Cold Chain Digital Tools Link to Sustainability Through Worker Behavior

A cross-sectional investigation of Chinese agri-food cold-chain enterprises has revealed that organizational digital maturity is positively linked to green human resource analytics, which in turn correlates with frontline employee loss-prevention actions and superior supervisor-rated sustainable logistics performance. Published in Frontiers in Sustainable Food Systems, the empirical study analyzed matched survey evaluations from 96 enterprise supervisors and 545 subordinate employees across diverse cold-chain logistics operations in China. The researchers sought to examine how overarching digital capabilities convert into frontline ecological actions under the conceptual umbrella of the Twin Transition perspective, which posits that digital transformation and environmental sustainability represent interdependent structural trajectories rather than isolated operational objectives.

Agri-food supply chains experience substantial pressure to maintain product safety and integrity while curtailing energy consumption and carbon emissions, particularly within refrigerated logistics networks where temperature deviations directly produce food degradation. Within modern cold storage facilities, transit corridors, and distribution hubs, continuous temperature surveillance platforms, vehicle tracking frameworks, and warehouse inventory databases capture voluminous operational records. However, technical systems alone cannot prevent product spoilage without reliable human operational intervention. The investigative team focused on resolving the operational mechanism connecting enterprise data collection with the everyday routines of frontline workers who handle, pack, transport, and monitor perishable commodities across complex distribution networks.

To capture this organizational transmission mechanism, the investigators conceptualized green human resource analytics as an applied organizational capability that merges workforce behavioral metrics with environmental tracking indicators. Rather than operating as an isolated computational routine or representing an entire suite of green human resource management policies, green human resource analytics evaluates operational performance gaps across distinct job functions and supplies direct guidance for employee training, routine feedback, performance appraisals, and labor schedules. In this operational model, digital maturity provides the prerequisite infrastructure and informational architecture necessary to track logistics exceptions and evaluate operational deviations, thereby enabling supervisors to direct targeted feedback toward routine cold-chain preservation practices.

The study specifically measured frontline activity through employee green loss-prevention behavior, operationalized as a combined behavioral construct encompassing both mandatory protocol compliance and voluntary, discretionary initiative. Compliant actions include strict adherence to refrigeration standards, regulated freight-handling protocols, and prompt mechanical exception notifications. Discretionary actions involve unprompted peer interventions, peer coaching, and proactive operational checks performed in the absence of direct supervisory monitoring. At the organizational outcome level, participating supervisors evaluated sustainable logistics performance across multiple operational dimensions, including food-loss mitigation, energy conservation, vehicular emissions reductions, cold-chain consistency, and overall physical resource efficiency across participating corporate departments.

Data collection was conducted in September 2025 and March 2026 across several Chinese geographic territories, including the Yangtze River Delta, Pearl River Delta, Beijing-Tianjin-Hebei corridor, Chengdu-Chongqing area, and central regional agricultural hubs. The sample encompassed third-party cold-chain logistics firms, fresh food e-commerce delivery services, internal distribution departments of food processors, wholesale market storage facilities, and specialized distributors handling meat, dairy, aquatic produce, and vegetables. To qualify for participation, respondents were required to be at least 18 years of age, possess a minimum of six months of specialized logistics experience, work within an organization utilizing at least one digital management platform, and demonstrate foundational awareness of food-loss mitigation and green operational standards.

To mitigate single-source survey bias, the researchers utilized a matched multi-respondent methodology linking each supervisor with between three and eight subordinate personnel within the same corporate facility. Supervisors provided ratings evaluating organizational digital maturity, green human resource analytics, and sustainable logistics performance, while frontline employees provided self-evaluations detailing their personal green loss-prevention behavior and perceived algorithmic pressure. Among the 545 surveyed employees, 58.35 percent were male and 41.65 percent were female, with over 69 percent aged between 26 and 45 years. Participating job classifications included frontline equipment operators, dispatchers, transit coordinators, warehouse personnel, temperature technicians, food quality monitors, and operational line managers.

Primary structural equation modeling of the pooled dataset confirmed that digital maturity maintained a statistically significant positive association with green human resource analytics, yielding a standardized path coefficient of 0.57. In turn, green human resource analytics demonstrated a significant positive association with employee green loss-prevention behavior, yielding a standardized coefficient of 0.46. Reported employee loss-prevention behavior was also positively linked with supervisor-rated sustainable logistics performance, exhibiting a path coefficient of 0.53. In contrast, the direct path linking green human resource analytics directly with sustainable logistics performance was statistically nonsignificant, showing a standardized estimate of 0.08, indicating that the operational influence of workforce analytics is linked through frontline employee behavioral enactment.

Bootstrap resamples evaluating indirect pathways confirmed statistically significant serial relationships across the operational framework. The indirect pathway connecting green human resource analytics to sustainable logistics performance via employee loss-prevention behavior produced an estimate of 0.244, while the full serial pathway spanning digital maturity, green human resource analytics, employee behavior, and final logistics performance yielded an estimate of 0.139. Supplementary unit-based cluster bootstrapping across 5,000 resamples of the 96 organizational units reinforced these indirect paths, generating 95 percent bias-corrected confidence intervals of 0.152 to 0.345 and 0.078 to 0.218, respectively. These sensitivity analyses demonstrated that the observed indirect relationships remained robust when explicitly accounting for cluster-level sampling variations.

Critically, the research team identified perceived algorithmic pressure as an influential boundary condition that moderated the relationship between analytics and worker behavior. Defined as subjective strain stemming from pervasive digital monitoring, automated performance metrics, and rapid algorithmic appraisals, algorithmic pressure exhibited a significant negative interaction with green human resource analytics, producing an interaction coefficient of negative 0.15. Interaction modeling demonstrated that while the relationship between analytics and loss prevention remained positive under high perceived algorithmic pressure, the slope was significantly attenuated compared to conditions of low perceived pressure.

The authors underscored several structural and methodological boundaries governing the interpretation of their empirical findings. Because the investigation relied on a cross-sectional observational design, the documented relationships represent correlational patterns rather than proven causal mechanisms or longitudinal transformations. Furthermore, because supervisor assessments were recorded once per enterprise unit and matched to multiple employees, the primary pooled structural equations reflect shared organizational information rather than isolated cross-level latent dynamics. Future investigations will need to track longitudinal operational metrics, separate compliance from voluntary initiatives, and incorporate objective sensor records of energy expenditure and food spoilage to further clarify the operational role of workforce analytics in sustainable logistics.

To rigorously evaluate the structural relationships within the survey data, the authors conducted confirmatory factor analyses and statistical tests for common method variance across the pooled responses. The hypothesized five-factor model demonstrated an acceptable overall fit to the data, outperforming collapsed alternatives such as a four-factor model that combined green human resource analytics with frontline employee behavior. Furthermore, diagnostic checks indicated that common method bias was unlikely to account for the primary results. An unrotated exploratory factor analysis revealed that the first factor accounted for 27.86 percent of the total variance, remaining comfortably below the common 40 percent diagnostic benchmark, while incorporating a common latent factor into the measurement model shifted structural path coefficients by less than 0.05.

The study also documented key descriptive baselines and correlational patterns across the participating cold-chain logistics operations. On five-point rating scales, mean scores averaged 3.72 for digital maturity, 3.55 for green human resource analytics, 3.78 for employee green loss-prevention behavior, and 3.67 for sustainable logistics performance. Notably, algorithmic pressure averaged 3.23 and exhibited distinct correlational directions: it correlated positively with digital maturity at 0.19 and green human resource analytics at 0.22, yet correlated negatively with employee loss-prevention behavior at negative 0.18 and supervisor-rated sustainable logistics performance at negative 0.14. This contrasting configuration demonstrates that while more digitally mature facilities deploy more extensive tracking, heightened perceived surveillance pressure does not naturally foster positive frontline engagement.

Complementary sensitivity analyses verified that the primary structural equation modeling results were not artifacts of ignoring nested data structures across the 96 organizational clusters. Because individual employee responses were grouped within specific operating facilities, intra-class correlation coefficients of 0.14 for loss-prevention behavior and 0.11 for algorithmic pressure indicated meaningful within-unit clustering. When the researchers re-estimated the structural relationships using composite-score mixed models that explicitly grouped responses by enterprise unit, the positive associations between digital maturity, analytics, employee behavior, and sustainable logistics performance remained statistically significant at the 0.001 level. Similarly, the negative moderating effect of algorithmic pressure remained significant at p equals 0.001, confirming that the structural conclusions held under alternative estimation techniques.

These analytical findings provide practical guidance for cold-chain managers seeking to align digital surveillance with environmental goals. The researchers emphasized that operational data—such as temperature-control deviations, inventory degradation alerts, and energy consumption spikes—must be deliberately translated into supportive employee feedback, targeted training, and fair performance appraisals.

Subject of Research: Digital maturity, workforce analytics, and sustainable logistics in agri-food cold chains

Article Title: Digital maturity, green HR analytics, and sustainable logistics performance in agri-food supply chains

Article References: He, X., Zhang, X., & Wang, H. (2026). Digital maturity, green HR analytics, and sustainable logistics performance in agri-food supply chains. Frontiers in Sustainable Food Systems, 10, Article 1925542. https://doi.org/10.3389/fsufs.2026.1925542

Image Credits: AI Generated

DOI: 10.3389/fsufs.2026.1925542

Keywords: agri-food supply chains, cold chain logistics, digital maturity, green human resource analytics, loss-prevention behavior, sustainable logistics, Twin Transition, algorithmic pressure, food loss mitigation, refrigerated transport, workplace monitoring, structural equation modeling, cluster bootstrapping, supply chain management, organizational behavior, energy efficiency, environmental performance