	0.447	0.525	0.526	0.526	21	0.826	0.465	0.513	0.536	0.525	18
D3.3	0.862	0.481	0.562	0.560	0.561	11	0.820	0.473	0.541	0.557	0.549	10	0.804	0.487	0.527	0.562	0.544	11
D3.4	0.899	0.444	0.540	0.518	0.529	27	0.851	0.441	0.521	0.520	0.521	25	0.841	0.450	0.504	0.519	0.511	27
D3.5	0.897	0.446	0.541	0.520	0.530	26	0.849	0.443	0.523	0.522	0.522	24	0.841	0.451	0.504	0.520	0.512	26
D3.6	0.929	0.414	0.522	0.483	0.502	40	0.884	0.408	0.502	0.481	0.492	37	0.872	0.420	0.486	0.484	0.485	40
D3.7	0.808	0.535	0.600	0.623	0.612	5	0.771	0.521	0.575	0.614	0.595	5	0.746	0.546	0.568	0.629	0.599	6
D3.8	0.798	0.545	0.608	0.636	0.622	3	0.764	0.529	0.581	0.623	0.602	3	0.732	0.559	0.579	0.645	0.612	3
D3.9	0.868	0.475	0.559	0.554	0.556	12	0.828	0.465	0.536	0.548	0.542	15	0.807	0.485	0.526	0.559	0.542	12
D3.10	0.919	0.424	0.528	0.494	0.511	34	0.872	0.421	0.509	0.496	0.502	31	0.863	0.429	0.491	0.494	0.493	35
D3.11	0.869	0.474	0.558	0.552	0.555	13	0.820	0.472	0.541	0.556	0.549	11	0.812	0.480	0.522	0.553	0.538	15
D3.12	0.913	0.430	0.531	0.502	0.516	30	0.875	0.417	0.507	0.492	0.500	33	0.851	0.441	0.498	0.508	0.503	30
D4.1	0.896	0.448	0.542	0.522	0.532	24	0.857	0.435	0.518	0.513	0.516	27	0.834	0.457	0.508	0.527	0.517	24
D4.2	0.873	0.470	0.555	0.547	0.551	16	0.837	0.455	0.530	0.536	0.533	17	0.811	0.481	0.523	0.554	0.539	14
D4.3	0.851	0.493	0.570	0.574	0.572	9	0.816	0.477	0.544	0.562	0.553	9	0.787	0.505	0.539	0.582	0.560	9
D4.4	0.873	0.470	0.556	0.548	0.552	15	0.836	0.456	0.531	0.538	0.534	16	0.810	0.482	0.523	0.555	0.539	13
D4.5	0.917	0.426	0.529	0.497	0.513	31	0.871	0.421	0.510	0.497	0.503	30	0.860	0.432	0.493	0.498	0.496	33
D4.6	0.944	0.400	0.514	0.466	0.490	45	0.899	0.393	0.494	0.464	0.479	43	0.885	0.407	0.479	0.469	0.474	45
D4.7	0.918	0.426	0.529	0.496	0.512	33	0.876	0.416	0.506	0.490	0.498	34	0.858	0.434	0.494	0.500	0.497	32
D4.8	0.926	0.417	0.523	0.486	0.504	38	0.885	0.408	0.502	0.480	0.491	39	0.868	0.424	0.489	0.488	0.489	39
D4.9	0.940	0.403	0.516	0.470	0.493	44	0.899	0.393	0.494	0.464	0.479	44	0.880	0.412	0.482	0.475	0.478	44
D4.10	0.887	0.456	0.547	0.532	0.539	19	0.846	0.447	0.525	0.527	0.526	20	0.827	0.465	0.513	0.536	0.524	19
D4.11	0.892	0.451	0.544	0.525	0.534	23	0.852	0.441	0.521	0.519	0.520	26	0.832	0.459	0.510	0.530	0.520	23
D5.1	0.887	0.456	0.547	0.531	0.539	21	0.841	0.451	0.528	0.532	0.530	19	0.828	0.463	0.512	0.534	0.523	20
D5.2	0.907	0.436	0.534	0.508	0.521	28	0.869	0.423	0.511	0.499	0.505	29	0.846	0.446	0.501	0.514	0.507	28
D5.3	0.940	0.403	0.516	0.470	0.493	43	0.899	0.393	0.494	0.463	0.478	45	0.879	0.412	0.482	0.475	0.479	43
D5.4	0.889	0.454	0.545	0.529	0.537	22	0.846	0.446	0.524	0.526	0.525	22	0.830	0.462	0.511	0.532	0.522	22
D5.5	0.885	0.458	0.548	0.534	0.541	18	0.840	0.453	0.529	0.534	0.531	18	0.830	0.462	0.511	0.532	0.522	21
D5.6	0.823	0.520	0.589	0.606	0.598	8	0.784	0.508	0.566	0.599	0.583	8	0.762	0.530	0.557	0.611	0.584	8
D6.1	0.908	0.435	0.534	0.507	0.521	29	0.862	0.431	0.515	0.507	0.511	28	0.851	0.441	0.498	0.508	0.503	29
D6.2	0.920	0.423	0.527	0.493	0.510	36	0.873	0.420	0.509	0.494	0.501	32	0.864	0.428	0.491	0.493	0.492	36
D6.3	0.780	0.564	0.622	0.657	0.639	1	0.744	0.549	0.597	0.647	0.622	1	0.716	0.576	0.592	0.664	0.628	1
D6.4	0.802	0.541	0.605	0.630	0.618	4	0.768	0.524	0.578	0.618	0.598	4	0.738	0.553	0.574	0.638	0.606	4
D6.5	0.782	0.561	0.620	0.654	0.637	2	0.749	0.543	0.592	0.640	0.616	2	0.717	0.574	0.591	0.662	0.626	2
D6.6	0.896	0.447	0.541	0.521	0.531	25	0.849	0.443	0.523	0.522	0.522	23	0.840	0.452	0.505	0.521	0.513	25
D6.7	0.952	0.391	0.509	0.456	0.482	50	0.912	0.380	0.487	0.448	0.467	50	0.891	0.400	0.476	0.461	0.469	48
D6.8	0.970	0.374	0.500	0.435	0.468	52	0.929	0.364	0.478	0.429	0.453	52	0.909	0.382	0.466	0.441	0.453	52
s0	0.485	0.858					0.444	0.849					0.424	0.867

and Figure 7.

The three most significant SSC drivers were D6.3 “Corporate Information Systems” (rank #1); D6.5 “Intelligent Transportation Systems” (rank #2); and D3.8 “Technical condition of vehicle fleet” (rank #3). The three least significant drivers were D1.1 “Environmentally friendly raw materials (at supplier)” (rank #54); D2.2 “Reusable or recyclable raw materials” (rank #53); and D6.8 “Recycling processes for waste, packaging, finished products” (rank #52).

The ranking of SCC drivers separately for each element of the supply chain allowed us to establish the most important (rank #1) drivers of each element: D1.8 “E-commerce with supplier” (input element), D2.3 “Eco-friendly equipment” (processing element), D3.8 “Technical condition of vehicle fleet” (transport element), D4.3 “Spatial organization of warehouse facilities” (cumulative element), D5.6 “E-commerce with consumer” (output element), and D6.3 “Corporate Information Systems” (control element).

The following drivers had the least impact: D1.1 “Environmentally friendly raw materials (at supplier)” (input element, rank #8), D2.2 “Reusable or recyclable raw materials” (processing element, rank #9), D3.6 “Compliance of transportation vehicles with legal regulations” (transport element, rank #12), D4.9 “Placement and storage of products and waste” (cumulative element, rank #11), D5.3 “Tare and packaging return system” (output element, rank #6), and D6.8 “Recycling processes for waste, packaging, finished products” (control element, rank #8).

The results of the SSC drivers’ ranking for each supply chain element are presented in Figure 8.

The evaluation of the quality of ranking by the three methods using MRER (17) showed a satisfactory result, MRER = 188/54 × 3 = 1.16. That is, the average difference between the ranks calculated by different methods was close to one.

4.3. Sensitivity Analysis

Sensitivity analysis of the ranking results was performed by comparing the results of CRADIS with the results of three other multi-criteria methods. We chose the methods TOPSIS (Technique for the Order of Preference by Similarity to Ideal Solution) [125], ARAS (Additive Ratio Assessment) [126], and MARCOS (Measurement of Alternatives and Ranking according to COmpromise Solution) for comparison [127]. The results are shown in Figure 9, and the numerical values of the ranks calculated by all methods are presented in Appendix B.

Spearman rank correlation coefficient calculation showed a high correlation between the ranking results of different MCDM methods. The average correlation coefficient was for all models 0.966, for Crisp models is 0.9679, for Fuzzy models is 0.9647, and for Grey models is 0.9658 (

Table 14. Spearman rank correlation coefficient.

MCDM Model	C-DEMATEL TOPSIS	C-DEMATEL ARAS	C-DEMATEL MARCOS	C-DEMATEL CARDIS	F-DEMATEL TOPSIS	F-DEMATEL ARAS	F-DEMATEL MARCOS	F-DEMATEL CARDIS	G-DEMATEL TOPSIS	G-DEMATEL ARAS	G-DEMATEL MARCOS	G-DEMATEL CARDIS
C-DEMATEL TOPSIS	1.000	0.945	0.917	0.918	0.994	0.934	0.884	0.884	0.995	0.955	0.929	0.929
C-DEMATEL ARAS	0.945	1.000	0.991	0.991	0.957	0.996	0.978	0.977	0.932	0.999	0.994	0.994
C-DEMATEL MARCOS	0.917	0.991	1.000	1.000	0.937	0.997	0.995	0.995	0.898	0.986	0.998	0.998
C-DEMATEL CARDIS	0.918	0.991	1.000	1.000	0.938	0.996	0.994	0.994	0.899	0.987	0.998	0.998
F-DEMATEL TOPSIS	0.994	0.957	0.937	0.938	1.000	0.951	0.909	0.908	0.983	0.965	0.945	0.945
F-DEMATEL ARAS	0.934	0.996	0.997	0.996	0.951	1.000	0.989	0.988	0.917	0.993	0.998	0.998
F-DEMATEL MARCOS	0.884	0.978	0.995	0.994	0.909	0.989	1.000	1.000	0.861	0.970	0.991	0.990
F-DEMATEL CARDIS	0.884	0.977	0.995	0.994	0.908	0.988	1.000	1.000	0.861	0.970	0.990	0.990
G-DEMATEL TOPSIS	0.995	0.932	0.898	0.899	0.983	0.917	0.861	0.861	1.000	0.945	0.912	0.913
G-DEMATEL ARAS	0.955	0.999	0.986	0.987	0.965	0.993	0.970	0.970	0.945	1.000	0.991	0.991
G-DEMATEL MARCOS	0.929	0.994	0.998	0.998	0.945	0.998	0.991	0.990	0.912	0.991	1.000	1.000
G-DEMATEL CARDIS	0.929	0.994	0.998	0.998	0.945	0.998	0.990	0.990	0.913	0.991	1.000	1.000

).

4.4. Managerial Implications of Assessing Supply Chain Sustainability Drivers

The proposed system of supply chain sustainability drivers and the results gained from assessing the degree of influence of drivers on supply chain elements can be used as a basis for the systematization and implementation of green solutions in supply chain management. Such solutions can be applied both to individual elements of the supply chain and comprehensively to the entire chain. It should be noted that the calculated driver ranks and the degree of their impact on supply chain elements should be considered as recommendations for decision makers on what specific initiatives should be implemented to improve supply chain sustainability. As such initiatives, we propose the use of a system of green logistics methods and tools [108].

Table 15. Matching green logistics methods and tools to supply chain sustainability drivers.

Supply Chain Elements	Highest-Ranked Drivers	Solutions for the Implementation of Green Logistics Methods and Tools [108]
		Green Logistics Method	Green Logistics Instrument
Input element	D1.8—E-commerce with supplier	Procurement planning, execution, and supply control	Electronic document management with organizations and suppliers
Processing element	D2.3—Eco-friendly equipment	The use of eco-friendly equipment and technologies	Equipment with minimal impact on the environment
Transport element	D3.8—Technical condition of vehicle fleet	Selection of eco-friendly vehicles	Vehicles with the least impact on the environment
			Selection of vehicles’ relevant requirements in the field of ecology
		Transport management and transport planning	Provision of technological unity for transport and warehouse process
Cumulative element	D4.3—Spatial organization of warehouse facilities	Environmental design of warehouse complexes	Environmentally sound spatial organization of elements of a warehouse complex
Output element	D5.6—E-commerce with consumer	Work with consumers of products and services	Electronic document circulation in the organization of interaction with consumers
Control element	D6.3—Corporate Information Systems	Development and implementation of corporate information systems	ERP (Enterprise Resource Planning System)
			CRM (Customer Relationship Management System)
			MES (Manufacturing Execution System)
			WMS (Warehouse Management System)
			EAM (Enterprise Asset Management)
			HRM (Human Resources Management)
	D6.5—Intelligent Transportation Systems	Development and implementation of intelligent transport systems	Implementation of advanced information technologies (RFID, GPS, GIS, EDI, GPRS, GSM)

presents an example of forming decisions on the implementation of green logistics tools based on the results of the assessment of supply chain sustainability drivers. In our opinion, the final decision on the implementation of certain green logistics tools should be made using a combination of multi-criteria decision-making methods, mathematical optimization methods, and simulation modeling.

4.5. Discussion

This paper analyzes SSC drivers based on a review of logistics research and supply chain management practices. Open-access publications from 2005 to 2024 were analyzed. This is the most comprehensive analysis of SSC drivers to date, as existing reviews analyze publications over the periods of 2003–2016 [8], 2005–2018 [16], 2006–2018 [10], 2009–2019 [128], and 2006–2020 [129] and poorly consider the results of current research and changes that have occurred recently. In addition, the number of analyzed articles on the research topic in this paper was 101 publications, in contrast to the reviews listed above, in which the numbers of publications were 54 [16], 47 [10], 58 [128], and 51 [129]. One of the shortcomings of the existing reviews, in our opinion, is the emphasis on certain subject areas or aspects of supply chain management. Thus, the works investigate factors affecting supply chain performance [9], critical factors for the realization of sustainable supply chain innovations [11], barriers and drivers of supply chain visibility [10], barriers and drivers for reverse logistics [16], barriers and drivers towards the circular economy in supply chains [130], challenges and drivers of the internet of things in supply chains [131], and drivers of and barriers to reducing carbon emissions from supply chain systems [132]. Exceptions are works that investigate barriers and drivers of green supply chain management [129,133] and sustainable supply chains [8].

We used supply chain type, chain structure and functions, driver characteristics, and driver evaluation methods and models as the main criteria for analyzing the articles. This approach allows us to focus on the methods of grouping SSC drivers, indicators, and models of their evaluation to establish the impact of drivers on known functional areas of logistics. This distinguishes this study from most existing reviews that use standard indicators of bibliometric analysis—trends in publications, journals and citations, collaboration, and keyword focus.

Based on the results of analyzing scientific articles, a system of SSC drivers is proposed in this paper. To group and systematize the drivers, we used the structural–functional approach to describe SSC. We have identified universal elements of the supply chain, each of which performs the main functions of promotion and processing of material flow, which are inherent only to this particular element. When systematizing drivers, this excludes their duplication of drivers at different stages of the logistics process. This approach is new and differs from the classification and systematization of drivers presented in recent review papers on this topic. For example, in [8], internal drivers (corporate strategy, organizational culture, organizational resources, organizational characteristics) and external drivers (regulatory pressures, societal pressures, market pressures) are chosen as a classification feature. Studies [129] identified 28 drivers and 32 barriers, classified into two categories, such as internal and external. The case studies for 2024 on the research topic use different approaches as attributes for grouping drivers. For example, [95] uses 4 groups of factors (financial constraints, technical challenges, organizational resistance, regulatory and compliance issues) and 12 sub-factors. In [97], two attributes are used: internal driver (organizational support, technology, knowledge, financial/economic) and external pressure (regulatory pressure, market pressure, supplier pressure, competition pressure). In [102], 30 factors are grouped into six groups—green suppliers, green technology and expertise, green regulations and support, green products, green organization-organization and communications, and the dimension of green human resources. In [98], four groups are used (technological, organizational, environmental, and motivation factors), and in [99], three groups are used (environmental factors, GSCM adoption, environmental uncertainty). Such various approaches to systematization and grouping of factors and drivers shows the lack of a single comprehensive approach to assessing the drivers of all known functional areas of logistics in SSC management.

A combination of DEMATEL and CRADIS multi-criteria methods was used to evaluate and rank SSC drivers. This combination was used for the first time to evaluate SSC drivers. We used DEMAEL to determine the importance of criteria and sub-criteria for SSC driver evaluation. The comparison was performed using Crisp, Gray, and Fuzzy numbers, as there was difficulty in collecting raw data, inconsistency, and a lack of information when evaluating the SSC driver criteria. A comparison of results using all three DEMATEL models showed that the Crisp, Fuzzy, and Grey DEMATEL models gave different results: the number of links and the values of the strength of influence of criteria on each other changed (See Figure 6). At the same time, the weights of the criteria and sub-criteria of SSC driver evaluation did not change significantly (See Table 12). The results of calculating the significance ranks of SSC drivers using the CRADIS method for all models also showed high correlation, and the average difference between the ranks calculated by different methods was close to one (MRER = 1.16).

The results of SSC driver ranking are the basis for the systematization and implementation of green initiatives in supply chain management. For such initiatives, we propose to use the system of methods and tools of green logistics [108] in combination with multi-criteria decision-making methods, mathematical optimization methods, and simulation modeling [134].

5. Conclusions

The paper presents a framework for assessing and ranking the drivers of sustainable supply chain development. The peculiarities of the framework include the use of a universal supply chain model, which includes structural elements that realize all known functional areas of logistics. An original two-level system of criteria and sub-criteria for evaluating logistics flows is used as the assessing criteria. It is shown how the use of these criteria allows for the evaluation of supply chain sustainability drivers. Fifty-four sustainability drivers are identified based on a literature review of logistics research and supply chain management practices. The assessment and ranking of the drivers are proposed to be performed using a combination of DEMATEL and CRADIS multi-criteria methods.

The three most significant SSC drivers were identified. These drivers are “Corporate Information Systems”, “Intelligent Transportation Systems”, and “Technical condition of vehicle fleet”. The three least significant drivers are “Environmentally friendly raw materials (at supplier)”, “Reusable or recyclable raw materials”, and “Recycling processes for waste, packaging, finished products”. A similar ranking was performed for each structural element of the studied supply chain model.

An example of using the ranked drivers to select sustainable development methods and instruments is shown. Thus, it is proposed to consider the identified drivers and the framework for their selection as a universal basis for the justification and implementation of management decisions regarding coordinated supply chain sustainability. The limitation of the study is the focus on the Control element. It is assumed that this element coordinates the functioning of the all supply chain elements to achieve sustainable development goals. The proposed framework and the approach used generally do not sufficiently account for supply chain actors’ competition and conflicting economic objectives.

The main theoretical contributions of this study to the existing body of knowledge in logistics and supply chain management are:

−

A new approach to describing the structure and functions of the supply chain, which views the chain as a system of six elements that perform specific functions in the passage and processing of material logistics flow. Such functions are: supply and delivery (input element); production (processing element); warehousing (accumulation element); transportation (transportation element); sales and distribution (output element); and management (control element). The allocation of functions in the supply chain based on the structural–functional approach allows the drivers of supply chain sustainability to be systematized by two main features—belonging to the element of the supply chain and the functions that each element implements. This eliminates the duplication of drivers at different stages of the logistics process, and will help to identify promising solutions for supply chain sustainability.

−

The system of supply chain sustainability drivers is justified based on an analysis of scientific research in the field of logistics and supply chain management practice over the last 20 years. Supporting functions of supply chain elements—a set of specific functions of each element which influence the material flow—were used as the main features of driver systematization.

−

The methodology of evaluation and ranking of sustainable supply chain drivers, including methods for describing the structure and functions of SCM, a system of SCM drivers, and a reasonable system of SCM evaluation criteria.

The most valuable results of this research for practice are the methodology of evaluation and ranking of drivers of supply chain sustainability using a combination of the multi-criteria methods DEMATEL and CRADIS. The proposed approach can be used by stakeholders to assess the extent to which drivers influence supply chain elements. Such a comprehensive assessment will be the basis for implementing green solutions to improve supply chain resilience.

A limitation of this study is the small number of experts involved in the evaluation of supply chain sustainability drivers, which does not allow the results to be interpreted for global supply chains. Therefore, an important direction for future research is to increase the number of experts as well as to involve different stakeholders both inside the supply chain (carriers, logistics operators, suppliers, vendors) and outside it (consumers, government). This approach implies the division of supply chain sustainability drivers into internal and external and, consequently, the adaptation of principles and approaches for the effective implementation of green solutions to improve the sustainability of chains, considering the interests of all stakeholders.

Promising directions for future research on sustainability drivers for supply chains, according to the authors of the article, may be related to:

−

The increasing number of driver evaluation attributes caused by changes in economics, politics, and international relations and influences on the formation and management of global supply chains.

−

Combination of multi-criteria analysis methods with other methods (mathematical optimization methods, heuristic methods) to improve the quality of driver assessment and efficiency of implementation of management decisions to improve supply chain sustainability.

−

Integration of the results of sustainability drivers’ assessment into a combined multi-criteria simulation model for the implementation of green logistics tools in supply chains. Such a combination will make it possible to systematically consider the functional complexity of the supply chain and the main drivers of its sustainability for the comprehensive implementation of green logistics tools.