The explosive growth of e-commerce has magnified last-mile delivery (LMD) inefficiencies, generating disproportionate transportation costs, empty vehicle-miles and rising greenhouse-gas emissions. Confronting this dual logistical-environmental problem, this study proposes an integrated framework that combines Robotic Process Automation (RPA), artificial intelligence (AI) and social-media analytics to create adaptive, low-carbon LMD ecosystems. RPA orchestrates repetitive fulfillment workflows, while generative-AI interfaces negotiate real-time delivery slots and consolidate orders using customer sentiment mined from social platforms. Dynamic routing engines then translate these preferences into eco-efficient tours, maximising vehicle load factors and minimising total distance travelled. The chief contribution is a detailed, end-to-end architecture that fuses AI-enhanced e-commerce portals, automated social engagement and sustainability-centred decision rules. Obtained results indicate up to 30 % cuts in delivery emissions and 18 % savings in cost per drop, validating the synergy between operational performance and environmental stewardship. Concurrently, the study outlines governance mechanisms like privacy-preserving data pipelines, explicit consent layers and transparent opt-in green options, crucial for trustworthy deployment. By framing LMD as a multi-objective optimisation problem solvable through cognitive automation, the work charts a research agenda extending RPA toward self-learning, priority-balancing agents that adapt to volatile demand, regulatory pressure and carbon targets. The framework thus offers retailers a practical pathway to resilient, customer-centric and climate-aligned logistics.
Building similarity graph...
Analyzing shared references across papers
Loading...
Khass et al. (Thu,) studied this question.
www.synapsesocial.com/papers/69c0de74fddb9876e79c1472 — DOI: https://doi.org/10.1016/j.procs.2026.02.034
Omar Al Khass
Omid Fatahi Valilai
Procedia Computer Science
Constructor University
Building similarity graph...
Analyzing shared references across papers
Loading...