Toward a theoretical framework for sustainable third-party logistics provider selection: an exploratory study using AHP in the Turkish machinery industry

Adopting sustainable practices has become an increasingly important strategic priority for organizations worldwide in the face of climate change, growing regulatory demands and rising customer expectations (Abbass et al., 2022; Singh et al., 2022). The integration of sustainability into supply chain operations has emerged as a critical challenge for firms, particularly in the area of logistics outsourcing. Third-party logistics (3PL) providers play a vital role in enabling companies to achieve supply chain sustainability by offering logistics services that align with environmental, social and economic performance goals (Govindan et al., 2013; Winter and Knemeyer, 2013). Despite the importance of 3PL selection, previous research has largely focused on evaluating service providers using descriptive criteria such as cost, quality, flexibility and reliability (Raut et al., 2018; Paul et al., 2020). Although multi-criteria decision-making (MCDM) methods such as AHP, fuzzy AHP and DEA have been applied to prioritize these factors (Zarbakhshnia et al., 2020), much of the prior work remains practice-oriented, with limited connection to established theoretical frameworks. There remains a significant research gap in understanding how theoretical perspectives can systematically explain the drivers of sustainable 3PL provider selection, especially in emerging markets (Zimmer et al., 2016). In particular, prior studies have discussed sustainability in logistics and supply chain management, but empirical evidence on how decision-makers weight economic, environmental and social criteria in specific industrial settings remains limited. This study does not claim a new theory; instead, it uses established perspectives to organize the criteria and to interpret the priorities observed in the empirical analysis.

This study addresses this gap by proposing a theory-based framework for sustainable 3PL provider selection that draws on the resource-based view (RBV), transaction cost economics (TCE) and stakeholder theory. The RBV suggests that firms outsource logistics to acquire resources and capabilities that are valuable, rare, inimitable and non-substitutable (Barney, 1991). TCE posits that outsourcing decisions aim to minimize transaction costs under conditions of uncertainty and asset specificity (Williamson, 1987). Recent studies in emerging-market settings also emphasize cost transparency and risk-related considerations in logistics-related evaluation (Chakraborty et al., 2024; Kiani Mavi et al., 2024). Stakeholder theory highlights the influence of external pressures from governments, customers and communities, which encourage firms to consider environmental and social sustainability criteria when selecting logistics partners (Freeman, 2010). Workforce-related issues such as training and health and safety are also increasingly discussed as practical social sustainability priorities in logistics evaluations (Phan Ha et al., 2024; Roy et al., 2020b).

To explore the application of these theories in a real-world setting, this study presents an empirical AHP-based investigation of sustainable 3PL provider selection criteria in the Turkish machinery industry. As one of Turkey's largest export sectors, the machinery industry is both a critical economic player and an emerging adopter of sustainability initiatives (Presidency of the Republic of Turkey Investment Office, 2025; İncekara et al., 2023). The Turkish machinery sector is closely connected to European markets and EU sustainability initiatives such as the European Green Deal and the Carbon Border Adjustment Mechanism (CBAM) are expected to shape exporters' compliance requirements and cost structures (World Bank, 2020; Acar et al., 2022). At the same time, firms operate under domestic cost pressure and uncertainty, which can heighten the salience of economic and risk-related considerations in outsourcing decisions (Cihan and Kabak, 2024). However, sector-specific and theory-informed empirical evidence on sustainable 3PL selection in this industry remains limited, which motivates this study.

This study aims to answer three key research questions:

The key contributions of this study are twofold. First, the paper proposes a theory-based conceptual framework to explain sustainable 3PL selection. Second, it provides exploratory empirical insights using AHP from industry practitioners to validate the proposed framework's relevance. By linking a practical prioritization exercise with established theoretical perspectives, the study offers a basis for subsequent validation in other sectors and settings.

The remainder of the paper is structured as follows. Section 2 reviews the relevant literature and presents the proposed theoretical framework and research propositions. Section 3 describes the research methodology and data collection process. Section 4 presents the empirical results and discusses them in relation to the theoretical model. Section 5 offers conclusions, theoretical and managerial implications, limitations and directions for future research.

This section reviews prior work on (1) third-party logistics (3PL) and provider selection, (2) sustainability criteria in logistics outsourcing and (3) MCDM approaches used to prioritize selection criteria. Building on existing theory-linked logistics literature, this study uses RBV, TCE and stakeholder theory as complementary lenses to organize sustainability-related criteria under the triple bottom line and to interpret prioritization patterns observed in the Turkish machinery context.

In supply chain management (SCM), 3PL refers to the use of external firms, either in its entirety or as part of logistics functions that have been traditionally carried out by the firm (Hertz and Alfredsson, 2003). In this sense, any externalizing of logistic services initially carried out by the organization includes 3PL (Skjoett-Larsen, 2000). Logistics service providers (LSPs) are increasingly critical not only for transportation and warehousing but also for other value-added services such as packaging, configuration, assembly, quality control and reverse logistics, which adds extra value to goods (Christopher, 2016).

Outsourcing logistics services to 3PL allows firms to focus better on their core capabilities, lower operational costs and improve service levels (Christopher, 2016). According to other studies, outsourcing reverse logistics to 3PL providers has proven to be economically and socially beneficial for many companies (Govindan et al., 2012). Therefore, many 3PL providers are shifting from cost-driven tactics to differentiation strategies based on service (Skjoett-Larsen, 2000) and emphasize being more responsive to customer needs (Tian et al., 2010).

In recent years, it has become increasingly vital to form alliances with 3PL providers for sustainability objectives to gain a broader perspective on creating corporate value through carbon tax reduction (Roy et al., 2020b). Evans et al. (2017) suggested that the integration of sustainability at the network level and achieving shared and individual goals within the network could be enhanced by new governance mechanisms. Although sustainability strategies in product supply chains have been thoroughly studied in the past, little is known about 3PL sustainability. Through a case study involving a shipper, a 3PL company and a hauler, Forslund et al. (2021) fill this gap by investigating sustainability practices and managerial problems in a transport supply chain. The study places strong emphasis on teamwork in enhancing sustainability practices. In addition, redesigning the firm's purpose as part of the value network could enable innovations in new sustainable business models.

Prior studies often report similar operational criteria across sectors, but the rationale behind those criteria is not always made explicit. In this study, RBV, TCE and stakeholder theory are used as interpretive lenses to clarify why capability-oriented, cost/risk-related and legitimacy-driven criteria may be emphasized in 3PL selection.

To address climate change and sustainability, many organizations are considering sustainable measures for their supply chains which can create additional logistics challenges (Zhou et al., 2008). This requires the integration of environmental and social sustainability into corporate operations and collaboration with 3PL providers who are both economically affordable and committed to social and environmental sustainability goals (Winter and Knemeyer, 2013). Businesses should examine the environmental and social sustainability of their 3PL operations, and they should incorporate these considerations into their selection decisions and other supply chain operating decisions (Govindan et al., 2019).

A sustainable approach shapes customers' perceptions, especially when it comes to critical areas such as the company's image, brand value and loyalty behavior (Gupta and Kumar, 2013; Ailawadi et al., 2014). In turn, it influences how customers evaluate businesses. Although it is evident that social and environmental responsibility efforts help increase an organization's profit, sustainable practices significantly impact customer expectations and behaviors (Aliakbari Nouri et al., 2018). Jamali and Rasti-Barzoki (2019) utilized game theory to study product price, greenness and third-party product variables. They found that supply chain competition is crucial for increased profitability and sustainability, with manufacturers earning the most money in the decentralized Stackelberg game. The other research conducts a systematic review of the literature on knowledge management (KM) and KM strategies in LSP' environmental sustainability efforts (Evangelista and Durst, 2015). These discussions imply that sustainability priorities can translate into concrete selection expectations that shippers ask logistics partners to meet.

As customers' priorities shift and the road haulage industry becomes more fragmented than before, one study examines how LSP activities affect the environment in Europe. Further, it investigates how effective LSPs are at monitoring the supply chain's logistics activities. As evidenced by the findings, customer concern over environmental solutions that increase the cost and length of logistics services is not yet a reality. Furthermore, it is suggested that LSP sustainability cannot be studied in isolation if a company does not manage its proprietary resources (such as trucks and drivers) but rather engages subcontractors. To fully appreciate the scale of the sustainability challenges confronting their corporations and markets, business leaders and executives must conduct an audit on the rising needs and expectations set by sustainability's TBL environmentally responsible behavior and enjoy the financial benefits as a result (Gimenez et al., 2012). Furthermore, SCM has strong interrelationships with the TBL of sustainability and the circular economy notion because many ecological and social elements can only be comprehended throughout the entire supply chain (Govindan and Hasanagic, 2018).

Finally, incorporating the principles of the triple bottom line (TBL) and sustainability into decision-making frameworks offers robust theoretical support for evaluating sustainable practices in logistics and supply chain management (LSCM). Developed by Elkington (1997), TBL emphasizes the interconnected nature of economic, environmental and social dimensions, providing a comprehensive perspective for assessing the multifaceted impacts of logistics decisions. This approach ensures that 3PL provider selection considers not only economic factors but also integrates environmental and social concerns, aligning with broader sustainability objectives. Additionally, stakeholder theory complements this perspective by emphasizing the importance of addressing the needs and interests of all stakeholders impacted by logistics decisions, further enhancing the alignment with sustainability principles.

Moreover, sustainability, as defined in the Brundtland Report (WCED, 1987), underscores resource preservation and long-term goals, fostering intergenerational equity. In this study, the TBL perspective is used to organize the selection criteria, while the prioritization procedure (AHP) is described in the methodology section.

Choosing the most appropriate 3PL requires consideration of a variety of criteria. Previously, the evaluation and selection processes relied heavily on subjective judgment based on the decision-makers’ individual perceptions of 3PL providers, with no theoretical underpinning or consideration of systems. As a result, firms employ various strategies to analyze, evaluate and select their 3PLs. Falsini et al. (2012) developed an AHP–DEA–linear-programming approach to support the evaluation and selection of 3PL based on a wide range of criteria. The model considered seven factors: quality assurance, service speed, flexibility, cost, equipment, worker safety and environmental protection. Table 1 provides the summary of these empirical studies and their respective methodologies. As Table 1 illustrates, most researchers have focused on three sustainable dimensions. In addition, some of them also studied other criteria such as risk and safety or operational dimensions. Ultimately, there are various methods to evaluate their research, such as fuzzy AHP, VIKOR, DEA and ANP. A unique hybrid, multiple-attribute decision-making (MADM) technique has also been proposed, combining the fuzzy analytic hierarchy process (fuzzy AHP) with gray multi-objective optimization via ratio analysis (MOORA-G). The findings of the study show that their suggested hybrid technique is more viable when dealing with qualitative data (Zarbakhshnia et al., 2020). Recent logistics MCDM studies increasingly adopt fuzzy/hybrid approaches (e.g. Nila et al., 2024; Nila and Roy, 2024), so we complement standard AHP with a sensitivity analysis to check ranking stability.

Notably, the supply chain has become more complex because of technological advancement and globalization in the current business environment. These initiatives have piqued the interest of policymakers who are concerned about environmental issues. For example, Ali et al. (2019) examine the role of sustainability and utilizing SEM, they found that except for packaging, which has no measurable impact, transportation and warehousing operations significantly impact environmental, economic, social and operational performance. The empirical results revealed no significant association between these factors except for social performance. As a result, decision-makers now better understand 3PL industries and may use that knowledge to solve sustainability problems (Ali et al., 2019). The environmental policies of various LSPs may look similar, but in practice, they differ significantly. Environmentally conscious logistics service buyers should pay attention to this variation in practice because it emphasizes the importance of ongoing follow-up inspections (Nilsson et al., 2017).

Zarbakhshnia et al. (2020) developed a model to outsource sustainable reverse logistics utilizing fuzzy AHP and MOORA-G. When people make decisions in the real world, they often receive inputs that are both qualitative and uncertain. Thus, the researchers used fuzzy and gray numbers to address these inputs. In addition to the TBL dimensions, the researchers also looked at risk and safety as dimensions. Roy et al. (2020b) constructed a MCDM model based on IVFRN by integrating factor relationship (FARE) and multi-attributive border approximation area comparison (MABAC) models with three sustainable dimensions. Raut et al. (2018) employed an integrated strategy of data envelopment analysis (DEA) and analytic network process (ANP) as an evaluation and selection methodology. Paul et al. (2020) created a new decision-making framework for evaluating transportation service providers in terms of economic, environmental, social, and operational sustainability. The decision-making framework incorporated both qualitative and quantitative expert opinions, as well as the best-worst method (BWM) and Višekriterijumsko kompromisno rangiranje (VIKOR) methods.

During the past 2 decades, many researchers have attempted to determine the most important factors for evaluating and 3PL providers. To capture the most significant parameters, researchers typically use survey methods. Zarbakhshnia et al. (2020) used three main dimensions and risk and safety as the fourth dimension. Paul et al. (2020) developed a total of 4 dimensions and 28 sub-criteria. It was observed that the operational aspect was the most important factor in evaluating a transport service provider. Roy et al. (2020b) identified 3 main criteria and 15 sub-criteria. In order to do so, they used a framework to select the best logistics provider.

Most of the previous research has focused primarily on pricing (cost) and service quality for evaluating 3PL providers. One of the recent research conducted to identify the most suitable 3PL company selection criteria in the food sector using fuzzy MCDM methods and using economic criteria (Erdem et al., 2023). In contrast, the adoption of social sustainability-related criteria, such as labor or management policy-related issues, has been tested in a few pilot projects. Contrastingly, the social benefits outlined in their research are more concerned with brand reputation than with social sustainability. The training and management capabilities of logistics providers were employed as evaluation factors for logistics provider selection in the Taiwanese automotive industry, according to Huang et al. (2013). Although some previous related research attempted to look at social sustainability-related criteria when evaluating 3PL providers, the majority of research only analyzes individual criteria that were primarily focused on labor relations instead of specific associated corporate governance within the company. Furthermore, other social sustainability topics (such as philanthropy and workplace health) were not considered significant in previous related research. As a result, there has been a gap in analyzing the full range of social sustainability as an essential criterion for evaluating 3PL providers (Jung, 2017).

There are several studies that look at selection and evaluation of 3PL in Turkey (Ecer, 2017; Özcan et al., 2018). Recent studies also point to the role of digital capabilities in logistics provider selection. For example, Şahin et al. (2025) highlight criteria related to automation and digital collaboration, which can support monitoring and reporting of operational and sustainability performance. An integrated cognitive map-based intuitionistic fuzzy multiple criteria decision aid was introduced by Goker (2021) which can be used to rank agile outsourcing provider options and to find the best performing provider among them. To evaluate Third-party suppliers of transportation services, Yayla et al. (2015) offer a hybrid technique that incorporates Buckley's extension and FMCDM. Bulgurcu and Nakiboglu (2018) looked at the selection criteria used by LSPs in Turkey's cement sector to choose Third-party provider and research only focused on economic factors. According to previous research (Ulutaş et al., 2024; Erdem et al., 2023) in Turkey's context, studies on sustainability of 3PL are lacking, which demonstrates a need for further research on sustainability.

MCDM approaches have been utilized in a few studies to analyze ecological sustainability issues for 3PL. However, this research is critical because there is a lack of research on expanding logistics sectors from an environmental and social standpoint. Furthermore, the strategies and processes outlined above necessitate large amounts of data and employ repetitious techniques, making the approach highly complex and time-consuming. Thus, much more effort on connecting research with actual practical application on the 3PL evaluation and selection process is required and needed to assist decision-makers and provide good theoretical backing. AHP is a flexible analytical technique that enables decision-makers to identify the best feasible solution to complicated situations by breaking down a problem into a systematic and hierarchical network of inter-relationships among the many levels and criteria. Essentially, literature has found that evaluating logistics service effectiveness parameters are interconnected (Kayakutlu and Buyukozkan, 2011). Therefore, the paper integrated the analytical hierarchy method to help logistics managers and decision-makers incorporate environmental sustainability and social criteria while selecting the most appropriate 3PL from an ecological standpoint.

While recent studies increasingly employ hybrid and fuzzy MCDM models, AHP remains widely used as a transparent approach for structuring criteria and eliciting expert judgments through pairwise comparisons. In this study, AHP is used to derive relative priorities in a context where expert evaluation is central, and the stability of the ranking is examined through a sensitivity analysis reported later.

The AHP method was used to analyze the survey in this research. This study uses standard AHP because it provides a transparent procedure for structuring criteria and eliciting expert judgments through pairwise comparisons, which is suitable when decision-makers’ assessments are central to the analysis. Recent applications also show that AHP continues to be used in industrial evaluation settings (Cihan and Kabak, 2024; Chakraborty et al., 2024). In addition, a decision-making framework was created using the AHP method. Figure 1 shows the systematic research steps used to construct the research design. According to the research design, if pairwise comparisons are not consistent, the decision hierarchy needs to be checked and constructed again.

To prioritize selection criteria, this study employs the AHP, which structures complex decisions and derives relative priorities from expert judgments (Cihan and Kabak, 2024). As demonstrated in recent frameworks for automotive tooling, AHP effectively structures complex decision-making problems where empirical performance data may be scarce. Following the collection of quantitative and qualitative data for the AHP supplier selection model, companies could use the following procedures to select the 3PL that is more suitable than the others. A snowball sampling (non-probability sampling approach) method was utilized to obtain the data. Given the specialized nature of logistics decision-making in the machinery sector, snowball sampling was used to identify qualified decision-makers through professional networks. The recruitment funnel was as follows: 21 invitations were distributed via email, 15 complete responses were received (71.4% initial response rate) and 5 respondents were excluded because they had less than five years of relevant industry experience. This resulted in a final qualified expert panel of 10 participants (n = 10), corresponding to 47.6% of the initial invitations (10/21). The respondents were managers or people who worked in the department and had enough experience to participate in the survey.

The research reviewed the literature on selecting appropriate LSPs to develop a complete framework for determining the most often utilized 3PL selection criteria; for the sub-criteria for supplier selection, preliminary verification was conducted with experts in their fields. The first survey was based on an in-depth interview and a five-point Likert scale. According to the findings of an in-depth interview and questions based on five-level Likert scale with the manager of the supply chain department, there are 15 sub-criteria that are important and necessary for selecting a 3PL provider.

Table 2 describes the main criteria and each sub-criterion with relevant references. The AHP model was built based on the main criteria and sub-criteria that were identified. As shown in Figure 2, the selection criteria hierarchy model has three levels: goal, main criteria and sub-criteria. Each main criterion has five sub-criteria. A hierarchy organizes procedures according to the organization's objectives.

MCDM is a modeling methodology that may be utilized to make choices while dealing with complex technical matters. One way to measure decisions is through the AHP, as discrete or continuous paired comparisons can be used to construct the ratio scales. A basic scale representing the relative intensity of preferences and feelings may also be employed. AHP is a hierarchical decision-making method that derives priority weights from pairwise comparisons and evaluates the internal consistency of those judgments. Nonetheless, the AHP is advantageous in multicriteria decision-making, planning, resource allocation and dispute resolution. Psychologists claim that expressing one's view on only two options is easier and more accurate than expressing one's opinion on all the alternatives simultaneously. In addition, the AHP facilitates consistency and cross-checking across many pairwise comparisons.

• Step 1: Define the problem and identify relevant criteria.

• Step 2: Construct the hierarchy and development of a pairwise comparison matrix of criteria.

• Step 3: Assess the consistency with which the solution has been achieved. In order to make sure that pairwise comparisons are consistent, a consistency ratio must be calculated. It was calculated using mathematical expressions given as CR=CI/RI. To assess whether decisions are coherent and reliable, the CR value must be used.

Judgments were considered acceptable when CR ≤ 0.10; otherwise, the pairwise comparisons were revisited.

• Step 4: For the main dimensions and criteria, normalization of weight should be determined. The dimensions have been ranked according to the normalized weight.

When ten professional decision-makers evaluated each judgment on the three main dimensions and 15 sub-criteria, the results were compiled exclusively as a matrix. The method starts by organizing a decision-making problem in the shape of an upside-down tree, with the primary goal at the top. The second level included incomplete tasks that contributed to the overall aim. It is possible to break down the second-level objectives into third-level objectives by looking at them from the level to which they are subordinate. These unfulfilled goals were used as the criteria in this study. The choices are given and then evaluated pairwise based on their contribution to accomplishing each lower-level objective or criterion at a lower level.

For conducting this research, four companies were surveyed. Company A, B and D have more than 200 employees, and C had less than 200 employees. A was the largest manufacturer in terms of product. However, C was a small manufacturer with 100–200 employees and had a share of international and national trade in Turkey. With four companies with different backgrounds, this survey displayed different perspectives, which helps to develop better conclusions for the criteria. Table 3 shows the details of the companies that have been surveyed, including the name of the company, the number of employees of the company and the products that they are producing.

Data were collected using snowball sampling to reach qualified decision-makers in the machinery sector. In total, 21 invitations were distributed via email and 15 complete responses were received (71.4%). To ensure that respondents had sufficient industry experience for the AHP comparisons, five responses were excluded because they did not meet the minimum five-year experience criterion. This resulted in a final panel of 10 experts (n = 10), equivalent to 47.6% of the initial invitations (10/21).

While this sample size is numerically small compared to statistical surveys, it is consistent with established AHP methodological standards which prioritize the depth of expert consistency over sample breadth. Similar sample sizes (ranging from 6 to 10 experts) have been successfully utilized in recent logistics MCDM studies to derive robust results (Safavi et al., 2025; Nila and Roy, 2023).

Firstly, analysis begins with the evaluation of the main criteria. The decision group assigned the pair-wise comparisons, and to obtain the priority vector, each priority vector value was multiplied by the sum of every element. After identifying the critical criteria, they were grouped into matrices. Each criterion has its own rows and columns in the AHP matrices. After calculating the consistency index, it was divided by the random index. A consistency ratio (CR) of 0.0043 was obtained, which indicates acceptable consistency (CR ≤ 0.10). The entire matrix is then used to calculate the importance weights, which indicate the relative relevance of each criterion in the manager's decisions. The greater the weight assigned to a particular criterion, the greater its influence on the manager's final selection. Out of the three dimensions for selecting 3PL, economic criteria are given the highest priority, receiving a weight of 77.63%. After that, environmental criteria are second with 12.56%, followed by social criteria, which is weighted at 9.81%. These results indicate that economic considerations receive the highest emphasis in the surveyed decision context, while environmental and social criteria follow.

After calculating the main criteria, the sub-criteria were calculated to find out their consistencies and weights. Economic sub-criteria consistency ratio came up as consistent and the ratio is 0.0046. To enable direct comparison across all criteria, the following values are reported as global weights (Table 4). Cost is the most important factor, accounting for 29.31% of the total. After that, Quality is second, with a weight of 17.89%, followed by Reliability (16.05%), Lead Time (9.28%) and Flexibility (5.09%). This finding suggests that cost- and service-related performance criteria dominate the overall decision priorities in the surveyed context. Logistics providers require competitive pricing to lower transportation costs and meet basic service expectations related to quality and reliability.

The environmental criteria findings have more interesting results. Consistency ratio is 0.0066. When the environmental sub-criteria weights were calculated, Environmental Legal and Policy Framework had the highest global weight (3.17%), followed by Green Packaging (2.83%). Carbon Footprint (2.45%) and Resource Consumption (2.43%) had similar global weights, while Green Transportation Channel had the smallest global weight (1.67%). According to the results, the policy framework was higher than others, and it could be because it is easier to verify and has legitimacy from governmental organizations. The second sub-criterion was green packaging. In preliminary interviews with decision-makers, some respondents noted that green packaging was beneficial because they could use the packaging in their warehouse or for other transportation reasons.

The social criteria show consistency with 0.0089. The AHP technique yields social criteria weights, which show that, among the five social criteria for selecting 3PL, Health and Safety is given the highest global weight (2.85%), followed by Reduction of Accidents (2.49%). Investment in Local Community (1.63%) and Client Relationship (1.49%) follow, while Interests and Rights of Employees has the smallest global weight (1.35%). This pattern is consistent with prior work that discusses workforce-related considerations – such as training and safety – as salient social sustainability issues in logistics evaluations (Phan Ha et al., 2024; Nila and Roy, 2023). Overall, the relatively high weights assigned to Health and Safety and Reduction of Accidents indicate that respondents place substantial emphasis on safety outcomes during logistics operations. Client relationship is ranked lower than other social criteria, which suggests a potential topic for future research, such as examining why relationship-oriented considerations are weighted less than safety-related concerns in this context.

The findings of this study reveal that the economic dimension is the most vital criterion for selecting a logistics provider, followed by environmental and social criteria. Table 4 reports the global weights and overall ranking across all criteria, which allows direct comparison of sub-criteria. Overall, environmental criteria receive a higher combined weight than social criteria in the surveyed sample. However, carriers who demonstrate social performance may be viewed as appealing employers who can attract and maintain a highly skilled workforce (Ehrgott et al., 2011). This suggests that social performance – particularly safety-related practices – may matter for workforce stability and service reliability in logistics operations.

To examine the stability of the results, a sensitivity analysis was conducted by perturbing the weights of the three main dimensions (Economic, Environmental, Social) by ±10%, generating six scenarios. In each scenario, one main-dimension weight was adjusted by ±10% as a relative change (w × 1.10 or w × 0.90), and the remaining main-dimension weights were re-normalized so that the three weights sum to 1. For each scenario, the global weights and the resulting rankings were recalculated. Across all scenarios, the top-ranked sub-criteria (Cost, Quality, Reliability, Lead Time and Flexibility) remained unchanged. At the same time, minor rank changes occurred among some mid-ranked criteria with very similar weights (e.g. Health and Safety and Green Packaging). Overall, the sensitivity check suggests that the main conclusions are not driven by small changes in the main-dimension weights, while criteria with close weights may show limited rank swapping (Table 5).

The purpose of this study was not only to report an AHP ranking but also to move toward a theory-informed framework that clarifies how decision-makers interpret sustainability criteria when selecting a 3PL provider in a specific industrial setting. The results provide an ordered structure of priorities across the TBL, and they also indicate how economic, environmental and social considerations can be understood through complementary theoretical lenses. In other words, the weighting pattern is useful as empirical evidence, but its value for research is stronger when the criteria are treated as mechanisms that reflect distinct strategic logics, rather than as a simple checklist.

From a TCE perspective, the dominance of the economic dimension and the high ranking of Cost, Quality and Reliability are consistent with an outsourcing logic that emphasizes uncertainty reduction and safeguarding. In logistics outsourcing, the “price” of a 3PL relationship is not limited to a quoted transportation rate; it also reflects the expected costs of monitoring, coordination, renegotiation and service recovery when disruptions occur. The relatively high global weights assigned to Reliability and Quality, alongside Cost, can therefore be interpreted as indicators of transaction risk management. A 3PL that performs reliably reduces exposure to late deliveries, rework, and customer dissatisfaction, all of which translate into additional transaction costs. This interpretation aligns with the broader outsourcing literature that treats operational stability as a form of risk mitigation under conditions of uncertainty and asset specificity, and it helps explain why cost-related criteria remain salient even in a sustainability-oriented evaluation.

The RBV provides a complementary interpretation of the same pattern. While TCE emphasizes safeguarding and cost control, RBV draws attention to capability acquisition through logistics outsourcing. In this study, the weight structure suggests that firms value 3PL partners for their ability to support core operational outcomes—especially service quality and reliability—rather than for isolated functional attributes. In RBV terms, these criteria can be read as proxies for the provider's logistics capabilities: process discipline, operational know-how and coordination ability that enable stable delivery performance. The prominence of Lead Time and Flexibility, even below the top three, reinforces this point. These are not purely “service features”, but reflect adaptive logistics capabilities that allow manufacturers to respond to demand variation and operational constraints. Therefore, the findings support a framework in which outsourcing to a 3PL is partly a capability strategy: firms rely on external logistics resources to sustain performance in their own production and distribution systems.

Stakeholder theory adds a third, distinct layer to the framework by clarifying why environmental and social criteria enter selection decisions even when they do not dominate overall weights. The environmental results are especially instructive. Within the environmental dimension, Environmental Legal and Policy Framework receives the highest global weight, followed by Green Packaging. This pattern is consistent with a legitimacy-oriented selection logic in which decision-makers prioritize criteria that are auditable and externally defensible. Compliance with environmental regulations and policy expectations is visible to regulators and customers, and it can be documented more readily than broader claims about environmental performance. In stakeholder terms, such compliance-related criteria function as legitimacy safeguards: they help firms demonstrate that outsourcing decisions do not expose them to reputational or regulatory risk. Green Packaging, which also ranks highly within the environmental criteria, appears to be interpreted as an actionable and operationally tangible sustainability practice. The preliminary interview comments suggest that decision-makers associate packaging with warehouse and transport routines, which makes it easier to translate into procurement requirements than more abstract environmental indicators. The broader implication is that stakeholder pressure does not necessarily translate into “green” priorities in the abstract; rather, it tends to privilege environmental practices that can be verified and operationalized.

The social dimension provides a similarly grounded insight. Even though the overall social weight is lower than the environmental weight, safety-related criteria receive the highest global weights within the social category. Health and Safety ranks first, followed by Reduction of Accidents. This is not a generic appeal to social responsibility; it indicates that decision-makers interpret social sustainability in logistics primarily through workplace and driver safety outcomes. This pattern is consistent with prior logistics-related studies that emphasize training and safety as salient social sustainability issues in supplier or provider evaluation, and it supports the view that internal stakeholders (employees and drivers) remain central in operationally intensive industries. In addition, the link between safety-related practices and reliability is not accidental: safety procedures, training and incident prevention can be interpreted as upstream conditions for stable service delivery. In that sense, the social criteria contribute to the framework not only as ethical requirements but also as operational risk signals that can reinforce economic performance objectives.

Taken together, the findings support a theoretical framework in which sustainable 3PL selection involves three interlocking logics. The first is a transaction-cost logic: firms prioritize criteria that reduce coordination and disruption costs (Cost, Quality, Reliability and related service performance). The second is a capability logic: outsourcing is used to access and sustain logistics capabilities that support stable and adaptable operations (Quality, Reliability, Lead Time and Flexibility). The third is a legitimacy logic: environmental and social criteria matter most when they represent externally auditable compliance (Environmental Legal and Policy Framework) or operationally enforceable practices tied to key stakeholder expectations (Green Packaging and safety-related criteria). Importantly, this does not imply that firms disregard sustainability; rather, the framework suggests that sustainability criteria are filtered through concerns about verifiability and operational translation. In practical terms, decision-makers appear more willing to prioritize sustainability items that can be specified contractually, monitored routinely and defended to stakeholders.

The robustness check further supports the use of this pattern as a foundation for framework development. The sensitivity analysis indicates that the ranking of the top criteria remains stable under small perturbations of the main-dimension weights, while minor swapping can occur among mid-ranked criteria with very similar weights. This matters for theory-building because it suggests that the dominant logics – cost/risk control and service capability – are not artifacts of small weighting changes, whereas boundary areas between environmental and social criteria may be more sensitive to context and managerial interpretation. Future research can build on this point by examining when and why those mid-ranked sustainability criteria become more decisive, for example under stronger regulatory enforcement, different customer requirements or distinct contractual arrangements.

In sum, this study advances framework development by linking an empirical priority structure to established theory. Rather than treating sustainability criteria as a flat list, the combined use of RBV, TCE and stakeholder theory clarifies how different categories of criteria correspond to different strategic logics in 3PL selection. The Turkish machinery setting provides context-specific evidence for this interpretation, and the resulting framework offers a grounded basis for future studies to test how these logics shift across sectors, institutional environments and sustainability pressure conditions.

This study develops an empirically informed, theory-based framework for sustainable 3PL provider selection in the Turkish machinery context. The criteria were identified through a literature review and consultation with a supply chain manager, resulting in 15 criteria organized under three dimensions: economic (Cost, Quality, Reliability, Lead Time and Flexibility), environmental (Environmental Legal and Policy Framework, Green Packaging, Resource Consumption, Carbon Footprint and Green Transportation Channel) and social (Health and Safety, Reduction of Accidents, Investment in Local Community, Interests and Rights of Employees and Client Relationship). AHP was then applied to derive priorities based on expert judgments.

Regarding RQ1, the results show that the economic dimension receives the highest overall weight, followed by environmental and social dimensions. At the sub-criterion level, Cost has the largest global weight, followed by Quality and Reliability. Among sustainability-related criteria, Environmental Legal and Policy Framework and Green Packaging are the most prominent environmental items, while Health and Safety and Reduction of Accidents are the most prominent social items. Overall, the priority structure indicates that decision-makers in the surveyed setting give the greatest emphasis to criteria that directly shape cost and service continuity, while also prioritizing sustainability requirements that can be evidenced and implemented in practice.

Regarding RQ2, the dominance of cost and service performance is consistent with a TCE interpretation that firms seek to reduce uncertainty and safeguard outsourcing relationships. From an RBV perspective, the prominence of quality, reliability and lead time-related considerations suggests that firms value 3PL partners as capability providers that support stable and adaptable operations. Stakeholder theory helps explain why environmental and social criteria enter selection decisions in ways that emphasize verifiability and legitimacy, particularly through compliance-related environmental expectations and safety-related workforce concerns. Taken together, these lenses clarify how the same selection structure can reflect cost/risk control, capability acquisition and legitimacy maintenance, rather than simply listing preferences.

Regarding RQ3, this study proposes a framework that combines (1) a structured set of sustainability-related criteria under the TBL, (2) an empirically derived weighting structure from the Turkish machinery context and (3) a theory-based interpretation of the observed trade-offs using RBV, TCE and stakeholder theory. The robustness of the main conclusions was examined through a sensitivity analysis: a ±10% relative change (w × 1.10 or w × 0.90) was applied to one main-dimension weight at a time, and the remaining weights were re-normalized to sum to 1 before recalculating global weights and rankings. The top-ranked criteria remained unchanged across scenarios, while minor rank changes occurred among some mid-ranked criteria with similar weights. This robustness check supports the stability of the overall priority pattern and strengthens the framework as a basis for further testing.

This study's theoretical perspective positions sustainable 3PL selection as the outcome of three concurrent logics. TCE highlights the role of cost and service performance as safeguards against uncertainty and disruption in outsourcing relationships. The RBV frames quality, reliability and lead-time related factors as capability signals that firms seek from logistics partners to support stable and adaptable operations. Stakeholder theory explains why verifiable sustainability items – particularly compliance-related environmental expectations and safety-related social concerns – enter the selection structure as legitimacy and accountability requirements. Together, these lenses provide a concise interpretive basis for the priority pattern observed in the AHP results and clarify why some sustainability criteria become salient when they are operationally actionable and defensible to stakeholders.

From a managerial perspective, the model is intended to be directly useable. Table 6 translates the 15 criteria into a decision checklist by converting each criterion into concrete screening questions. Practitioners can use the checklist to structure requests for proposals and supplier discussions, document evidence for compliance-related criteria and compare providers consistently across economic, environmental and social dimensions.

Several limitations should be noted. The study focuses on the Turkish machinery industry and relies on a small expert panel, which limits generalizability. In addition, AHP captures expert preferences rather than observed operational performance. Future research could validate the framework using larger samples, additional industries or alternative methods (e.g. fuzzy AHP or ANP) where appropriate, and could also examine the provider side by assessing how 3PL firms design and improve sustainability practices in response to client expectations.