Multi-Agent Multimodal Large Language Model Framework for Automated Interpretation of Fuel Efficiency Analytics in Public Transportation

Ma, Zhipeng; Bahja, Ali Rida; Burgdorf, Andreas; Pomp, André; Meisen, Tobias; Jørgensen, Bo Nørregaard; Ma, Zheng Grace

doi:10.3390/app152111619

This is an early access version, the complete PDF, HTML, and XML versions will be available soon.

Open AccessArticle

Multi-Agent Multimodal Large Language Model Framework for Automated Interpretation of Fuel Efficiency Analytics in Public Transportation

by

Zhipeng Ma

¹

,

Ali Rida Bahja

²,

Andreas Burgdorf

²,

André Pomp

²

,

Tobias Meisen

²

,

Bo Nørregaard Jørgensen

^1,*

and

Zheng Grace Ma

¹

SDU Center for Energy Informatics, The Maersk Mc-Kinney Moller Institute, Faculty of Engineering, University of Southern Denmark, DK-5230 Odense, Denmark

²

Institute for Technologies and Management of Digital Transformation, University of Wuppertal, D-42119 Wuppertal, Germany

^*

Author to whom correspondence should be addressed.

Appl. Sci. 2025, 15(21), 11619; https://doi.org/10.3390/app152111619

Submission received: 13 October 2025 / Revised: 27 October 2025 / Accepted: 28 October 2025 / Published: 30 October 2025

(This article belongs to the Special Issue Enhancing User Experience in Automation and Control Systems)

Download Versions Notes

Abstract

Enhancing fuel efficiency in public transportation requires the integration of complex multimodal data into interpretable, decision-relevant insights. However, traditional analytics and visualization methods often yield fragmented outputs that demand extensive human interpretation, limiting scalability and consistency. This study presents a multi-agent framework that leverages multimodal large language models (LLMs) to automate data narration and energy insight generation. The framework coordinates three specialized agents, including a data narration agent, an LLM-as-a-judge agent, and an optional human-in-the-loop evaluator, to iteratively transform analytical artifacts into coherent, stakeholder-oriented reports. The system is validated through a real-world case study on public bus transportation in Northern Jutland, Denmark, where fuel efficiency data from 4006 trips are analyzed using Gaussian Mixture Model clustering. Comparative experiments across five state-of-the-art LLMs and three prompting paradigms identify GPT-4.1 mini with Chain-of-Thought prompting as the optimal configuration, achieving 97.3% narrative accuracy while balancing interpretability and computational cost. The findings demonstrate that multi-agent orchestration significantly enhances factual precision, coherence, and scalability in LLM-based reporting. The proposed framework establishes a replicable and domain-adaptive methodology for AI-driven narrative generation and decision support in energy informatics.

Keywords: multimodal LLM; data narration; multi-agent system; fuel efficiency; industrial decision support; public transportation

Share and Cite

MDPI and ACS Style

Ma, Z.; Bahja, A.R.; Burgdorf, A.; Pomp, A.; Meisen, T.; Jørgensen, B.N.; Ma, Z.G. Multi-Agent Multimodal Large Language Model Framework for Automated Interpretation of Fuel Efficiency Analytics in Public Transportation. Appl. Sci. 2025, 15, 11619. https://doi.org/10.3390/app152111619

AMA Style

Ma Z, Bahja AR, Burgdorf A, Pomp A, Meisen T, Jørgensen BN, Ma ZG. Multi-Agent Multimodal Large Language Model Framework for Automated Interpretation of Fuel Efficiency Analytics in Public Transportation. Applied Sciences. 2025; 15(21):11619. https://doi.org/10.3390/app152111619

Chicago/Turabian Style

Ma, Zhipeng, Ali Rida Bahja, Andreas Burgdorf, André Pomp, Tobias Meisen, Bo Nørregaard Jørgensen, and Zheng Grace Ma. 2025. "Multi-Agent Multimodal Large Language Model Framework for Automated Interpretation of Fuel Efficiency Analytics in Public Transportation" Applied Sciences 15, no. 21: 11619. https://doi.org/10.3390/app152111619

APA Style

Ma, Z., Bahja, A. R., Burgdorf, A., Pomp, A., Meisen, T., Jørgensen, B. N., & Ma, Z. G. (2025). Multi-Agent Multimodal Large Language Model Framework for Automated Interpretation of Fuel Efficiency Analytics in Public Transportation. Applied Sciences, 15(21), 11619. https://doi.org/10.3390/app152111619

Note that from the first issue of 2016, this journal uses article numbers instead of page numbers. See further details here.

Article Menu

Multi-Agent Multimodal Large Language Model Framework for Automated Interpretation of Fuel Efficiency Analytics in Public Transportation

Abstract

Share and Cite

Article Metrics

Article Access Statistics

Further Information

Guidelines

MDPI Initiatives

Follow MDPI