AutoMARS: Automatic and Scalable Multi-Agent Large Language Model Framework for Supply Chain Risk Assessment

Yinzhu Quan; Zefang Liu; Frederick Benaben; Benoit Montreuil

cs.AIOct 12, 2026

AutoMARS: Automatic and Scalable Multi-Agent Large Language Model Framework for Supply Chain Risk Assessment

Authors: Yinzhu Quan, Zefang Liu, Frederick Benaben, Benoit Montreuil

Organizations: Physical Internet Center, H. Milton Stewart School of Industrial and Systems Engineering, Georgia Institute of Technology, Atlanta, 30332, GA, USA · School of Computational Science and Engineering, Georgia Institute of Technology, Atlanta, 30332, GA, USA · Centre Genie Industriel, IMT Mines Albi, Albi, France

Abstract

We propose AutoMARS (Automatic Multi-Agent Risk Assessment), an agentic framework for automatic and scalable comparative risk assessment in supply chain networks. The framework enables end-to-end execution of analytical workflows by automatically planning tasks, coordinating specialized agents, invoking external analytical tools, and synthesizing domain-specific results into integrated risk evaluations. AutoMARS further supports scalable risk comparison across multiple locations by parallelizing independent analytical tasks. The framework adopts a multi-domain, multi-agent, and multi-tool architecture that integrates heterogeneous risk factors from socioeconomic, disaster, environmental, and transportation domains. Through coordinated agent interactions and tool integration, AutoMARS enables structured, systematic, and interpretable comparative analysis to support supply chain risk-aware location evaluation.

Figures & tables

Figure 1: Framework of AutoMARS for supply chain risk assessment.

Tool	Agent	Type	Description
Labor Exposure Tool 8 8 8 https://www.census.gov/programs-surveys/ces/data/restricted-use-data/lehd-data.html	Census	API	Extract labor composition indicators
Workforce Dynamics Tool 9 9 9 https://www.census.gov/data/developers/data-sets/qwi.html	Census	API	Capture employment flow and turnover indicators
Socioeconomic Profile Tool 10 10 10 https://www.census.gov/programs-surveys/acs.html	Census	API	Retrieve core socioeconomic indicators
Community Resilience Tool 11 11 11 https://www.census.gov/data/developers/data-sets/community-resilience-estimates.html	Census	API	Characterize population risk factor distributions
Industry Scale Tool 12 12 12 https://www.census.gov/data/developers/data-sets/economic-census.html	Census	API	Quantify industry-level economic scale indicators
Disaster Declaration Tool 13 13 13 https://www.fema.gov/openfema-data-page/disaster-declarations-summaries-v2	Disaster	API	Identify federal disaster declaration events

Table 1: Tools used in AutoMARS, including their associated agents, data source types, and functional descriptions.

Figure 2: Workflow of AutoMARS case study for pairwise risk assessment.

Tool	Paulding County	Cobb County
Socioeconomic Profile Tool (2019-2023)	Population: $\sim$ 174k; Labor Force: $\sim$ 90k; Median Household Income: $94.6k; Unemployment Rate: 4.46%	Population: $\sim$ 769k; Labor Force: $\sim$ 433k; Median Household Income: $98.7k; Unemployment Rate: 4.55%
Community Resilience Tool (2023)	Population with 3+ Risk Factors: 12.13% ( $\sim$ 22k)	Population with 3+ Risk Factors: 14.96% ( $\sim$ 115k)
Labor Exposure Tool (2023)	In-county Jobs: $\sim$ 29k; Commuting Pattern: Net outflow	In-county Jobs: $\sim$ 414k; Commuting Pattern: Net inflow

Table 2: Tools invoked by Census Agent and corresponding comparative socioeconomic indicators for Paulding and Cobb Counties.

Tool	Paulding County	Cobb County
Multi-Hazard Risk Index Tool	Composite Risk: 69.97; Expected Annual Loss: 79.52; Community Resilience: 44.12	Composite Risk: 94.37; Expected Annual Loss: 95.85; Community Resilience: 28.72
Disaster Declaration Tool (2020-Present)	6 declarations; includes Tropical Storm Zeta (2020), Hurricane Helene (2024), Winter Storm (2026)	4 declarations; includes Hurricane Helene (2024), Winter Storm (2026)
Severe Weather Tool (2023)	Events: 355; Damage: $\sim$ $37.4M; Deaths: 2; Injuries: 8	Events: 319; Damage: $\sim$ $37.3M; Deaths: 1; Injuries: 8
Climate Tool (2023)	Annual TMIN: $\sim$ 10.2–10.6 ∘ C; Precipitation: $\sim$ 1174–1179 mm	Annual TMIN: $\sim$ 11.9 ∘ C; Precipitation: $\sim$ 1080–1416 mm

Table 3: Tools invoked by Disaster Agent and corresponding comparative hazard and disaster risk indicators for Paulding and Cobb Counties.

Tool	Paulding County	Cobb County
Air Quality Tool (2023)	No county-coded AQS observations for PM2.5, Ozone, or CO	PM2.5: $\sim$ 10.82 $\mu g/m^{3}$ ; Ozone: $\sim$ 0.030 ppm
Environmental Facility Tool (2023)	$\sim$ 7 industrial plants; 1 landfill; $\sim$ 1 chemical facility	$\sim$ 30 industrial plants; $\sim$ 6 landfills; $\sim$ 35 chemical facilities; includes a major natural gas power plant

Table 4: Tools invoked by Environment Agent and corresponding comparative environmental exposure indicators for Paulding and Cobb Counties.

Indicator	Paulding County	Cobb County
Traffic Accident Tool (2023)	Total Crashes: 13; Severe/Fatal Crashes: 0 (0%); Mean Severity: 2.00	Total Crashes: 606; Severe/Fatal Crashes: 18 (3.0%); Mean Severity: 2.06

Table 5: Tools invoked by Traffic Agent and corresponding comparative transportation safety indicators for Paulding and Cobb Counties.

Figure 3: Execution timeline of AutoMARS for pairwise risk assessment.

County A	County B	More Risky	Percentage
Fulton	Cobb	Fulton	100%
Cobb	Fulton	Fulton	100%
Clayton	Gwinnett	Gwinnett	100%
Gwinnett	Clayton	Gwinnett	85%
DeKalb	Henry	DeKalb	100%
Henry	DeKalb	DeKalb	100%

Table 6: Pairwise comparison stability across repeated runs.

Panel of Figure 4 — Figure 4: Ranking stability across repeated runs at different spatial scales.

Panel of Figure 5 — Figure 5: Average and standard deviation of county rankings across repeated runs.

Panel of Figure 6 — Figure 6: Spatial distribution of mean county rankings across Atlanta regions.

Manager	Domain	Reflection	Clayton	Cobb	Gwinnett	DeKalb	Fulton
GPT-5	GPT-5	GPT-5	$1.4\pm 0.55$	$1.6\pm 0.55$	$3.4\pm 0.55$	$3.6\pm 0.55$	$5.0\pm 0.00$
GPT-5 mini	GPT-5 mini	GPT-5 mini	$1.8\pm 0.84$	$1.6\pm 0.89$	$2.6\pm 0.55$	$4.0\pm 0.00$	$5.0\pm 0.00$
GPT-5 nano	GPT-5 nano	GPT-5 nano	$1.0\pm 0.00$	$2.4\pm 0.55$	$2.8\pm 0.84$	$3.8\pm 0.45$	$5.0\pm 0.00$

Table 7: Performance comparison of end-to-end agent pipelines on Atlanta Urban Core with mean and standard deviation.

Setting	Total Runtime (s)
Sequential execution	$622.11\pm 59.84$
Parallel execution	$327.49\pm 64.70$

Table 8: Impact of parallel execution on overall runtime with mean and standard deviation.

Setting	Tool Invocation Time (s)
Air Quality Tool w/o caching	76.68
Air Quality Tool w caching	0.00
Environmental Facility Tool w/o caching	37.98
Environmental Facility Tool w caching	0.58

Table 9: Impact of caching mechanism on tool invocation time.

Explore similar work

A Hybrid Agentic AI Framework for Intelligent Supply Chain Analytics
Sep 11, 2026Xian Yeow Lee, Teppei Inoue, Haiyan Wang +1Data Analysis Agents Multi-Agent LLM Systems
Industrial AI Lab Hitachi America
LLM-Grounded Explainable AI for Supply Chain Risk Early Warning via Temporal Graph Attention Networks
Mar 5, 2026Zhiming Xue, Yujue Wang, Menghao HuoSupply Chain Management LLM Reasoning with Graphs
College of Engineering, Northeastern University, Boston, USA · Department of Chemistry and Chemical Biology, University of New Mexico, Albuquerque, USA · School of Engineering, Santa Clara University, Santa Clara, USA
TrinityGuard: A Unified Framework for Safeguarding Multi-Agent Systems
Mar 16, 2026Kai Wang, Biaojie Zeng, Zeming Wei +7LLM Agent Safety Multi-Agent System Security
Shanghai AI Laboratory

Sep 11, 2026J/K move · Enter open · S save

A Hybrid Agentic AI Framework for Intelligent Supply Chain Analytics

Xian Yeow Lee, Teppei Inoue, Haiyan Wang, Chetan Gupta

Industrial AI Lab Hitachi America

Data Analysis Agents Multi-Agent LLM Systems Multi-Agent Orchestration Supply Chain Management

Open paper AskPDF arXiv

Efficient utilization of supply chain analytics for decision making remains a significant challenge for planners, as critical tasks such as database querying, key performance indicator (KPI) analysis, demand forecasting, and performance diagnosis require heterogeneous expertise spanning data engineering, operations research, and domain knowledge. In this work, we propose an agentic system for supply chain analytics that bridges the gap between business decision-making and technical expertise, where a coordinator agent interprets user intent and delegates sub-tasks to specialized agents. The system supports both exploratory analysis and deterministic workflows, enabling planners to transition between ad hoc questions and structured processes. Domain logic is encapsulated within specialist agents and prompts, yielding a scalable, modular, and auditable design and lowering the cost of functional extension through prompt-centric development. We evaluate the proposed architecture on a test environment that replicates multi-echelon inventory management operations. Results show that our multi-agent design achieves a 90% accuracy, which is competitive with a single agent baseline while reducing input token usage by roughly fourfold, substantially improving scalability and cost-efficiency. Furthermore, we provide case studies to demonstrate interpretable suboptimality detection and automated forecast optimization, illustrating how agentic architectures can effectively combine open-ended exploratory analysis and deterministic supply chain analytics workflows, and provide a practical pathway toward more accessible and extensible decision-support systems.

Abstract

Explore similar work

A Hybrid Agentic AI Framework for Intelligent Supply Chain Analytics

LLM-Grounded Explainable AI for Supply Chain Risk Early Warning via Temporal Graph Attention Networks

TrinityGuard: A Unified Framework for Safeguarding Multi-Agent Systems