AI-Driven Digital Twin Framework for Warehouse Logistics Automation Optimization

Md Mahinur Alam; Kanita Jerin Tanha; Md Tanjin Alam; Shadman Sakib; Min Gyu Kim

cs.AIOct 28, 2026

AI-Driven Digital Twin Framework for Warehouse Logistics Automation Optimization

Authors: Md Mahinur Alam, Kanita Jerin Tanha, Md Tanjin Alam, Shadman Sakib, Min Gyu Kim

Organizations: Georgia Southern University, GA, USA · Kumoh National Institute of Technology, Gumi, South Korea · VosuAI · University of Maryland, Baltimore County, MD, USA

Abstract

In recent years, warehouse logistics optimization has increasingly focused on intelligent, data-driven systems to enhance operational efficiency. Conventional static product placement strategies are being progressively replaced by dynamic slotting approaches that can adapt to real-time variations in warehouse conditions. In this work, we present a novel framework that employs a reinforcement learning agent integrated within a FlexSim-based Digital Twin environment for real-time product placement optimization. The proposed agent continuously learns from streaming operational data, enabling it to reduce picking time while improving overall system throughput. The simulation environment models a four-zone warehouse layout with forklift-based material handling and leverages the Asset Administration Shell framework to ensure seamless interaction between the learning agent, physical asset, and the digital twin. Experimental evaluation demonstrates substantial performance gains, with the proposed approach achieving a 29.66% reduction in average picking time, a 35.42% increase in throughput, and an 18.54% decrease in travel distance.

Figures & tables

Fig. 1 : Proposed AI-Driven Digital Twin Framework for Smart Warehouse Logistics, Physical-to-Digital Synchronization via Asset Administration Shell with Reinforcement Learning Based Dynamic Reslotting Optimization.

Fig. 2 : Digital Twin Warehouse Logistics Simulation in FlexSim Four Zone Layout with Forklift-Based Product Handling.

Parameters	Values	Parameters	Values
Inbound Queues Max Content	100	Racks	16
Outbound Queues Max Content	100	Zones	4
Inbound Conveyor Speed	1.75 m\s	Speed Picker Transporter	1.25 m\s
Outbound Conveyor Speed	1.25 m\s	Acceleration Picker Transporter	1.75 m\s
Operators	2	Speed Placer Transporter	1.0 m\s
Sinks	1	Acceleration Placer Transporter	1.25 m\s

TABLE I : FlexSim Digital Twin Warehouse Simulation Asset Configuration Parameters.

Fig. 3 : Ablation study comparing average picking time across Q-network architecture variants.

Fig. 4 : FlexSim dashboard comparison between the proposed framework DQN-Agent and Static Slotting baseline across per-transporter travel distance, utilization state distribution, and warehouse throughput.

Metrics	Static Slotting	DQN-Agent	Changes (%)
Avg. Travel Distance (units)	0.8203 $\pm$ 0.012	0.6682 $\pm$ 0.018	$\downarrow$ 18.54%
Avg. Picking Time (s)	169.22 $\pm$ 2.41	118.32 $\pm$ 3.15	$\downarrow$ 29.66%
Throughput (orders/hour)	23.35 $\pm$ 0.87	31.62 $\pm$ 1.04	$\uparrow$ 35.42%

TABLE II : Performance comparison averaged over 5 independent runs with different random seeds.

Warehouse Automation

See all 15 papers

Offline Reinforcement Learning for Warehouse SLAM Throughput Control
Jun 22, 2026Tina Dongxu Li, Mouhacine Benosman, Rajat Kumar +3RL Control Offline RL
Amazon.com, USA
Learning-guided Prioritized Planning for Lifelong Multi-Agent Path Finding in Warehouse Automation
Mar 25, 2026Han Zheng, Yining Ma, Brandon Araki +2Warehouse Automation Multi-Agent Path Finding
Massachusetts Institute of Technology, Cambridge, MA, USA · Symbotic, Wilmington, MA, USA
Learning to Staff: Offline Reinforcement Learning and Fine-Tuned LLMs for Warehouse Staffing Optimization
Mar 25, 2026Kalle Kujanpää, Yuying Zhu, Kristina Klinkner +1LLM Decision-Making Supply Chain Management
Amazon, Fulfillment Technologies & Robotics

Jun 22, 2026J/K move · Enter open · S save

Offline Reinforcement Learning for Warehouse SLAM Throughput Control

Tina Dongxu Li, Mouhacine Benosman, Rajat Kumar, Kevin Tan+2

Amazon.com, USA

RL Control Offline RL

Open paper AskPDF arXiv

We present an offline reinforcement learning (RL) framework for optimizing SLAM throughput control in a warehouse fulfillment environment. SLAM (Scan/Label/Apply/Manifest) throughput directly influences system congestion and operational efficiency. Our RL-based control approach dynamically recommends SLAM throughput settings that adaptively balance throughput maximization with downstream stability through intelligent adjustment of throttling behavior. We include a history-informed state representation, action space abstraction for delayed-impact control, and a reward function that captures both upstream and downstream operational metrics. Our approach is algorithm-agnostic, enabling integration of multiple offline RL methods under a unified architecture. We instantiate our framework with three state-of-the-art offline RL algorithms, and trained the models offline using de-identified historical operational logs from a large-scale warehouse. Policy performance is evaluated using a comprehensive multi-method strategy. These include model-free approaches including immediate reward estimation via regression models and long-horizon Fitted Q Evaluation (FQE), as well as model-based Deep Koopman dynamics evaluation. Empirical results reveal that the CQL policy consistently outperforms alternatives, improving system health by 22.97% and reducing average throttling duration by 3.18%. These findings demonstrate the potential of offline RL for safe and scalable warehouse throughput control optimization.

Abstract

Warehouse Automation

Explore similar work

Offline Reinforcement Learning for Warehouse SLAM Throughput Control

Learning-guided Prioritized Planning for Lifelong Multi-Agent Path Finding in Warehouse Automation

Learning to Staff: Offline Reinforcement Learning and Fine-Tuned LLMs for Warehouse Staffing Optimization