The empirical analysis draws on the ‘Supply chain analysis dataset’, obtained from the Kaggle open data repository. Transaction-level datasets of this type are widely used in supply chain and operations management research because they enable the simultaneous examination of cost-related and performance-related dimensions within a unified analytical framework (Chae et al. 2014; Kamble & Gunasekaran 2020).

The dataset includes variables related to order quantities, lead times, manufacturing and transportation costs, supplier performance indicators, and inventory levels, making it suitable for analysing operational efficiency, responsiveness, and reliability across supply chain activities (Gunasekaran et al. 2008; Maestrini et al. 2018).

However, it is important to acknowledge that the dataset represents a simplified and partially constructed abstraction of real-world supply chains. While such datasets are widely used for methodological transparency and replicability, they may not fully capture the institutional complexity, strategic decision-making processes, and contextual constraints present in actual industrial environments.

Therefore, the findings of this study should be interpreted as indicative of general operational patterns rather than as directly generalisable to all real-world supply chain settings. This limitation is explicitly considered when interpreting the empirical results.

Prior to analysis, the dataset was cleaned and preprocessed following best practices in empirical research (Hair et al. 2019). Missing numerical values were imputed using median values, while missing categorical values were imputed using modal values to reduce the influence of skewed distributions and extreme observations (Field 2018).

Potential outliers in lead-time and cost-related variables were identified using the interquartile range (IQR) method and were retained only when they reflected plausible operational conditions, consistent with recommendations for maintaining external validity in operational datasets (Hair et al. 2019). To ensure comparability across variables, cost-related measures and order quantities were normalised using min–max scaling prior to inclusion in regression models (Chavez et al. 2017).

No missing values were observed in the dataset. Nevertheless, data preprocessing procedures were carefully designed to address potential missingness. Median imputation was selected as a robust approach for skewed distributions, and sensitivity analyses using alternative methods (e.g. mean substitution and complete-case analysis) were considered to ensure consistency of results. These procedures support the robustness and reliability of the empirical analysis.

Records exhibiting logical inconsistencies – such as negative lead times or zero demand combined with non-zero shipment costs – were removed from the sample. Following these preprocessing steps, the final dataset provided a balanced representation of transactions across product categories and suppliers, supporting the robustness and interpretability of the empirical analysis.

Consistent with the integrated theoretical framework, study variables were organised into three complementary dimensions commonly used in strategic operations research (Barney 1991; Teece 2007; Vanpoucke et al. 2014). Resource-based indicators – manufacturing costs, transportation costs, inventory levels, and supplier ratings – reflect resource endowments related to cost efficiency and reliability, emphasised by the RBV (Barney 1991; Huo et al. 2014).

Dynamic capability indicators capture firms’ responsiveness and adaptability to uncertainty, operationalised primarily through lead-time performance, which reflects the ability to sense demand changes and reconfigure logistics processes (Aslam et al. 2018; Teece 2007). Supply chain integration indicators reflect coordination mechanisms across the supply chain, including supplier reliability and delivery-related consistency measures, consistent with SCI theory (Flynn et al. 2010; Schoenherr & Swink 2012).

Supply chain performance is modelled as a composite dependent variable to reflect its multidimensional nature. Following established performance measurement frameworks, the performance index is constructed by normalising and averaging delivery timeliness (the inverse of lead time), cost efficiency (the inverse of total operational cost), and fulfilment accuracy (Gunasekaran et al. 2008; Maestrini et al. 2018). Fulfilment accuracy is operationalised using delivery-status consistency and order-completion indicators available in the dataset.

However, it is important to acknowledge that the dataset represents a simplified and partially constructed abstraction of real-world supply chains. While such datasets are widely used for methodological transparency and replicability, they may not fully capture the institutional complexity, strategic decision-making processes, and contextual constraints present in actual industrial environments.

Therefore, the findings of this study should be interpreted as indicative of general operational patterns rather than as directly generalisable to all real-world supply chain settings. This limitation is explicitly considered when interpreting the empirical results.

To ensure a direct correspondence between the theoretical framework and the empirical model, each theoretical perspective is explicitly operationalised in the model specification. Consistent with the RBV, manufacturing cost, transportation cost, inventory level, and supplier rating are treated as observable proxies of firm-level operational resources.

Dynamic capabilities are operationalised through lead-time performance, which reflects the firm’s ability to adapt and reconfigure operational processes under uncertainty. Supply chain integration is captured through supplier rating and its interaction with lead time, representing relational coordination mechanisms that condition the effectiveness of responsiveness.

In this study, inventory levels are interpreted as proxies for resource endowments (RBV), delivery time reflects adaptive responsiveness (dynamic capabilities), and supplier performance represents relational coordination (SCI), thereby providing a structured – although simplified – operationalisation of the theoretical framework

Accordingly, the regression model is interpreted as providing statistically grounded evidence on the associations among key operational variables, while the integrated theoretical framework serves as a lens for interpreting these relationships rather than being directly tested within the empirical model.

A multistage analytical procedure was employed. Firstly, descriptive statistics were computed to summarise the central tendency and variability of key variables (Field 2018). Secondly, Pearson correlation coefficients were calculated to examine linear associations among operational indicators and to inform regression model specification (Hair et al. 2019).

Finally, a multivariate linear regression model was estimated to assess the joint effects of key operational indicators on supply chain performance (Equation 1): Performancei=β0+β1 (Supplier Ratingi)+β2 (Lead Timei)       +β3 (Manufacturing Costi)+       β4 (Transportation Costi)+β5 (Inventory       Leveli)+εi\text{Performance}_i=\beta_0       +\beta_1 (\text{Supplier Rating}_i)+\       beta_2(\text{Lead Time}_i)+\beta_3(\       text{Manufacturing Cost}_i)+\beta_4(\       text{Transportation Cost}_i)+\beta_5(\       text{Inventory Level}_i)+\varepsilon_       iPerformancei=β0+β1 (Supplier Ratingi)       +β2 (Lead Timei)+β3 (Manufacturing       Costi)+β4 (Transportation Costi)+β5       (Inventory Leveli)+εi.[Eqn 1]

To examine whether transport responsiveness operates as a conditional, rather than an independent, performance mechanism, additional interaction models were estimated.

Specifically, interaction terms between lead time and supplier rating and between lead time and manufacturing cost were introduced to test the capability logic proposed in the theoretical framework. This approach allows assessment of whether the performance impact of transport responsiveness depends on the presence of supporting relational and cost-based capabilities, consistent with the integrated RBV–Dynamic Capabilities–Supply Chain Integration perspective (Equation 2): Performance_i=β0+β1 (Lead Time_i)+β2(Supplier Rating_i)+         β3(Manufacturing Cost_i)+β4(Lead Time×         Supplier Rating)_i+β5(Lead Time×         Manufacturing Cost)_i+ε_i[Eqn 2]

All continuous variables included in the interaction terms were mean-centred prior to estimation to reduce potential multicollinearity concerns.

Given the cross-sectional nature of the dataset, the estimated coefficients are interpreted as associational rather than causal relationships (Wooldridge 2016). The objective is to identify statistically robust performance drivers that align with mechanisms proposed in strategic management and supply chain theory.

Standard diagnostic tests were conducted to assess regression assumptions, including checks for residual normality, homoscedasticity, and multicollinearity (assessed using variance inflation factors, VIFs), following established econometric guidelines (Hair et al. 2019).

Robustness checks included alternative model specifications estimated via stepwise regression and the least absolute shrinkage and selection operator (LASSO), as well as sensitivity analyses that employed alternative weighting schemes for the composite performance index (Hair et al. 2019; Maestrini et al. 2018). These procedures confirm the stability of the signs and magnitudes of coefficients across specifications.

The study relies exclusively on publicly available, non-proprietary data and does not involve human participants or personal information. Consequently, formal ethical approval was not required. Principles of data transparency, replicability, and responsible reporting were observed throughout the research process.

This article followed all ethical standards for research without direct contact with human or animal subjects. This study was based exclusively on secondary data obtained from an open-access repository. The dataset did not include human participants, animal subjects, personal identifiers, or confidential information. As the research did not involve human or animal subjects and posed no ethical risk, formal ethical approval from an institutional review board or ethics committee was not required. All data were analysed and reported in accordance with principles of transparency, responsible data use, and research integrity.

Table 1 reports descriptive statistics for the core operational variables used in the empirical analysis. The dataset comprises transactions spanning multiple product categories and suppliers, exhibiting substantial dispersion in both cost-related and time-based measures.

The descriptive statistics indicate substantial variability across observations, particularly for lead-time and cost-related variables, while delivery-related variables exhibit relatively limited variability. These summary statistics provide an overview of the distributional properties of the data and serve as a basis for the subsequent correlation and regression analyses.

Pearson correlation coefficients were calculated to examine linear associations among the operational indicators included in the analysis. Table 2 reports the correlation matrix for the key variables.

The correlation matrix indicates statistically significant associations between the performance index and several operational indicators, including lead time, supplier rating, and manufacturing cost. In addition, manufacturing and transportation costs exhibit a relatively strong positive correlation. Inventory levels exhibit weak correlations with most variables.

A multivariate linear regression model was estimated to quantify the associations between key operational variables and supply chain performance. Table 3 shows the estimated coefficients, standard errors, and model fit statistics.

A series of diagnostic and robustness checks were conducted to assess the reliability of the regression results. Multicollinearity was examined using VIF, which ranged from 1.12 to 2.87, and was well below the conventional threshold of 5, indicating no multicollinearity concerns.

Residual diagnostics further support the adequacy of the model specification. Visual inspection of residual plots reveals no evidence of heteroscedasticity, and the Shapiro–Wilk test suggests approximate normality of the residuals (W = 0.97, p = 0.06).

To assess the robustness of the estimated relationships, alternative model specifications were also estimated. Least absolute shrinkage and selection operator regression retained the same set of statistically significant predictors – namely lead time, supplier rating, and manufacturing cost – while stepwise regression yielded an identical model specification. Collectively, these robustness and diagnostic checks indicate that the reported results are stable and not sensitive to alternative estimation approaches or modelling assumptions.

Table 4 reports the results of the interaction models examining whether the performance impact of transport responsiveness is contingent on supporting supply chain capabilities. The interaction between lead time and supplier rating is negative and statistically significant, indicating that reductions in lead time yield stronger performance improvements when supplier reliability is high. By contrast, when supplier performance is weak, the marginal performance gains from lead-time reductions are substantially attenuated.

Similarly, the interaction between lead time and manufacturing cost is significant, suggesting that transport responsiveness is associated with superior performance only when supported by efficient internal cost structures. The interaction terms are not statistically significant, suggesting that the performance impact of lead time does not systematically depend on supplier rating or manufacturing cost within this dataset. This result does not support the hypothesised second-order capability effect and indicates that responsiveness may operate more independently than theoretically expected in this empirical context.

The negative association between lead time and supply chain performance and the positive relationship between supplier rating and supply chain performance are consistent with the RBV, which posits that strategic performance arises from the effective deployment of valuable and difficult-to-imitate operational capabilities (Barney 1991). These relationships are evident in both the correlation analysis (Table 2) and the multivariate regression results (Table 3), where lead time and supplier rating emerge as statistically significant predictors of performance.

From an RBV perspective, lead-time efficiency reflects underlying process maturity, coordination routines, supplier integration, and effective use of informational resources. Prior research similarly conceptualises delivery reliability and lead-time reduction as strategically embedded capabilities that generate performance advantages through superior coordination and responsiveness (Peng & Lu 2017; Rungtusanatham et al. 2003). Because these capabilities are embedded in firm-specific routines and interorganisational processes, they constitute an important source of performance heterogeneity (Chae et al. 2014; Wu et al. 2006).

In terms of magnitude, lead time exhibits the largest standardised effect among all predictors, indicating that a one-unit reduction in lead time is associated with a substantially larger change in performance compared to cost-related variables. This suggests that time-based competition may be more critical than cost efficiency in shaping performance outcomes in the analysed dataset.

Supplier rating captures the quality and reliability of upstream partnerships and functions as a relational resource that enhances operational stability and responsiveness. The positive association between supplier rating and performance reported in Table 3 is consistent with research emphasising that relational assets embedded in governance mechanisms and trust-based integration structures can generate sustained competitive advantage (Hitt et al. 2016; Vanpoucke et al. 2014; Zhang et al. 2018). Although the dataset does not directly capture strategic intent, the operational indicators analysed here can be interpreted as proxies for the underlying resource configurations emphasised by RBV (Yu et al. 2018).

The statistically significant effect of manufacturing cost further aligns with prior research highlighting cost efficiency as a foundational strategic capability in supply chains (Christopher 2016; Slack & Brandon-Jones 2022). Firms characterised by more efficient internal cost structures appear better positioned to manage competitive pressures and sustain performance under volatile conditions (Ivanov & Dolgui 2020; Singh & Modgil 2025).

It is important to observe, however, that the applicability of RBV in inter-organisational contexts such as supply chains has been subject to debate in the literature, as value creation often extends beyond firm boundaries and depends on relational and network-level capabilities. This study partially addresses this limitation by incorporating SCI as a complementary perspective.

The findings underscore the importance of dynamic responsiveness, as articulated in Dynamic Capabilities Theory (Teece 2007; Teece et al. 2016). The strong negative association between lead time and performance, observed in Table 2 and Table 3, suggests that agility in responding to demand and supply fluctuations is a critical driver of strategic outcomes (Aslam et al. 2018).

From a dynamic capabilities perspective, lead time captures firms’ ability to sense changes in operational conditions, seize emerging opportunities, and reconfigure logistics and production processes accordingly. Longer lead times can, therefore, be interpreted as indicators of weaker sensing and seizing mechanisms, whereas shorter lead times indicate more effective adaptive routines (Blome et al. 2013; Vanpoucke et al. 2014).

These findings are consistent with prior research demonstrating that dynamic capabilities – manifested through demand sensing, operational flexibility, and supplier reconfiguration – enhance resilience and performance in uncertain and volatile environments (Ivanov & Dolgui 2020; Raj et al. 2025; Wetsandornphong et al. 2025). By empirically linking lead time to performance, this study reinforces the role of responsiveness as a key adaptive mechanism in contemporary supply chain systems.

It is important to acknowledge that the empirical model employed in this study is static and cross-sectional, which limits its ability to fully capture the dynamic and temporal nature of capability development. While lead time is interpreted as a proxy for responsiveness, it does not directly capture sensing, seizing, and reconfiguration processes over time.

Therefore, the representation of dynamic capabilities in this study should be understood as a simplified operational approximation rather than a full empirical test of dynamic capability theory. Future research using longitudinal data would be better suited to capture the evolution of such capabilities.

The statistically significant association between supplier rating and supply chain performance is broadly consistent with SCI theory. As shown in Table 3, higher supplier ratings are associated with superior performance outcomes, suggesting that upstream integration may play an important role in shaping supply chain effectiveness (Flynn et al. 2010; Huo 2012).

Supplier integration enhances delivery reliability and coordination by reducing information asymmetries and operational friction. Consistent with this interpretation, the correlation analysis (Table 2) indicates that supplier rating is negatively associated with lead time and cost-related indicators. This pattern aligns with prior research demonstrating that relational integration improves performance through enhanced coordination and information alignment (Danese & Romano 2011; Schoenherr & Swink 2012; Zhang et al. 2018).

Interestingly, inventory level does not exhibit a statistically significant relationship with performance. This finding may reflect measurement limitations or context-specific dynamics within the dataset. Alternatively, it may indicate a structural shift away from inventory-based buffering towards more responsive and information-driven supply chain strategies.

By contrast, inventory level does not exhibit a statistically significant relationship with performance. This finding suggests that traditional buffer-based mechanisms may be less influential than relational- and capability-based integration mechanisms in the empirical context examined. Rather than relying on inventory buffers, performance advantages appear to be associated with coordinated relationships and information-rich interactions among supply chain partners (Ketchen & Hult 2007; Vanpoucke et al. 2014).

Taken together, the findings suggest that lead-time reduction should not be interpreted as a universally effective best practice. While responsiveness is associated with improved performance, the empirical results do not provide evidence of systematic interaction effects with supplier relationships or cost structures. This indicates that the role of responsiveness may be more independent and context-specific than theoretically expected. Accordingly, the findings do not fully support the assumption that agility compensates for weak relational or resource foundations.

From a multitheoretical perspective, supply chain performance can be interpreted as being shaped by interdependent strategic mechanisms rather than isolated operational drivers. Resource-based view emphasises the role of efficient cost structures and relational resources; Dynamic Capabilities Theory explains how responsiveness sustains performance under uncertainty; and SCI highlights how coordination amplifies the effectiveness of both resources and adaptive routines. Integration thus operates as a performance multiplier that conditions the returns to both operational resources and dynamic responsiveness.

The findings are broadly consistent with established research in supply chain and strategic management. The strong association between lead time and performance reinforces prior evidence identifying temporal efficiency as a critical determinant of service quality and responsiveness (Gunasekaran et al. 2008; Peng & Lu 2017; Slack & Brandon-Jones 2022).

Similarly, the positive relationship between supplier rating and performance aligns with extensive research emphasising supplier quality and integration as key relational capabilities (Danese & Romano 2011; Flynn et al. 2010; Schoenherr & Swink 2012). The statistically significant effect of manufacturing costs further consistent with research highlighting cost efficiency as a foundational strategic capability (Barney 1991; Hitt et al. 2016).

Finally, the limited role of inventory levels contrasts with traditional buffer-based models but aligns with recent studies suggesting that responsiveness, digital visibility, and relational coordination increasingly substitute for inventory buffers in modern supply chains (Ivanov & Dolgui 2020; Kamble & Gunasekaran 2020).

Building on these findings and their alignment with prior literature, the following section outlines the key contributions of this study to transport and supply chain management research and practice.

Firstly, this study contributes to the transport and supply chain management literature by empirically demonstrating that transport responsiveness, captured through lead-time performance, plays a central role in shaping overall supply chain performance across interconnected value chain activities. These findings contribute to the literature by demonstrating that operational responsiveness (delivery time) functions as a dominant capability mechanism, while resource-based elements and relational coordination play supporting but context-dependent roles. This nuanced differentiation extends existing theory by highlighting the asymmetric impact of capability dimensions in supply chain performance.

Secondly, the study advances existing research by showing that transport-related performance outcomes are associated with responsiveness, supplier coordination, and cost efficiency, from rather than isolated operational factors.

Thirdly, by integrating transport-relevant operational indicators with strategic supply chain theories, the study bridges transport operations research and strategic supply chain management, offering a unified framework for analysing value creation across logistics and supply chain processes.

Fourthly, the findings provide practical insights for logistics managers and policy-oriented stakeholders by highlighting that investments in transport responsiveness and supplier integration may yield greater performance improvements than traditional inventory-based buffering approaches.

Finally, the use of a transparent and publicly available operational dataset enhances the reproducibility of transport and supply chain performance research and supports evidence-based decision-making in transport planning and supply chain design.

This study contributes to the strategic management and supply chain literature in several ways.

Firstly, it brings together the RBV, Dynamic Capabilities Theory, and SCI Theory within a unified analytical framework. Rather than providing a full empirical test of these perspectives, the study offers a theoretically informed empirical examination of how key operational indicators can be interpreted in the light of these complementary frameworks.

The findings of this study suggest that supply chain performance is associated with the alignment of internal resource endowments, responsiveness-related capabilities, and coordinated interorganisational relationships, although these relationships should be interpreted as context-dependent rather than universally generalisable.

Secondly, the results provide quantitative support for capability-based explanations of performance by linking abstract strategic constructs – such as responsiveness and relational coordination – to observable operational indicators, including lead time and supplier rating. This strengthens the empirical connection between strategic management theory and operations-focused research.

Thirdly, the use of a composite performance index integrating cost efficiency, delivery timeliness, and fulfilment accuracy advances holistic performance measurement in supply chain research and supports systems-oriented perspectives that move beyond single-dimensional metrics.

From a managerial perspective, the results suggest that superior supply chain performance is more likely to result from coordinated investments across multiple operational domains than from isolated interventions. Managers should prioritise lead-time reduction through process optimisation, digital visibility, and logistics coordination, while strengthening supplier evaluation and development practices. Although responsiveness is critical, manufacturing cost efficiency remains a foundational capability that supports competitiveness, underscoring the continued relevance of operational excellence initiatives such as lean practices and advanced production planning.

Several limitations should be acknowledged. The dataset represents a simplified and partially synthetic abstraction of real-world supply chains and may not fully capture firm-specific strategic intent or industry-specific dynamics. In addition, the cross-sectional design restricts the analysis to associations and does not permit examination of how capabilities evolve over time. Potential omitted variables – such as demand uncertainty, customer collaboration, and environmental volatility – may also influence performance, but are not observable in the available data.

Future research could address these limitations by employing longitudinal datasets to examine dynamic capability development, conducting industry-specific comparative analyses, and incorporating digital transformation indicators such as Internet of Things (IoT), Artificial Intelligence (AI)-driven forecasting, and real-time visibility systems. Advanced analytical approaches, including structural equation modelling, machine learning, and network-based analyses, could further capture nonlinear and systemic performance mechanisms.

Overall, this study provides a data-driven and theoretically grounded account of how strategic and operational capabilities jointly shape supply chain performance. By integrating multiple strategic perspectives and operational analytics, it offers scholarly and managerial insights for designing more resilient, adaptive, and strategically aligned supply chains. However, the findings should be interpreted with caution, as the empirical model captures selected dimensions of the integrated theoretical framework and does not fully reflect the dynamic and interactional mechanisms implied by dynamic capabilities and integration theories. This limitation highlights the need for more advanced empirical designs in future research.