Original Research

Overcoming barriers to green procurement: Insights from an integrated model of decision-making trial and evaluation laboratory-interpretive structural modelling in logistics industry

Aries Susanty, Nia B. Puspitasari, Darminto Pujotomo, Bambang Purwanggono, Muhammad S. Akbar

Received: 31 July 2025; Accepted: 08 Oct. 2025; Published: 08 Jan. 2026

Copyright: © 2026. The Author(s). Licensee: AOSIS.
This work is licensed under the Creative Commons Attribution 4.0 International (CC BY 4.0) license (https://creativecommons.org/licenses/by/4.0/).

Abstract

Background: Indonesia’s logistics industry boosts economic growth but harms the environment via plastic waste and carbon emissions – single-use plastics and fossil fuel transport cause persistent pollution. Green procurement (GP) is essential to reduce these impacts by prioritising sustainable suppliers and products despite cost and regulatory challenges.

Objectives: The objective of this study is to explore the interconnections and significance of barriers affecting the implementation of GP in Indonesia’s logistics sector.

Method: The study employs content validity analysis, decision-making trial and evaluation laboratory (DEMATEL) and interpretive structural modelling (ISM) approaches for data processing. These methods are used to identify, analyse and structure the relationships and hierarchical positioning of the barriers.

Results: The study identifies 14 key barriers to the successful implementation of GP in Indonesia’s logistics industry. Key barriers include the lack of eco-friendly materials, insufficient supplier readiness, the absence of regulatory frameworks and inadequate government incentives. These foundational barriers trigger cascading effects that impact both operational and strategic procurement levels.

Conclusion: The findings suggest overcoming barriers through targeted supplier training, increasing eco-friendly materials, adopting effective GP regulations and enhancing government incentives, which will reduce environmental impact, promote sustainable logistics practices and support broader environmental sustainability goals.

Contribution: This study helps Indonesia’s logistics by providing practical solutions that enable the industry to adopt sustainable practices, reduce environmental impacts and support national sustainability goals.

Keywords: barriers; green procurement; logistics industry; decision-making trial and evaluation laboratory; interpretive structural modelling.

Introduction

The logistics industry is a cornerstone of economic advancement in emerging markets, particularly Southeast Asia. In Indonesia, the sector plays a pivotal role in stimulating trade, supporting employment and enhancing infrastructural connectivity, thereby contributing significantly to the country’s national competitiveness in the global marketplace (Wong, Mak & Ho 2022). Recent data reveal that Indonesia’s logistics industry grew by 15.93% in Gross Domestic Product (GDP) terms in the first quarter of 2023 compared to the same period in the previous year, and its export value increased by 26.03% in 2022. These indicators highlight the sector’s crucial role in maintaining macroeconomic stability and fostering development. Moreover, the rapid expansion of domestic and international trade networks, accelerated by urbanisation and digital commerce, reinforces logistics as a vital driver of economic modernisation in the archipelago.

Despite these contributions, the industry continues to face mounting criticism for its environmental impact. The logistics sector generates ecological degradation primarily through two vectors: the proliferation of plastic packaging waste and the heavy reliance on fossil fuel–based transportation. Single-use plastics remain a dominant material in packaging, with over 300 million tons produced annually, a substantial share of which is used in logistics applications (Letunovska et al. 2023). Meanwhile, freight transport is identified as a major contributor to global greenhouse gas (GHG) emissions, intensifying climate-related challenges (Quan, Li & Wu 2022; Tseng et al. 2019).

In Indonesia, the environmental burden is exceptionally pressing. The transportation sector accounts for over 23% of the country’s total carbon emissions, with the number of vehicles increasing at a rate of approximately 5% per year. Government regulations, such as limiting vehicular emissions to 1.0 g/km for carbon monoxide and 0.18 g/km for nitrogen oxides, have been enacted to mitigate this trend. Nevertheless, the effectiveness of these measures remains limited because of gaps in enforcement, infrastructure constraints and the slow adoption of clean technologies (Dahliani et al. 2023). These challenges illustrate the structural inefficiencies and policy fragmentation that complicate efforts towards sustainable logistics practices.

One proposed pathway to reconcile economic imperatives with environmental sustainability is green procurement (GP). Green procurement integrates environmental considerations into procurement decisions by prioritising suppliers, products and services that demonstrate sustainability performance (Letunovska et al. 2023; Pinto 2023). By embedding environmental criteria into purchasing decisions, logistics firms can indirectly influence both upstream and downstream actors in the supply chain to adopt more responsible practices. Empirical research indicates that the implementation of GP has resulted in significant reductions in plastic use, the expansion of reverse logistics systems and increased efficiency in resource utilisation (Balm 2022).

Beyond its environmental benefits, GP also delivers tangible economic and reputational advantages. Firms adopting GP practices report cost savings through reduced material usage, optimised waste management and enhanced supplier reliability. Moreover, these practices serve as strategic differentiators in competitive markets, where consumers increasingly value corporate responsibility and environmental awareness. Studies further indicate that GP adoption can catalyse systemic supply chain transformations, advancing corporate alignment with global sustainability goals while simultaneously bolstering competitiveness (Balm 2022; Letunovska et al. 2023). Case studies of third-party logistics (3PL) providers reinforce this notion, demonstrating how strategic procurement practices can drive sector-wide sustainability transitions (Wong et al. 2022).

Nevertheless, the implementation of GP in logistics is constrained by several barriers. These include high perceived upfront costs, insufficient government incentives, ambiguous regulatory frameworks and the lack of supplier readiness to deliver environmentally friendly products (Ahmed, Thaheem& Maqsoom 2020; Simion, Nicolescu & Vrîncut 2019). Additionally, the limited availability of eco-friendly materials poses practical constraints for logistics companies attempting to redesign procurement processes. While such barriers have been thoroughly investigated in industries like construction, manufacturing and real estate, empirical research specific to the logistics sector remains scarce. The fragmented institutional environment and inconsistent policy enforcement further exacerbate these limitations, creating uncertainty for firms considering GP integration (Vejaratnam et al. 2023).

A review of the literature reveals that prior research has primarily employed analytical techniques, such as interpretive structural modelling (ISM) and the decision-making trial and evaluation laboratory (DEMATEL), to categorise and analyse barriers to GP adoption (Fang, Wang & Song 2020; Luthra et al. 2011). These methodologies are particularly effective in mapping causal relationships and structuring barriers hierarchically, thereby identifying root causes. However, studies that combine ISM and DEMATEL in the logistics sector are exceedingly rare. Considering the sector’s operational complexity and significant environmental footprint, there is a pressing need for integrative frameworks that can capture interdependencies among GP barriers while providing actionable insights.

Against this backdrop, the present study seeks to address the identified research gap by focusing on the Indonesian logistics sector and employing a combined ISM-DEMATEL methodology. This integrated approach enables the construction of a hierarchical model of barriers while simultaneously mapping their causal interrelationships. By leveraging these complementary tools, the study aims to offer a comprehensive understanding of the systemic impediments to GP adoption in logistics.

The contribution of this research is threefold. Firstly, it advances theoretical understanding by applying and integrating ISM and DEMATEL in a context where such approaches remain underutilised. Secondly, it provides practical insights by identifying leverage points that policymakers and industry practitioners can target to overcome GP barriers. Thirdly, by focusing on Indonesia’s logistics sector – an area often overlooked in the GP literature – this study highlights a context of high relevance for both emerging markets and global sustainability agendas. Ultimately, the findings aim to inform strategies that support the transition to environmentally sound, economically viable and socially responsible logistics practice.

Literature review

Theoretical background: Institutional theory, stakeholder theory and green procurement

Institutional theory can be used as a strong foundation for understanding how organisational norms and external forces influence an organisation’s adoption and application of GP practices. According to DiMaggio and Powell (1983, 1991; Scott 2007), institutional theory has historically focused on how groups and organisations can better secure their positions and legitimacy by adhering to the rules and norms of the institutional environment. These rules and norms include regulatory structures, governmental agencies, laws, courts, professions, scripts and other societal and cultural practices that exert conformance pressures. According to institutional theory, firms’ strategies and organisational decision-making are influenced by external social, political and economic pressures as they attempt to legitimise their current practices or adopt new ones in the eyes of other stakeholders (Jennings & Zandbergen 1995).

Three types of drivers that produce isomorphism in organisational strategies, structures and procedures are described by institutional theory. According to DiMaggio and Powell (1983), these drives are coercive, normative and mimetic. Coercive power stems from the influence of individuals in positions of authority, as seen in the logistics industry’s GP sector. According to Kilbourne, Beckmann and Thelen (2002), coercive forces play a critical role in promoting sustainability and environmental management. According to Sarkis, Zhu and Lai (2011), normative drivers ensure that organisations comply in order to be seen as engaging in lawful activities. According to Ball and Craig (2010), normative constraints push businesses to be more ecologically conscious. They contend that institutional research is necessary to comprehend changing societal norms (such as ecological thinking and ethical principles) and organisational reactions to environmental problems. According to March and Olsen (1989), normative drivers exert influence because of a social obligation to conform, which is based on societal necessity or what an organisation or individual is expected to do. Mimetic isomorphic drivers occur when businesses attempt to replicate the path to success and, consequently, legitimacy by copying the strategies of successful competitors in their industry. Examples of these strategies in this research context include improved supplier reliability, optimised waste management and GP for material usage.

Stakeholder theory, in conjunction with institutional theory, can serve as a foundation for examining how competing interests and potential disputes among different stakeholders influence the logistics sector’s adoption of green purchasing practices. Stakeholder theory focuses on the mutual connections between an organisation and the different groups that can affect or be affected by the achievement of an organisation’s objectives (Freeman 1984). According to Hörisch, Freeman and Schaltegger (2014), this theory has been frequently used as a lens to evaluate green supply chain management (GSCM) strategies, considering a variety of perspectives, roles, competencies and conflicting performance expectations. Despite the large number of stakeholders that may be considered, prior research has tended to divide stakeholders into two basic groups: internal and external (e.g., Sarkis, Gonzalez−Torre & Adenso−Diaz [2010]). Internal stakeholders, including senior administrators, procurement staff, users (workers) and internal specialists (such as Information Technology [IT] or environmental specialists), frequently create and manage GP programmes. Crucial, yet equally important, are the differing levels of influence or control that various stakeholder groups have over procurement objectives, data collection and disclosure (Domingues et al. 2017). For example, the successful implementation of green IT procurement, which aims to increase organisational understanding and engagement in new initiatives, requires the assistance of senior administrators (Walker, Di Sisto & McBain 2008). This kind of support could lead to more financing for information gathering or training (Sharma & Henriques 2005). Additionally, finding talented individuals and retaining employees whose beliefs align with GP objectives may be made easier by top administrators’ commitment to environmental sustainability (Sarkis et al. 2010).

Unlike internal stakeholders, external stakeholders typically do not have direct influence over organisational resources or the process (Liu et al. 2021). By influencing public opinion and drafting new legislation through various democratic and social channels, they have an indirect effect (Brammer & Walker 2011). These stakeholders may be more inclined to emphasise – and exert greater pressure to incorporate – green criteria into procurement contracts than internal stakeholders (Roman 2017), even if the green criteria are only given symbolic weight in subsequent decision-making (Igarashi, De Boer & Michelsen 2015). Last but not least, suppliers can be a hindrance to the adoption of GP practices (Ahsan & Rahman 2017; Amann et al. 2014), despite being essential to any organisation’s attempts to enhance environmental sustainability (Kelly et al. 2021). By managing information and producing eco-friendly products and services, suppliers, as stakeholders, exercise influence (New, Green & Morton 2002).

Previous research on barrier to practising green procurement

This study employs the Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) to gather previous research on barriers to practicing GP in the logistics industry. Initially, this study identified 250 manuscripts, with 248 indexed in SCOPUS and two others discovered through Emerald Insight that are not indexed in SCOPUS. During the search through SCOPUS, the study utilised the following keywords: TITLE-ABS-KEY ((green*) AND (procure* OR purchase*) AND (barrier*)), applying a filter to limit the publication years to the last 12 years. Additionally, this study set other limitations, such as selecting sources exclusively from manuscripts. This study excludes book chapters because of difficulties obtaining full texts and omits manuscripts from conference proceedings because some articles are considered lower quality for research reference. In the second stage, this study screened titles and abstracts relevant to the study. They deemed 31 manuscripts relevant (219 manuscripts were excluded because their titles were less relevant and their abstracts did not contain several keywords pertinent to the research, which could potentially lead to bias in the review process). Next, this study conducted a thorough examination (full-text review for eligibility) of these 31 articles. This examination aimed to determine whether these manuscripts mentioned barriers to implementing GP and whether they identified significant barriers associated with it. Eight manuscripts specifically discussed barriers in the context of implementing GP in companies (23 articles were excluded because they did not mention barriers in their research or because the barriers cited were not relevant to this study’s context).

Additionally, some manuscripts had research objectives that were less aligned with the investigated topic. The next step involved conducting backward or forward citations from the eight manuscripts that had previously passed the eligibility check. This study performed this citation process to discover any additional relevant manuscripts related to the study. As a result, this study identified two additional manuscripts relevant to this study through backward or forward citations. These two manuscripts provided unique perspectives on the barriers to GP, enriching the review process. The process of gathering previous research on barriers to implementing GP in the logistics industry is illustrated in Figure 1. Then, Table 1 outlines 10 articles that identified the primary barriers to implementing GP in the logistics industry.

FIGURE 1: The process of gathering the previous research on barriers to implementing green procurement in the logistics industry.

Based on the information provided in Table 1, it is evident that no studies have specifically focused on identifying and analysing the barriers to implementing GP within the logistics industry. Most existing research in this area has concentrated on the construction and manufacturing industries, leaving the logistics sector underexplored. In addition, Table 1 highlights that while some studies explore barriers to GP, they predominantly employ methods such as ISM, ANOVA, BWM Fuzzy or qualitative approaches. Notably, only one study integrates both ISM and DEMATEL to examine the barriers and their interrelationships. This unique integration of ISM and DEMATEL in this context is a key aspect that distinguishes this study and contributes to its originality. By discussing studies in related sectors, even those that employ different methodologies, the research gap in the logistics industry can be further highlighted, thereby strengthening the justification for using the ISM-DEMATEL integration in this study. This methodological approach provides a comprehensive and nuanced understanding of the barriers to GP, which have not been sufficiently addressed in logistics-related studies.

The ISM-DEMATEL integration is a hybrid multi-criteria decision-making (MCDM) method, or a part of the broader MCDM family, which combines multiple decision-making techniques to address complex sustainability challenges. Expanding the review to include related logistics-sector studies, even if not identical in methodology, enhances the validity of the gap statement and supports the argument for methodological originality. In this case, hybrid MCDM methods have proven effective in addressing complex sustainability challenges in logistics and supply chain management. The following are several relevant studies that highlight the application of hybrid MCDM methods in related areas of logistics and supply chain.

Supplier selection

Hybrid models: Several studies have applied hybrid MCDM models for sustainable supplier selection. For instance, a combination of the fuzzy Delphi method, analytic network process (ANP) and Technique for Order Preference by Similarity to Ideal Solution (TOPSIS) has been used to select optimal suppliers by considering interdependencies among criteria and ranking alternatives effectively (Wang et al. 2021; Wu, Hsieh & Chang 2013; Zambujal-Oliveira & Fernandes 2024).

Comparative approaches: Another study compared two hybrid approaches, Interpretive Structural Modeling-Analytic Network Process-Elimination and Choice Translating Reality (ISM-ANP-ELECTRE) and Interpretive Structural Modeling-Analytic Network Process-VlseKriterijumska Optimizacija I Kompromisno Resenje (ISM-ANP-VIKOR), for sustainable supplier selection, highlighting the effectiveness of these methods in ranking and selecting suppliers based on sustainability criteria (Girubha, Vinodh & Kek 2016).

Route selection in multi-modal supply chains

Integrated techniques: A novel hybrid MCDM approach integrating analytic hierarchy process (AHP), data envelopment analysis (DEA) and TOPSIS was proposed for route selection in multimodal supply chains. This method reconciles conflicting criteria and provides a preferred order of multimodal routes, validated through an empirical case study (Jayant & Neeru 2020).

Sustainable packaging design

Reverse logistics: In the context of food supply chains, hybrid MCDM methods have been used to facilitate the selection of sustainable product-package designs. These methods help minimise food losses and consider the entire life cycle of packaging, emphasising the importance of reverse logistics in sustainability (Koohathongsumrit & Meethom 2021).

Green supply chain management (GSCM)

Group evaluation: A hybrid decision-making approach combining fuzzy DEMATEL, Fuzzy Step-wise Weight Assessment Ratio Analysis (Fuzzy SWARA) and Fuzzy Technique for Order of Preference by Similarity to Ideal Solution (Fuzzy TOPSIS). was developed for evaluating GSCM implementation criteria. This approach facilitates the quantification of trade-offs between economic, social and environmental criteria, as illustrated through a real-life case study (Alastal et al. 2025).

Third-party logistics (3PL) provider selection

Fuzzy methods: The selection of 3PL providers has been enhanced using hybrid MCDM methods such as fuzzy AHP and fuzzy VIKOR. These methods consider economic, service level, environmental, social and risk aspects, providing a robust evaluation and selection process (Erdebilli & Sıcakyüz 2024).

By reviewing these studies, despite their differing methodologies, the gap in the logistics sector, where the application of integrated ISM and DEMATEL remains scarce, is clearly visible. This condition underscores the methodological contribution of this study, which aims to provide a nuanced and comprehensive analysis of the barriers to GP in logistics – an area that has not been sufficiently addressed in previous research.

Method of research

Barriers to maximising halal industry performance

Although the research in Table 1 does not explicitly address barriers to implementing GP in the logistics industry, several previous studies on the concept of GP and its barriers in various sectors provide valuable insights for this research. Based on these insights, this study has identified 20 factors that could serve as barriers to implementing GP in the logistics industry. These factors include the following:

• Lack of cost of implementing GP procedures in the logistics industry (BI1).
• Increased costs because of the use of environmentally friendly materials for packaging and transportation fuel (BI2).
• There is no special budget for green procurement practices in the packaging, fuel and electric transportation rental (BI3).
• Lack of awareness about the importance of environmental management system certification in the logistics industry (PI1).
• Lack of knowledge about eco-friendly raw material criteria in the logistics industry (PI2).
• Failure to integrate GP policies into existing organisational policies (PI3).
• Lack of expert human resources who understand GP work procedures in the logistics industry (PI4).
• Lack of commitment from top management in implementing GP (KO1).
• Lack of vision, mission and organisational culture related to environmental sustainability (KO2).
• Lack of client interest in projects and products related to the environment (KO3).
• Lack of comprehensive guidance in implementing GP in the logistics industry (KH1).
• Lack of incentives provided by the government for the adoption of GP in the logistics industry (KH2).
• There are no relevant regulations related to GP work procedures and criteria for environmentally friendly materials in the logistics industry (KH3).
• Lack of social awareness from consumers and top management of the organisation (KK1).
• There is a fear of failure to implement GP (KK2).
• Lack of staff training on GP work procedures in the logistics industry (PL1).
• Lack of adequate information technology support system in the GP process (TK1).
• Lack of initiatives taken by organisations related to green programmes (TP1).
• Lack of availability of eco-friendly materials in the local market (KP1).
• Lack of supplier readiness in the supply of environmentally friendly materials (KP2).

As these 20 barriers were not directly derived from primary logistics-sector data, an inherent transferability risk exists. The logistics sector in Indonesia has unique operational characteristics, including reliance on transportation networks, varying fuel types and client-specific service requirements, which may influence the relevance of barriers borrowed from other industries. To mitigate this potential contextual mismatch, all preliminary barriers were subjected to expert validation with professionals experienced in the Indonesian logistics industry. The validation process assessed the relevance, feasibility and applicability of each barrier within the local context. For instance, financial barriers (BI1–BI3) reflect high cost sensitivity in logistics operations, knowledge and awareness barriers (PI1–PI4, PL1) correspond with limited formal training and certification in environmental management systems, organisational and cultural barriers (KO1–KO3, KK1–KK2) align with observed practices and challenges in local logistics firms, and market and regulatory barriers (KP1–KP2, KH1–KH3, TK1, TP1) are consistent with local supply constraints and emerging government incentives for green practices. Through this validation process, the study ensures that the identified barriers are contextually relevant to the Indonesian logistics environment, while acknowledging the potential need for further refinement through direct field research.

The barriers to implementing GP in the logistics industry can be effectively explained through the lens of institutional theory and stakeholder theory. Institutional theory emphasises that external pressures, including regulations, norms and social expectations, shape organisational practices. Barriers such as the absence of a specific budget (BI3), lack of relevant regulations (KH3) and limited government incentives (KH2) reflect weak coercive pressures, which should ideally come from regulators and government institutions. Meanwhile, the lack of awareness regarding environmental management system certification (PI1) and insufficient knowledge about eco-friendly material criteria (PI2) reveal weak normative pressures. These weak pressures need to be strengthened by professional norms and societal expectations, which can motivate organisations in the logistics sector to adopt green procurement practices and take on greater environmental responsibility. Barriers such as the fear of failure (KK2) and a lack of initiative (TP1) also demonstrate weak mimetic isomorphism, as companies fail to replicate the successful green practices of their competitors.

From the perspective of stakeholder theory, internal barriers such as weak top management commitment (KO1), lack of organisational vision and culture for sustainability (KO2), inadequate expert resources (PI4) and insufficient staff training (PL1) highlight the limited role of internal stakeholders – including managers, procurement staff and specialists – in driving GP. On the other hand, external barriers such as low client interest (KO3), limited consumer social awareness (KK1) and insufficient supplier readiness (KP2) illustrate how external stakeholders also hinder adoption.

Therefore, these barriers are interconnected, indicating that the implementation of GP in the logistics industry is not only a matter of costs (BI1, BI2) but also involves institutional weaknesses, a lack of normative pressure and limited stakeholder engagement. A synergy between institutional pressures and stakeholder involvement is essential to overcome these challenges, enhance legitimacy and advance organisational sustainability.

Content validation and combined interpretive structural modelling- decision-making trial and evaluation laboratory methodology

The following section explains the content validation method, which ensures that barrier factors or items are relevant and accurately represent the concept, typically through expert assessment and the use of quantitative indices such as the content validity index (CVI). Then, follow up with an explanation of the systematic combination of DEMATEL-ISM, which maps the relationships and dependencies among barrier factors.

Content validation

The validation of GP barrier factors in logistics uses the CVI with six experts. Two evaluation methods, I-CVI and kappa statistic, are applied. A smaller respondent group requires higher expert consensus for validity (Almanasreh, Moles & Chen 2019):

• The first step is determining the I-CVI validation score acceptance limit, based on expert respondents, to ensure each identified factor is relevant and significant in the context of GP barriers in logistics. The I-CVI assessment is calculated by dividing the total barrier factors with values of 3 and 4 by the total number of expert respondents. With six experts, the minimum acceptance limit for validation is 0.833 (Almanasreh et al. 2019).
• The second step involves using the kappa statistic for in-depth validation to avoid accidental agreements. Kappa values below 0.4 (k < 0.40) are considered very bad, between 0.40 and 0.59 (k = 0.40-0.59) bad, between 0.60 and 0.74 (k = 0.60–0.74) good and above 0.74 (k > 0.74) excellent (Almanasreh et al. 2019). Barrier factors are accepted if the I-CVI is ≥ 0.833, and the kappa value is categorised as ‘Excellent’ (k > 0.74).
• The third step is to calculate scale-level validity or S-CVI/Ave based on the average I-CVI scores across all barriers. Excellent content validity is indicated by an S-CVI/Ave of at least 0.9 (Shi, Mo & Sun 2012).

Combined interpretive structural modelling- decision-making trial and evaluation laboratory methodology

The combined ISM-DEMATEL methodology is a system analysis approach that uses matrix theory and graphs to evaluate the relationships between various factors in a system (Manoharan et al. 2021). Compared to other MCDM techniques, such as AHP-TOPSIS and ANP, ISM-DEMATEL offers more thorough insights because it goes beyond simple prioritisation. Firstly, it identifies causal connections. Although AHP, TOPSIS and even ANP are effective at rating options, they often fail to differentiate between drivers and results. By identifying which barriers or enablers need to be removed first to unleash systemic progress, DEMATEL, on the other hand, clearly identifies cause-and-effect factors. Secondly, the hierarchical structure is captured. Interpretive structural modelling establishes a multi-level hierarchy that clearly identifies which barriers serve as intermediate drivers, surface-level results or underlying causes. When it comes to policy formulation, this hierarchical mapping is especially effective as it enables decision-makers to focus on the most critical layers rather than just addressing symptoms. Thirdly, interdependence is handled more openly in ISM-DEMATEL. Even while ANP enables dependency modelling, experts may find its pairwise comparisons to be cognitively taxing. While ISM arranges these interactions into a hierarchical structure, making the linkages easier to understand and act upon, DEMATEL streamlines this process using influence matrices. Fourthly, the approach offers managers interpretability that is useful. The numerical rankings that AHP-TOPSIS frequently provides might not adequately explain why a factor has a higher score. On the other hand, ISM-DEMATEL creates visual representations that facilitate managers’ utilisation of the results in strategic planning, such as hierarchical digraphs and cause-and-effect diagrams. Lastly, ISM-DEMATEL provides more than simply a ranking; it also provides a strategic plan. In addition to identifying the most critical barriers, it also illustrates how they interact and the order in which they should be addressed. This methodical approach provides more practical advice than linear prioritisation techniques. For instance, AHP-TOPSIS may show that barriers related to cost are placed higher than those related to knowledge in the context of GP. Analytic network process may also exhibit some linkages, but with intricate weight distributions. However, ISM-DEMATEL will reveal that the primary barrier leads to downstream barriers, allowing policymakers and managers to target systemic, fundamental barriers first with this level of awareness, which in turn leads to more practical and long-lasting solutions.

Decision-making trial and evaluation laboratory methodology

• Step 1: Using the DEMATEL questionnaire, each expert evaluates the relationship strength between barrier factors to assess their influence. The assessment scale ranges from 0 to 4, with 0 indicating no impact between two barriers, 1 indicating low impact between two barriers, 2 indicating moderate impact between two barriers, 3 indicating high impact between two barriers and 4 indicating very high impact between two barriers (Safdari Ranjbar, Akbarpour Shirazi & Lashkar Blooki 2014).
• Step 2: The evaluation list of the relationship between two barriers compiled by experts is presented in a direct relationship matrix, with the average score calculated, as this study involves nine experts. The matrix element Xij shows the impact of barrier factor i on barrier factor j, and the main diagonal is set to 0, as a barrier factor does not influence itself (Safdari Ranjbar et al. 2014):

  

• Step 3: Normalise the direct relationship matrix into Equation 2 and Equation 3. The main diagonal matrix remains at value 0 with the sum of each row and column having a maximum value of 1 (Safdari Ranjbar et al. 2014).

  

  

• Step 4: Calculate the direct and indirect influence matrix through Equation 4 (Safdari Ranjbar et al. 2014)

  

• Step 5: Calculate the total rows and total columns from the direct and indirect influence matrix through Equation 5 and Equation 6 (Safdari Ranjbar et al. 2014).

  

  

• Step 6: Create table prominence and cause value for each barrier and causal diagram. This causal diagram uses (Di + Ri) as the horizontal axis and (Di − Ri) as the vertical axis. The axis (Di + Ri) indicates the overall degree of variables that affect each other, while the axis (Di − Ri) reflects the difference in the degree of variables that affect and are influenced by others. The value (Di + Ri) also indicates the degree of relationship between the variables, so a larger value indicates a stronger relationship. This graph can be generated by setting a certain threshold value to determine the relationship between variables (Safdari Ranjbar et al. 2014).

Interpretive structural modelling methodology

• Step 1: Create the final reachability matrix by assigning binary values of 0 and 1 to each barrier in direct and indirect influence matrix from DEMATEL. Values exceeding the threshold are assigned a binary value of 1, while those below it receive a value of 0. The threshold value is calculated from the average of all values in the direct and indirect influence matrix (Lin, Wu & Xu 2021).
• Step 2: Classify the barriers into different levels of the ISM structure by determining their partitioning levels. To make level partitions in ISM, it starts by identifying the reachability set for each barrier, which includes the barrier itself and all barriers it can help achieve. Next, determine the antecedent set for each barrier, consisting of the barrier itself and all barriers that can help achieve it. Then, find the intersection set by identifying common elements between the reachability and antecedent sets for each barrier. Barriers whose reachability set equals their intersection set are assigned to the highest level because they do not help achieve any barriers above their level. These barriers are separated out, and the process is repeated for the remaining barriers until all levels are identified (Susanty, Purwanggono & Situmorang 2023).
• Step 3: Formation of a hierarchical model. Through the level partitioning, a hierarchical model was developed that describes various barriers according to their importance and describes them in a structured model based on the level of resistance.

Respondents of the research

The validation process, using the content validation method, involved six experts with extensive experience in procurement, logistics and environmental management. This study carefully selected the experts to ensure a diverse range of expertise, and their profiles are as follows: (1) a professor in supply chain management with 34 years of experience, representing academia and the Indonesian Supply Chain and Logistics Institute (ISLI) Board of Trustees; (2) a vice-president of procurement and logistics with 28 years of experience, representing the Indonesia Logistics and Forwarders Association (ILFA) and the Indonesian Association of Procurement Professionals (IAPI); (3) a business development and strategic procurement director with 20 years of experience, representing the Indonesian Procurement Society (IPS); (4) two experts with 15 and 13 years of experience, respectively, in government policy and digital logistics adoption, representing the Ministry of Communication and Information Technology (KOMINFO) and (5) a staff member with 2 years of experience, representing logistics company PT Cakraindo Mitra International.

However, for processing the data using the DEMATEL and ISM methodologies, it was necessary to involve three additional experts. These experts were: (1) a trucking operations staff with 3 years of experience, representing the Indonesia Logistics Association (ALI) at Dimerco Indonesia; (2) an Human Capital (HC) and General Administration (GA) supervisor with 3 years of experience, representing the Indonesian Procurement Association (API) at PT Cakraindo Mitra International and (3) a logistics staff with 1 year of experience, representing PT Cakraindo Mitra International. The inclusion of these additional experts was crucial because the DEMATEL and ISM methodologies require detailed, practical input from those with direct operational experience. These experts contributed valuable insights into the logistics and procurement processes, ensuring the study accurately reflects the challenges of GP in the logistics industry. Their participation enabled a more comprehensive understanding of the interrelationships between the barriers, thereby enhancing the credibility and relevance of the study’s findings.

Shortly, all of the experts were affiliated with diverse organisations in the supply chain management field. This diverse panel of professionals brings a wealth of expertise from various sectors, ensuring that the validation process is comprehensive and representative of the expertise required for this study, thereby enhancing the credibility of the findings. While larger panels of experts are typically used for statistical robustness, a smaller panel was chosen to provide deeper qualitative insights, following best practices in qualitative decision modelling. Feedback from the experts was collected through online questionnaires over 1 month for each process (validation and DEMATEL-ISM process). Responses were verified and cross-validated to ensure consistency and reliability. This iterative process ensured the credibility of the data used in DEMATEL-ISM model construction. By combining a validated instrument, expert evaluation and analytical modelling, the study provides both theoretical insights and practical relevance for developing GP strategies in the logistics sector.

Results

Result of content validity method

Based on the content validity calculation (Table 2), six barrier factors (BI1, KO1, KO3, KK1, PL1 and TP1) were excluded because of I-CVI values below 0.833 (I-CVI ≤ 0.833) and kappa coefficients under 0.74 (k < 0.74). Remaining factors with I-CVI ≥ 0.833 and kappa > 0.74 are considered relevant. Excluding these factors improved the initial S-CVI/Ave value of 0.833 to a final value of 0.91, exceeding the threshold. The research will proceed with 14 barrier factors for the next stage.

Result of data processing with decision-making trial and evaluation laboratory

Data processing using the DEMATEL approach is conducted to determine the degree of influence among barriers, resulting in a priority ranking for each barrier based on the obtained values. This process follows feedback from nine expert judgments on the DEMATEL questionnaire, with the list of barriers derived from the previous content validity assessment (CVA) approach. The DEMATEL data processing involves several steps: creating the direct relation matrix using Equation 1 (see Table 3), normalisation direct relation matrix using Equation 2 and Equation 3 (see Table 4), influence matrix using Equation 4 (see Table 5) and total row and total columns from the direct and indirect influence matrix – prominence and cause value. In this case, the tables related to the direct relation matrix, the normalised direct relation matrix and the influence matrix are presented in Online Appendix 1. However, the prominence and causal value of each barrier are evident in Table 3, and the causal relationship diagram based on these values is presented in Figure 2.

FIGURE 2: Causal relationship diagram.

In Figure 2:

• Quadrant one contains barriers with high prominence and cause values, making them the most influential. These include: lack of guidance for GP (KH1), lack of government incentives (KH2), absence of relevant regulations (KH3) and lack of environmental vision (KO2), with D + R and D − R values of (9.48; 0.078), (9.029; 1.544), (9.193; 1.547) and (9.388; 0.094).
• Quadrant two has low prominence but high cause values, indicating less influence but significant impact. Barriers include: lack of eco-friendly materials (KP1) and supplier readiness (KP2), with D + R and D − R values of (8.167; 1.132) and (7.687; 1.083).
• Quadrant three has low prominence and cause values, indicating minimal influence. Barriers include: lack of IT support (TK1), increased costs (BI2) and no special budget for GP (BI3), with D + R and D − R values of (8.149; −0.093), (8.438; −0.796) and (8.57; −0.618).
• Quadrant four has high prominence but low cause values, indicating they influence others but are less significant. Barriers include: lack of environmental certification awareness (PI1), lack of knowledge on eco-friendly materials (PI2), failure to integrate policies (PI3), lack of expertise (PI4) and fear of failure (KK2), with D + R and D − R values of (9.077; −0.603), (9.068; −0.958), (9.050; −1.246), (9.320; −0.284) and (9.851; −0.879).

Result of data processing with interpretive structural modelling

Interpretive structural modelling method starts from creating the final reachability matrix from influence matrix (result of DEMATEL). Final reachability matrix (see Table 4) begins with determining the threshold, calculated from the average of all values in the influence matrix (Table 3), resulting in a threshold of 0.318. Afterward, the values in the matrix are converted into binary numbers, 0 and 1. Relationships between factors exceeding the threshold are assigned a value of 1, while those below are given a value of 0. This approach simplifies the analysis by highlighting significant relationships (1) and disregarding insignificant ones (0), focusing the evaluation on key factors and avoiding ambiguity. The use of binary values also facilitates the creation of a structural matrix, which forms the basis for the ISM, providing a clearer understanding of the structure and relationships among barrier factors (Lin et al. 2021).

Level partitioning in ISM is used to divide barrier factors into hierarchical levels based on their interrelationships and dependencies. The process begins by constructing the reachability matrix from the direct relation matrix, which shows which barrier factors can be reached either directly or indirectly. Each barrier factor is analysed to determine the reachability set, which includes the barrier factors influenced by factor i (with a binary value of 1), and the antecedent set, which includes the barrier factors influencing barrier factor j (with a binary value of 1). The intersection set is the set of elements that are common to both sets. Hierarchical levels are determined through iterations, where barrier factors with identical reachability and intersection sets are placed at the same level (Lin et al. 2021). After each iteration, barrier factors assigned a level are removed, and the process continues until all barrier factors have been assigned a level for the ISM hierarchical model (see Table 5 for the results of level partition). This iteration process is used to define the levels of barrier factors in implementing GP in the logistics industry.

The levels presented in Table 5 correspond to different sets of barriers identified during the validation process, reflecting the complexity and hierarchy of challenges faced in implementing GP practices in the Indonesian logistics industry. In this case, barriers in level 1, related to operational barriers (e.g. costs, supply chain issues), will be influenced by the barriers in the level 2, which is related to knowledge and awareness barriers (e.g. lack of information or knowledge). Then, the barrier in the level 2 (lack of information or knowledge) will be influenced by barriers in the level 3, which is related to organisational barriers (e.g. company policies, management support). Moreover, the barriers at level 3 will be influenced by the barriers at level 4, which are related to resource availability (e.g. lack of human resources and information technology support systems). The barriers in level 4 will be influenced by the barriers in level 5, which are related to the direction of implementation (e.g. lack of guidance and lack of vision, mission and organisational culture). The barriers will influence the last, those in levels 5 and 6, which are related to systemic or structural barriers (e.g. government incentives, supply chain infrastructure and local regulations). In this case, barriers in level 6 represent the root causes of the barriers to GP in Indonesia’s logistics industry. Without addressing these systemic issues, barriers at higher levels – such as cost concerns, regulatory confusion and a lack of supplier readiness – will persist, continuing to delay the adoption of GP practices.

The four level 6 barriers – KH2, KH3, KP1 and KP2 – are foundational because they represent systemic issues that must be addressed to enable the successful adoption of GP in Indonesia’s logistics sector. These barriers are policy-driven and market-based, necessitating coordination among the government, logistics companies and suppliers to establish an ecosystem where GP is both feasible and scalable. In the Indonesian context, government incentives (KH2) and regulations (KH3) are essential to providing clear guidance and financial support to logistics companies in Indonesia. The local availability of eco-friendly materials (KP1) and the readiness of suppliers (KP2) are critical to ensuring that companies can source sustainable materials locally and at scale. Together, these barriers – KH2, KH3, KP1 and KP2 – interact and create a vicious cycle that impedes the transition to more sustainable logistics practices. Overcoming these barriers is critical to unlocking the potential of GP in Indonesia’s logistics sector. Addressing the root causes in level 6 (systemic barriers, such as incentives, regulations and market readiness) would create a strong foundation for the widespread adoption of sustainable practices, helping logistics companies meet their environmental targets, reduce carbon footprints and contribute to Indonesia’s broader sustainability goals:

• KH2: Lack of government incentives is a significant barrier because logistics companies, such as PT. Tiki Jalur Nugraha Ekakurir (PT JNE), struggle to justify the high initial costs of adopting green technologies like electric vehicles and eco-friendly packaging. These investments are often considered financially unviable without government subsidies, tax breaks or grants to offset the upfront costs. In other countries, government incentives make the shift to sustainability more affordable, but in Indonesia, the lack of such financial support results in delayed or limited adoption of GP practices. Without these incentives, logistics companies may continue to rely on traditional, less sustainable methods because they cannot absorb the added cost burden. This barrier is compounded by the limited scope of Presidential Regulation No. 16/2018 on Government Procurement, which primarily focuses on the public sector and does not extend sufficient support to the private logistics sector. Without comprehensive incentives for green technologies in the private sector, logistics companies are left with limited financial justification for green investments, which stalls progress in sustainability initiatives.
• KH3: Lack of relevant regulations further complicates the adoption of GP by leaving companies uncertain about which materials or technologies meet environmental standards. While Indonesia has begun implementing green public procurement policies, such as Presidential Regulation No. 16/2018, their application has been slow, and they do not yet comprehensively cover the private sector. Many companies, such as PT Citra Van Titipan Kilat (TIKI), seek to adopt eco-friendly packaging or other green practices but are hesitant because of the lack of clear guidelines or certification standards. The uncertainty about compliance with future regulations prevents companies from committing to GP because they are unsure whether their choices will align with national standards or global sustainability goals. Furthermore, although Presidential Regulation No. 22/2017 on the National Energy General Plan (RUEN) encourages renewable energy development, it does not provide sufficient regulatory clarity on how these goals can be practically integrated into the logistics industry. The lack of specific guidelines and certifications for eco-friendly logistics practices leaves businesses uncertain about the regulatory environment and hesitant to make long-term investments in sustainability.
• KP1: Lack of availability of eco-friendly materials is another major challenge. While Indonesia is a major producer of palm oil, which could be used for biofuels or other sustainable products, eco-friendly materials like biodegradable packaging are often not available locally or are too expensive. This lack of availability forces logistics companies, such as TIKI, to rely on imported materials, which increases costs and delays the implementation of GP practices. The scarcity of eco-friendly alternatives at a competitive price point makes it difficult for companies to scale up their sustainable efforts. This situation also leads to the continued use of traditional, less environmentally friendly materials simply because they are more readily available and affordable. This challenge is made significantly more significant by the Ministry of Environment and Forestry Regulation No. 75/2019 on waste management, which mandates manufacturers, retailers, and food and beverage services to develop and implement plans to reduce waste from their products and packaging by 30% by 2029. In this case, logistics service providers should be actively involved in this effort, primarily as partners in reducing packaging waste and managing waste throughout the distribution process. This regulation, part of Indonesia’s broader shift towards a circular economy, requires producers to create a waste reduction roadmap that includes their baseline, reduction plans, implementation strategies and reporting. It encourages innovation in product and packaging design, supporting the circular economy by focusing on waste minimisation, recycling and reuse. Although this regulation promotes recycling and waste reduction, it does not yet sufficiently address the development of local eco-materials. While regulations encourage sustainable practices, the infrastructure for producing and sourcing these materials is underdeveloped, especially within the logistics sector.
• KP2: Lack of supplier readiness highlights another critical issue in the GP process. Even when eco-friendly materials are available, many suppliers in Indonesia often lack the necessary capacity, infrastructure or certification to meet the demand for sustainable products. For instance, PT Pelabuhan Indonesia (Persero) (PT. Pelindo), a state-owned company managing Indonesian ports, has faced difficulties sourcing biofuels because of limited supplier readiness. Many biofuel suppliers lack the quality control systems necessary to provide reliable and consistent products at the scale required for large-scale operations. This lack of supplier capability results in delays and inconsistent supply, preventing logistics companies from fully committing to GP and scaling up their use of sustainable materials. This issue is linked to the Regulation of the Ministry of Industry on Green Industry (Industri Hijau), which encourages industries to adopt green practices but does not sufficiently develop the capacity of local suppliers to meet GP standards. While this regulation aims to drive the green industry, it has not been effectively extended to improve supplier readiness within the logistics sector, leaving companies struggling to find suitable local suppliers.

Result of data processing with integrating decision-making trial and evaluation laboratory-interpretive structural modelling

The final stage involves integrating the DEMATEL-ISM model. This stage begins by determining the ISM hierarchy model based on the partition level. It is followed by analysing relationships and the strength of the relationships between barrier factors through DEMATEL data, using the influence matrix for guidance. The integration of the DEMATEL-ISM model provides a comprehensive understanding of the relationships between barrier factors, offering both a visual hierarchy and insights into the factors that significantly influence or are influenced by others. This approach helps identify direct and indirect influences among factors and supports more informed decision-making in complex systems. The result of this integrated model is illustrated in Figure 3 – Integration Model of Diagraph DEMATEL-ISM.

FIGURE 3: Integration model of decision-making trial and evaluation laboratory-interpretive structural modelling.

In this study, the threshold value for significant influence is derived from the average of all values in the influence matrix (see Online Appendix 1, Table 3-A1), resulting in a threshold of 0.318. Any influence value below 0.318 can be considered non-significant in terms of its direct or indirect impact on other barrier factors and is not represented by an arrow in Figure 3. This threshold allows us to classify causal effects as either significant or non-significant based on whether their influence values exceed or fall below this threshold. The relationship is very complex, and this is an example of how to read the relationship between the four critical barriers in level 6, as they become the first barriers that should be evaluated in relation to the other barriers in the lower levels:

• Lack of availability of environmentally friendly materials (KP1):
  • Direct influence on lack of comprehensive guidance in the GP process (KH1) = 0.346 (significant, as it exceeds 0.318).
  • Direct influence on lack of vision, mission and organisational culture related to the environment (KO2) = 0.341 (significant, as it exceeds 0.318).
  • Indirect influence on increased costs from using environmentally friendly materials (BI2) = 0.362 (significant).
  • Indirect influence on the absence of a budget for green procurement (BI3) = 0.343 (significant).
  • Indirect influence on lack of awareness about the importance of environmental management system certification (PI1) = 0.357 (significant).
  • Lack of availability of environmentally friendly materials directly influences the lack of guidance in green procurement, the absence of vision and mission related to the environment, and indirectly impacts increased costs, the absence of a budget for green procurement and the lack of awareness about environmental management system certification.
• Lack of supplier readiness (KP2):
  • Direct influence on lack of comprehensive guidance in the green procurement process (KH1) = 0.330 (significant).
  • Indirect influence on fear of failing to implement GP (KK2) = 0.381 (significant).
  • Lack of supplier readiness directly impacts the lack of guidance in the GP process and indirectly affects the fear of failure in GP implementation.
• Lack of incentives provided by the government (KH2):
  • Direct influence on lack of comprehensive guidance in the GP process (KH1) = 0.410 (significant).
  • Direct influence on lack of vision, mission and organisational culture related to the environment (KO2) = 0.396 (significant).
  • Direct influence on absence of relevant regulations (KH3) = 0.340 (significant).
  • Indirect influence on lack of knowledge about eco-friendly raw material criteria (PI2) = 0.429 (significant).
  • Lack of incentives provided by the government directly influences guidance, organisational culture and regulations, while indirectly affecting knowledge about eco-friendly raw material criteria in GP.
• Absence of relevant regulations (KH3):
  • Direct influence on lack of comprehensive guidance in the GP process (KH1) = 0.402 (significant).
  • Direct influence on lack of vision, mission and organisational culture related to the environment (KO2) = 0.411 (significant).
  • Direct influence on lack of incentives provided by the government (KH2) = 0.344 (significant).
  • Indirect influence on fear of failing to implement GP (KK2) = 0.451 (significant).
  • Indirect influence on lack of adequate information technology (IT) support system (TK1) = 0.345 (significant).
  • Absence of relevant regulations directly influences guidance, organisational culture and incentives, while indirectly affecting fear of failure and IT support in GP.

Overall, the lack of availability of environmentally friendly materials, supplier readiness and government incentives directly influence guidance, organisational culture and regulations, while indirectly affecting costs, knowledge, fear of failure and IT support in GP. This integrated approach provides a powerful tool for managers and policymakers, highlighting the barriers that must be addressed first to unlock systemic improvements. Literature supports this causal-dependency distinction, asserting that addressing root barriers, such as lack of regulatory support or supplier readiness, can resolve a chain of associated problems (Dimand 2022; Guenther, Scheibe & Farkavcová 2010). Furthermore, the integration highlights the importance of adopting a holistic approach to sustainability management. The decision-making trial and evaluation laboratory quantifies the influence among variables, while ISM structures them hierarchically. Together, they guide targeted interventions by distinguishing between high-leverage drivers and dependent symptoms (Pinto 2023).

The results validate the methodological efficacy of combining DEMATEL and ISM to uncover the causal weight and structural positioning of barriers to GP in logistics. The insights gained from this integrative model serve as a strategic roadmap for future interventions to enhance environmental performance through effective procurement reforms.

Discussion

The study identified four primary barriers that significantly inhibit the implementation of GP practices in the logistics sector: a lack of availability of local eco-friendly materials (KP1), a lack of supplier readiness (KP2), the absence of relevant GP regulations (KH3) and the lack of government-provided incentives (KH2). The results of this study can be better understood when examined through the dual perspectives of procurement transformation and environmental governance. From a procurement transformation standpoint, the barriers identified – particularly the lack of supplier readiness (KP2) and the scarcity of eco-friendly materials (KP1) – indicate that procurement in the Indonesian logistics sector remains largely transactional and cost-driven, rather than functioning as a strategic lever for sustainability and innovation. This aligns with Walker and Brammer (2012) and Miemczyk, Johnsen and Macquet (2012), who argue that sustainable procurement requires procurement functions to evolve into strategic enablers of market transformation and supplier development. At the same time, the findings also reveal critical shortcomings in environmental governance. The absence of clear regulations (KH3) and limited government incentives (KH2) illustrate governance gaps, where state institutions have not provided coherent rules, incentives or enforcement mechanisms to steer firms towards sustainable practices. As Lemos and Agrawal (2006) highlight, weak governance frameworks undermine the alignment of market actors with environmental goals. Taken together, the evidence suggests that procurement reform in logistics cannot advance without the simultaneous strengthening of governance structures. Organisational efforts to transform procurement must be supported by institutional reforms that establish clear regulatory frameworks, provide adequate incentives and foster supplier capacity building. In this way, the DEMATEL-ISM results underscore the interdependence of procurement transformation and environmental governance in shaping the feasibility and effectiveness of GP adoption.

Local material and supplier challenges

The barrier labelled KP1, referring to the unavailability of local eco-friendly materials, reflects persistent structural limitations in the sustainable materials market. High market prices for such materials are driven by limited supplier competition and low economies of scale, which are typical in emerging economies where green technologies have not yet achieved widespread adoption. Ahmed et al. (2020) argue that the lack of a stable and scalable eco-material supply leads to cost unpredictability and risks in procurement continuity, which are further exacerbated in the logistics sector because of volume and frequency of transactions. Ashari (2021) echoes this view, pointing out that Indonesian logistics firms often encounter severe material unavailability, resulting in operational delays and inflated costs.

Supplier readiness (KP2), another critical foundation-level barrier, underscores the insufficiency of suppliers in meeting green standards. This includes a technical component, such as a lack of capability to produce eco-certified goods, and a managerial component, such as a limited understanding of sustainability expectations. Letunovska et al. (2023) emphasise that GP success depends mainly on supplier alignment with sustainability goals. Pinto (2023) notes that delays in procurement cycles and the need for frequent compliance revalidation often stem from supplier deficiencies. As these suppliers struggle to meet green specifications, logistics firms are often forced to compromise sustainability goals or seek expensive international alternatives.

Regulatory and policy gaps

The lack of well-defined regulations (KH3) and incentive structures (KH2) was identified as a highly influential systemic barrier. In line with Ahmed et al. (2020), the research finds that regulatory clarity is a precondition for market transformation. When procurement laws lack environmental clauses or fail to delineate performance criteria for sustainable practices, firms face ambiguity in compliance requirements, which deters green innovation. Vejaratnam et al. (2023) and Banihashemi et al. (2023) confirm that weak environmental monitoring and poor policy enforcement remain dominant issues in Malaysia and other developing nations.

The lack of government incentives (KH2) also diminishes the financial feasibility of transitioning to GP. As Mahardhika (2024) indicates, policies like tax reliefs or vehicle electrification subsidies remain underutilised in Indonesia, which stalls corporate transitions to environmentally friendly logistics models. The lack of such supportive policies maintains the status quo and discourages early adopters. Governmental and stakeholder engagement is thus essential in generating institutional momentum and in embedding sustainability within the procurement system (Zambika 2022).

Strategic recommendations

Several strategic recommendations can be explained as follows.

Enhancing supplier readiness (KP2)

One of the most impactful interventions to improve supplier readiness involves providing education and training. Workshops, technical assistance and sustainability certification programmes tailored to local suppliers can foster compliance and build trust within procurement systems. McKinsey & Company, as cited by Swartz and Bové (2016), reported a 10% emissions reduction by Unilever and Walmart because of supplier-focused sustainability training initiatives implemented between 2014 and 2020. As supported by Pinto (2023), such corporate training enhances internal capacity while enabling upstream integration of green practices.

In the Indonesian context, supplier development can be institutionally anchored in the Ministry of Industry. In this case, the Ministry of Industry indeed runs the Industri Hijau (Green Industry) programme. It issues the Sertifikat Industri Hijau (SIH). While SIH currently applies primarily to the manufacturing sector, its scope could be expanded to cover logistics suppliers and service providers as part of a national GP policy.

Collaboration with the Indonesian Logistics Association (ALI) would ensure that supplier development programmes are not only technically aligned with green industry standards but also adapted to the specific operational challenges of the logistics sector. This integration would provide suppliers with formal recognition, compliance incentives and practical capacity building, thereby addressing KP2 in a way that strengthens both market trust and regulatory alignment.

Increasing local eco-friendly material supply (KP1)

To address local material scarcity, governments can promote the production of bioplastics and recyclable materials through targeted incentives such as tax deductions for green manufacturers, R&D subsidies and public procurement prioritisation. For short- to medium-term needs, temporary importation from markets with advanced green standards, such as Germany and the Netherlands, can mitigate supply gaps. International producers like Stora Enso and BioLogiQ are strategic partners offering scalable and certified eco-materials.

In Indonesia, policy interventions should be anchored in the National Energy General Plan (RUEN), established under Presidential Regulation No. 22/2017 and updated by Presidential Regulation No. 73/2023. It is noteworthy that RUEN not only sets renewable energy and energy efficiency targets but also serves as a cross-sectoral framework that justifies incentives for biomass- and bio-based industries. As many eco-materials (e.g. bioplastics) are derived from agricultural waste and biomass, their development directly supports RUEN’s objectives of optimising renewable resources and reducing energy intensity.

At the same time, eco-material policies should be linked to the broader Circular Economy Roadmap (2020), developed through multi-stakeholder collaboration involving National Development Planning Agency (Badan Perencanaan Pembangunan Nasional-Bappenas), United Nations Development Programme (UNDP), the Government of Denmark, universities, research institutions and the private sector. This roadmap emphasises waste reduction, recycling and resource circulation – making it complementary to RUEN’s energy transition agenda.

Regional governments can reinforce this framework by enforcing local waste and recycling regulations, such as segregation mandates, recycling quotas and extended producer responsibility schemes, to stimulate domestic production of green materials. Meanwhile, the Indonesian Investment Coordinating Board (Badan Koordinasi Penanaman Modal – BKPM) could operationalise fiscal incentives, including tax holidays, tax allowances or investment facilitation packages, for investors establishing eco-material plants in Indonesia. By embedding eco-material policies within both RUEN and the circular economy framework, Indonesia can ensure coherence between its national energy strategy, waste reduction initiatives and industrial development, thereby reducing reliance on imports while building a competitive domestic eco-material supply chain.

Adopting global green procurement laws (KH3)

International policy benchmarks offer viable blueprints for regulation enhancement. The European Union’s Directive 2014/24/EU institutionalises sustainability in public procurement by incorporating life cycle costing and environmental compliance into the selection process (De Giacomo et al. 2019). Similarly, Sweden’s Environmental Management Act and Japan’s Green Purchasing Law (2000) mandate sustainability standards in public and private sector procurement. These frameworks promote clarity, supplier innovation and institutional accountability – outcomes transferable to the Indonesian logistics context if appropriately localised.

A concrete entry point is Presidential Regulation No. 16 of 2018 on Government Procurement, which already contains sustainability provisions but only as optional. Revising this regulation to mandate life cycle costing and eco-friendly standards would provide regulatory clarity and align with Indonesia’s National Action Plan for the Sustainable Development Goals (SDGs).

Expanding government incentives (KH2)

Fiscal incentives are pivotal in overcoming the initial capital barriers associated with transitioning to GP. Leal et al. (2020) found that green public procurement practices, combined with fiscal incentives and policy mandates, encourage broader market participation and long-term investments in sustainability. Dimand (2022) underscores linking government policy with organisational leadership to drive sustainable change. Mechanisms such as grants, performance-based subsidies and subsidised loan programmes can reduce investment risks and encourage firms to reorient their procurement strategies. Mahardhika (2024) confirms that such measures are vital to expanding corporate participation and scaling up GP initiatives.

In Indonesia, this strategy can be operationalised by broadening existing policy instruments beyond the current electric vehicle subsidies (introduced in 2023) towards logistics-specific incentives. Firstly, the government could establish a dedicated Green People’s Business Credit (KUR) scheme – a government-backed loan facility with subsidised interest rates for micro, small and medium enterprises (MSMEs) – to support sustainable logistics investments such as eco-friendly vehicle fleets and green warehousing. Secondly, matching-fund programmes could be developed between the government and major state-owned enterprises (SOEs), including the national postal company Pos Indonesia and the Indonesia Port Corporation (Pelindo), to implement pilot projects in GP and logistics infrastructure. Thirdly, corporate income tax reductions could be granted to companies that can demonstrate that at least 30% of their procurement volume is sourced from certified eco-friendly products. Collectively, these mechanisms would reduce the financial risks associated with transitioning to sustainable practices while creating more substantial incentives for logistics firms to align with Indonesia’s long-term green development agenda.

Integrated barrier dynamics and policy implications

The integration of DEMATEL and ISM methodologies revealed which barriers matter and how they interact across the procurement system. The hierarchical representation and the influence pathways in Figure 3 reinforce the idea that upstream barriers like KP1 and KH2 trigger a cascade of challenges at higher structural levels. This insight aligns with Guenther et al. (2010), who argue that causal barriers, if unresolved, perpetuate systemic inefficiencies across interdependent processes.

Policymakers can allocate resources more effectively by identifying root causes and tracing their effects. For example, investing in KH2 and KH3 reforms yields multiplier effects, influencing mid-level barriers such as KO2 and KH1 and then operational outcomes like PI1 and PI3. This causal logic supports a phased intervention model, addressing systemic enablers first and thus ensuring sustainable improvement across the logistics procurement ecosystem.

Furthermore, dependent barriers such as PI3 (failure to integrate policies) and KK2 (fear of implementation) highlight the importance of internal organisational transformation. As illustrated in the ISM model, these barriers are consequences of unresolved foundational challenges. Addressing them through training, internal policy development and leadership alignment becomes effective only after the base-level drivers have been mitigated.

The discussion also underscores the importance of cross-sectoral collaboration. Logistics firms, suppliers, government agencies and international partners must engage in structured dialogue to co-develop GP criteria and implementation strategies. To move beyond a generic call for cooperation, this collaboration could be institutionalised through multi-stakeholder forums jointly led by the Ministry of Industry, the Ministry of National Development Planning, the Ministry of Investment, the Ministry of Environment and Forestry and the Ministry of Transportation. Such forums should be mandated to set measurable targets, for example, increasing the proportion of GP within national and regional government budgets, while also monitoring compliance and fostering continuous innovation across the logistics sector.

The national context significantly shapes the operational effectiveness of GP. Letunovska et al. (2023) and Pinto (2023) report that developed countries benefit from comprehensive policy support, stable institutions and advanced technological infrastructure, which streamline GP integration. In contrast, developing countries – including Indonesia – face barriers such as limited regulatory enforcement, fragmented institutional mandates and constrained access to sustainable alternatives (Ali et al. 2020; Dahliani et al. 2023). These differences lead to variation in adoption levels, emphasising the need for context-sensitive policy formulation.

Conclusion

Based on a systematic literature review regarding applying the concept of GP in various industries, this study identified 20 barriers. The validation process, a crucial step in our research, confirmed the relevance of about 14 barriers, which were then involved in further analysis. A structural model and cause-and-effect diagram were then developed for the 14 barriers using a combined DEMATEL-ISM method to identify the main barriers. Four barriers have the most significant impact, including ‘Lack of availability of eco-friendly materials in the local market’; ‘Lack of supplier readiness in the supply of environmentally friendly materials’; ‘There are no relevant regulations related to GP work procedures and criteria for environmentally friendly materials in the logistics industry’ and ‘Lack of incentives provided by the government for the adoption of GP in the logistics industry’. These barriers must be overcome first to overcome other barriers effectively. These barriers significantly influence a network of dependent challenges, making them strategic leverage points for intervention.

The analysis demonstrated how these causal factors propagate through the procurement system to shape various outcome-based barriers, such as increased costs and poor policy integration. The hierarchical structure produced by the ISM model and the causal pathways mapped through DEMATEL offer a clear roadmap for prioritising interventions. Strategic recommendations emphasise the importance of enhancing supplier capabilities, investing in eco-material supply chains, adopting international best practices in procurement legislation and expanding government incentive schemes. The study contributes to the academic discourse by applying a robust methodological framework rarely used in logistics and offering a sector-specific understanding of GP barriers. The findings serve scholarly inquiry and policy development, enabling more targeted, systemic reforms.

This study has several limitations, primarily because of the limited body of research on barriers to implementing GP in the logistics industry. As a result, the identification of barriers was partly informed by findings from other industries and subsequently validated by expert assessments. This reliance introduces the potential for bias in self-reported expert judgements, where subjective perceptions may not fully capture the operational realities of the logistics sector. Future research should address this by expanding the expert panel to include a more diverse range of stakeholders and conducting in-depth focus group discussions. These steps would minimise bias, improve generalisability and capture a broader range of perspectives.

Furthermore, the study does not incorporate quantitative validation against actual procurement performance data, which limits the robustness of the findings. The absence of such empirical benchmarking reduces the ability to confirm whether the identified barriers translate directly into measurable inefficiencies or costs in practice. Future studies should therefore combine expert-based approaches with quantitative analyses, such as procurement performance metrics, cost-efficiency indicators or environmental impact data, to provide stronger empirical validation.

In addition, the methodological approach does not fully capture unmodelled dynamic feedback loops, meaning that the interactions between barriers may evolve differently over time than suggested in the current system mapping. Future research should adopt longitudinal designs or system dynamics modelling to explore how barrier interactions change over time, allowing for the identification of feedback effects and evolving causal pathways.

Another critical limitation concerns the absence of a sensitivity analysis or robustness check to strengthen confidence in the prioritisation of barriers. While we recognise the value of such an analysis, it was not performed in this study because of the complexity of the DEMATEL model, the focus on identifying significant barriers and the limitations in available data. Future research should address this gap by conducting sensitivity analyses – testing different thresholds for significant influence, recalculating barrier rankings under alternative weighting assumptions and applying techniques such as Monte Carlo simulations or bootstrapping. Scenario-based sensitivity tests could also be used to explore how results vary under different regulatory or market conditions. These approaches would enhance the robustness and reliability of conclusions. In addition, this study was limited to examining only the four primary barriers identified through the DEMATEL-ISM analysis, namely the lack of eco-friendly material availability (KP1), insufficient supplier readiness (KP2), the absence of relevant regulations (KH3) and the lack of government incentives (KH2). This scope will therefore be acknowledged as a limitation of the study. Furthermore, the study does not provide an in-depth discussion on how logistics firms could overcome internal barriers, such as organisational culture, without external support. Future studies should explore strategies for addressing these internal barriers independently, in conjunction with the systemic issues identified here, to provide a more comprehensive framework for advancing GP in the logistics sector.

Acknowledgements

Thank you to the Dean of the Faculty of Engineering at Diponegoro University for funding and providing the opportunity to conduct research on the implementation of green procurement in the logistics sector.

The authors declare that they have no financial or personal relationships that may have inappropriately influenced them in writing this article.

Aries Susanty: conceptualisation, methodology, formal analysis, writing – original draft, data curation, writing – review and editing, supervision and funding acquisition. Nia B. Puspitasari: formal analysis, visualisation, project administration, data curation and resource allocation. Darminto Pujotomo: writing – original draft, visualisation and writing – review and editing. Bambang Purwanggono: methodology, writing – original draft, and writing – review and editing. Muhammad S. Akbar: methodology, investigation, writing – original draft, software development, data curation and resource provision. All authors reviewed the article, contributed to the discussion of results, approved the final version for submission and publication, and take responsibility for the integrity of its findings.

Ethical clearance to conduct this study was obtained from Diponegoro University and the Health Research Ethics Committee (No. [192/EA/KEPK-FKM/2025]).

This research received a Strategic Grant from the Faculty of Engineering, Diponegoro University No.30/UN7.F3/HK/V/2025.

The data have been included in the manuscript, and if more details are required, please contact the corresponding author.

The views and opinions expressed in this article are those of the authors and are the product of professional research. They do not necessarily reflect the official policy or position of any affiliated institution, funder, agency or publisher. The authors are responsible for this article’s results, findings and content.

References