Dynamic Multi-Echelon Order Fulfillment Optimization in Omnichannel Logistics Empowered by Real-Time Edge-Inferred Spatial-Temporal Signals

• Asher Noah

  covenant University

  Author

The convergence of omnichannel retailing and edge computing has created unprecedented opportunities for dynamic order fulfillment optimization, yet existing multi-echelon inventory models remain largely static and fail to leverage real-time spatial-temporal signals for adaptive decision-making. This study addresses the critical gap between advanced demand forecasting capabilities and operational fulfillment execution by developing a framework that integrates edge-inferred spatial-temporal signals into multi-echelon order fulfillment decisions. Drawing on a systematic literature review of multi-echelon inventory optimization and empirical analysis of omnichannel fulfillment operations, this research designs and validates a Dynamic Edge-Informed Fulfillment Optimization framework. The methodology combines spatial-temporal signal processing through graph attention networks and temporal convolutional networks with reinforcement learning-based fulfillment decision optimization. Experimental validation using synthetic and real-world inspired retail datasets demonstrates that the proposed framework achieves 89.4% fulfillment accuracy while reducing average order cycle time by 22% compared to static baseline methods. The findings reveal that edge-inferred signals significantly improve inventory allocation decisions across echelons, particularly during demand disruptions. The study contributes a validated framework for integrating real-time spatial-temporal intelligence into omnichannel fulfillment operations and provides practical guidance for retailers seeking to enhance fulfillment responsiveness in competitive markets.

• pdf

Copyright (c) 2026 Asher Noah (Author)

This work is licensed under a Creative Commons Attribution 4.0 International License.