AI-Optimized Sustainable Last-Mile Delivery in E-Commerce: Challenges, Solutions, and Policy Recommendations for Azerbaijan
Main Article Content
Abstract
Last-mile delivery constitutes the most cost-intensive, environmentally burdensome, and operationally complex segment of e-commerce supply chains. In Azerbaijan, where e-commerce turnover has expanded by over 200 per cent since 2019 while logistics infrastructure maturity remains constrained, the sustainability challenges inherent in last-mile operations are intensifying. This paper investigates how artificial intelligence (AI) technologies can be leveraged to optimize last-mile delivery in ways that simultaneously improve operational efficiency, reduce environmental impact, and enhance service quality within Azerbaijan’s e-commerce ecosystem. Integrating the Technology–Organization–Environment (TOE) framework with the dynamic capability’s perspective, the study develops and empirically tests a moderated mediation model. Primary survey data from 285 logistics professionals are analyzed using Partial Least Squares Structural Equation Modelling (PLS-SEM). The results demonstrate that AI adoption significantly strengthens smart logistics capability (β = 0.437, p < 0.001), which in turn improves last-mile delivery performance (β = 0.523, p < 0.001). Capability mediates 64 per cent of AI’s total effect on performance. Data quality and organizational readiness significantly moderate the respective pathway stages, with the conditional indirect effect nearly quadrupling under favorable enabling conditions compared to unfavorable ones. The paper contributes context-specific evidence from an under-studied emerging market, demonstrates the empirical validity of capability-based AI value creation in sustainable last-mile logistics, and derives five actionable policy recommendations for Azerbaijan’s national digital logistics strategy. The findings underscore that technology procurement alone is insufficient; sustainable last-mile improvement requires coordinated investment in data governance, workforce analytics capabilities, and organisational transformation.