Sistema multiagente enfocado en procesos de inteligencia artificial distribuida para la captura y el ruteo de pedidos

Anny Astrid Espitia Cubillos; Robinson Jiménez Moreno; Mateo Andrés Pulido Aponte

doi:10.24054/rcta.v2i46.3530

Authors

Anny Astrid Espitia Cubillos Universidad Militar Nueva Granada https://orcid.org/0000-0002-4791-0250
Robinson Jiménez Moreno Universidad Militar Nueva Granada https://orcid.org/0000-0002-4812-3734
Mateo Andrés Pulido Aponte Universidad Militar Nueva Granada https://orcid.org/0009-0006-5858-7184

DOI:

https://doi.org/10.24054/rcta.v2i46.3530

Keywords:

app, reasoning and action agent, optimization, databases, routing, artificial intelligence, multi-agent system

Abstract

This paper presents the design of a multi-agent system focused on text-operated distributed artificial intelligence processes that works following a dialog model, oriented to the efficient management of industrial product orders. This system integrates five reasoning and action agents: the general agent powered by Google's Gemini artificial intelligence language model that carries the conversation flow and redirects the tasks to the other agents, an agent that identifies customers, an agent that recognizes the products, an agent that generates the orders and an agent that routes. The multi-agent system is designed to interact virtually and conversationally with users, facilitating order creation and management through an innovative approach that integrates advanced natural language processing technologies, vector and relational databases, and optimization methods. Finally, for its validation in a real environment, data from a company that produces and distributes cleaning products are used, allowing the development of different tests for the identification of customers, products, quantities and the establishment of order delivery routes. It is concluded that the conversational flow and techniques used allow users to make queries about customers with an average accuracy of 77.38% and about products with an average accuracy of 88.57%, even in scenarios with semantic ambiguities, managing orders in an intuitive way. It is also possible to optimize order routing by simultaneously considering two criteria that can be weighted: customer importance and distance traveled.

Downloads

Download data is not yet available.

References

A. Börold, D. Schweers y M. Freitag, «Towards Multimodal Information Systems for Assisting Humans in Production and Logistics Processes,» Procedia CIRP, vol. 120, pp. 1089-1094, 2023. https://doi.org/10.1016/j.procir.2023.09.130.

A. Dörner, P. Gerald, S. Michal y B. Marek, «Use of Lean Management Methods based on Eye-Tracking Information to make User Interfaces in Production more Human-centered,» vol. 128, pp. 514-519, 2024. https://doi.org/10.1016/j.procir.2024.04.014.

S. Vernim, H. Bauer, E. Rauch, M. T. Ziegler y S. Umbrello, «A value sensitive design approach for designing AI-based worker assistance systems in manufacturing,,» Procedia Computer Science, vol. 200, pp. 505-516, 2022. https://doi.org/10.1016/j.procs.2022.01.248.

M. Pellicelli, «Chapter Two - A long road to maximizing efficiency,» de The Digital Transformation of Supply Chain Management, Elsevier, 2023. https://doi.org/10.1016/B978-0-323-85532-7.00009-8, pp. 37-61.

I. Ghannouchi, «Examining the dynamic nexus between industry 4.0 technologies and sustainable economy: New insights from empirical evidence using GMM estimator across 20 OECD nations,» Technology in Society, vol. 75, p. 102408, 2023. https://doi.org/10.1016/j.techsoc.2023.102408

A. Salis, A. Marguglio, G. D. Luca, S. Razzetti, W. Quadrini y S. Gusmeroli, «An Edge-Cloud based Reference Architecture to support cognitive solutions in Process Industry,» Procedia Computer Science, vol. 217, pp. 20-30, 2023. https://doi.org/10.1016/j.procs.2022.12.198.

T. A. Aniceski, L. T. Pontes Miranda, O. Canciglieri Junior y G. Brittes Benitez, «The Four Smarts of Industry 4.0 and barriers for technology deployment: A TOE perspective,» Computers & Industrial Engineering, vol. 193, p. 110345, 2024. https://doi.org/10.1016/j.cie.2024.110345.

J. S. Dhatterwal, K. S. Kaswan y R. P. Ojha, «Chapter 13. The role of multiagent system in industry 4.0.,» de A Roadmap for Enabling Industry 4.0 by Artificial Intelligence, 2022. https://doi.org/10.1002/9781119905141.ch13, pp. 227-246..

B. G. Mark, E. Rauch y D. T. Matt, «Systematic selection methodology for worker assistance systems in manufacturing,» Computers & Industrial Engineering, vol. 166, p. 107982, 2022. https://doi.org/10.1016/j.cie.2022.107982.

G. Bitsch, P. Senjic y J. Askin, «Dynamic adaption in cyber-physical production systems based on ontologies,» Procedia Computer Science, vol. 200, pp. 577-584, 2022. https://doi.org/10.1016/j.procs.2022.01.255.

S. Mantravadi, C. Møller, C. LI y R. Schnyder, «Design choices for next-generation IIoT-connected MES/MOM: An empirical study on smart factories,» Robotics and Computer-Integrated Manufacturing, vol. 73, p. 102225, 2022. https://doi.org/10.1016/j.rcim.2021.102225.

X. Sun y Y. Song, «Unlocking the Synergy: Increasing productivity through Human-AI collaboration in the industry 5.0 Era,» Computers & Industrial Engineering, vol. 200, p. 110657, 2025. https://doi.org/10.1016/j.cie.2024.110657.

C. Toukmaji y A. Tee, «Retrieval-Augmented Generation and LLM Agents for Biomimicry Design Solutions,» de Proceedings of the AAAI Symposium Series, 2024. https://doi.org/10.1609/aaaiss.v3i1.31210.

S. Yao, J. Zhao, D. Yu, N. Du, I. Shafran, K. Narasimhan y Y. Cao, «ReAct: Synergizing Reasoning and Acting in Language Models,» 2023. https://doi.org/10.48550/arXiv.2210.03629.

H. R. Saeidnia, «Welcome to the Gemini era: Google DeepMind and the information industry,» Library Hi Tech News, 2023. https://doi.org/10.1108/LHTN-12-2023-0214.

Z. Jing, Y. Su y Y. Han, «When large language models meet vector databases: A survey,» 2024. https://doi.org/10.48550/arXiv.2402.01763 .

P. N. Singh, S. Talasila y S. V. Banakar, «Analyzing Embedding Models for Embedding Vectors in Vector Databases,» de 2023 IEEE International Conference on ICT in Business Industry & Government (ICTBIG), Indore, India, 2023. https://doi.org/10.1109/ICTBIG59752.2023.10455990.

A. Makris, K. Tserpes, G. Spiliopoulos, D. Zissis y D. Anagnostopoulos, «MongoDB Vs PostgreSQL: A comparative study on performance aspects,» Geoinformatica, vol. 25, p. 243–268, 2021. https://doi.org/10.1007/s10707-020-00407-w.

R. K. Malviya, V. Javalkar y R. Malviya, «Scalability and Performance Benchmarking of LangChain, LlamaIndex, and Haystack for Enterprise AI Customer Support Systems,» de IJGIS Fall of 2024 Conference. The New World Foundation., 2024. https://doi.org/10.21428/e90189c8.43aeb06e.

Q. Guan, X. Hong, W. Ke, L. Zhang, G. Sun y Y. Gong, «Kohonen Self-Organizing Map based Route Planning: A Revisit,» de 2021 IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS), Prague, Czech Republic, 2021. https://doi.org/10.1109/IROS51168.2021.9636025.

Y. Akkem, B. S. Kumar y A. Varanasi, «Streamlit Application for Advanced Ensemble Learning Methods in Crop Recommendation Systems – A Review and Implementation,» Indian Journal of Science and Technology, vol. 16, nº 48, pp. 4688-4702, 2023. https://doi.org/10.17485/IJST/v16i48.2850.

G. Schuh, M.-F. Stroh y L. Johanning, «Framework To Design Compliance Rules For Digital Technologies In Manufacturing Companies,» Procedia CIRP, vol. 119, pp. 103-108, 2023. https://doi.org/10.1016/j.procir.2023.04.005.

M. Habib, M. Faris, A. Alomari y H. Faris, «AltibbiVec: A Word Embedding Model for Medical and Health Applications in the Arabic Language,» IEEE Access, vol. 9, pp. 133875-133888, 2021. https://doi.org/10.1109/ACCESS.2021.3115617.

Y. Matsuo, Y. LeCun, M. Sahani, D. Precup, D. Silver, M. Sugiyama, E. Uchibe y J. Morimoto, «Deep learning, reinforcement learning, and world models,» Neural Networks, vol. 152, pp. 267-275, 2022. https://doi.org/10.1016/j.neunet.2022.03.037.