Examinando por Materia "TOMA DE DECISIONES - MODELOS MATEMÁTICOS"
Mostrando 1 - 6 de 6
Resultados por página
Opciones de ordenación
Publicación Desarrollo de un agente de inteligencia artificial para la gestión preventiva de proyectos en una PMO del sector tecnológico mediante predicción explicable y recuperación aumentada de conocimiento(Universidad EAFIT, 2026-06) Jimeno Junca, Pablo Andrés; Saldarriaga Aristizábal, Pablo AndrésProject Management Offices (PMOs) accumulate historical records of risks, issues, decisions, and lessons learned that are rarely transformed into actionable knowledge during the planning of new projects. This gap limits the ability to anticipate situations that may affect project success and keeps risk management largely reactive. In response to this problem, this work develops and evaluates a prototype artificial intelligence agent, named the PMO Intelligence Hub, to support preventive project management by integrating explainable risk prediction, retrieval-augmented knowledge, and a conversational interface designed for project managers. The predictive component uses real historical data from a corporate PMO in the technology sector. The model was developed using 576 historical projects, with an overall risk rate of 18.4 %. The pipeline incorporates feature engineering, explicit treatment of missing data as an informative signal, class imbalance handling through SMOTE, isotonic probability calibration, and local explainability through SHAP values. On the test set, which had a positive-class base rate of 18.1 %, the Random Forest model achieved a Recall@0.35 of 0.810, a PR-AUC of 0.326 against a reference value of 0.181, and a Lift@20% of 1.528. The model outperformed the random baseline and obtained better point estimates than logistic regression; however, this comparison should be interpreted in light of the limited size of the test set. Consequently, the results position the model as an early prioritization and decision-support tool rather than as a deterministic classification mechanism. The retrieval-augmented knowledge component indexes 5 329 records of risks, issues, and lessons learned using semantic embeddings. Semantic search obtained better results than a lexical baseline across the three corpus tables, with MRR improvements ranging from +0.20 to +0.35 and ranking quality evaluated through nDCG@5. The generation evaluation using RAGAS achieved an average Faithfulness of 0.86, Answer Relevancy of 0.89, Context Precision of 0.60, and Context Recall of 0.37. These results suggest that the generated responses were mostly faithful and relevant, while also indicating opportunities to improve the selection and coverage of the retrieved context. The architecture is integrated through a conversational agent orchestrated by an LLM (GPT-4.1) with tool-use, capable of retrieving project information, calculating its risk probability, explaining the main contributing factors through SHAP, and automatically triggering an Intelligence Briefing mechanism. This mechanism retrieves similar projects, historical risks, issues, and relevant lessons learned, and presents the results through a functional web interface developed for the prototype. The results support the technical feasibility of integrating explainable predictive models and organizational knowledge retrieval into a unified conversational experience for preventive project planning. The system does not replace the project manager’s judgment; rather, it acts as a decision-support mechanism aimed at prioritizing attention, surfacing actionable historical knowledge, and explaining the main factors associated with the estimated risk. The evaluation was conducted in a prototype environment and did not include production validation, an experimental comparison against an LLM without RAG, or complete manual annotation of the corpus. This work contributes a reproducible architecture for corporate PMOs in the technology sector seeking to evolve from reactive risk management toward a preventive, explainable, and data-driven approach.Publicación Diagnóstico de la gestión financiera de la empresa Surtipag S.A.S.(Universidad EAFIT, 2026) Pérez de Oro, Ketty Laura; Uribe Marín, Ricardo; Giraldo Hernández, Gina MaríaPublicación Diseño de un laboratotio de alfabetización microempresarial desde el enfoque de desarrollo local(Universidad EAFIT, 2026-02-04) Martín Medina, Nubia Leonor; Avendaño Fernández, Eduardo; Ocensa - Alcaldía de Monterrey - EAFITThis research answers the question: ¿How to design a business literacy laboratory, anchored in local experiences, that strengthens the decision-making of microentrepreneurs in Monterrey (Casanare)? An applied qualitative study (single case, participatory action research) was developed with eight microentrepreneurs, integrating semi-structured interviews, participant observation, and co-creation workshops. The transcripts were processed in IRaMuTeQ using similarity analysis, descending hierarchical classification (Reinert), and factorial correspondence analysis, complemented by triangulation of sources. The findings reveal decisions made under high uncertainty, recurrent dependence on accounting advice, and tensions between immediate liquidity and professionalization. In addition, formalization and digitization (electronic invoicing, taxes, and banking) reconfigure routines and are modulated by attitudinal factors (fear, discipline, perseverance). As a result, the “Raíces y Estrategia Living Laboratory” was co-designed, with micro-challenges, a toolbox, and digital/community mediation aimed at continuous improvement, replicable and evaluable in future implementations.Publicación Gestión de riesgos financieros caso aplicado en I lab Colombia(Universidad EAFIT, 2025) Cabarcas Toscano, Jhon; Rodríguez Ramírez, Alexander; Guerrero Latorre, Jorge HarleyÍtem Knowledge modelling for supporting decision making in optimal distributed design process(IEEE, 2007-12-02) Mejía Gutiérrez, Ricardo; Fischer Estia, Xavier; Bennis, Fouad; Universidad EAFIT. Departamento de Ingeniería de Diseño; Ricardo Mejia (rmejiag@eafit.edu.co); Ingeniería de Diseño - GRIDA methodology for distributed knowledge modeling will be presented in this article, in order to contribute to a better decision making during design problems analysis -- An optimal design process refers to setting up coherent numerical models and it requires a well structured problem definition -- However, a lack of downstream information in early stages of product development leads to a complicated elicitation task that became an issue due to nowadays distributed environments -- A Multi-Agent approach is proposed to support the distributed knowledge elicitation process -- A set of agents will guide members from the distributed design team, throughout the product life cycle, to extract relevant information and analyze it -- The interaction among agents will highlight potential incoherencies during the modeling process, in order to enable partners to avoid inconsistent information -- A coherent knowledge base is then constructed and ready to be used to create models to be analyzed by traditional inference engines such as optimization solvers,constraint satisfaction programming, etcPublicación Modelación de una situación empresarial para la enseñanza de simulación discreta(Asociación Colombiana de Facultades de Ingeniería (ACOFI), 2013-09-24) Carmona González, Guillermo L.; Montoya Agudelo, Juan Sebastián; Cano Escobar, María Adelaida; Álvarez Zapata, Daniela; Rodríguez Betancur, María Antonia; Universidad EAFIT. Departamento de Ingeniería de Producción; mrodri23@eafit.edu.co; dalvar35@eafit.edu.co; mcanoes@eafit.edu.co; jmonto41@eafit.edu.co; gcarmona@eafit.edu.co; Gestión de Producción y LogísticaDiscrete simulation is one of the tools that are used for modeling real systems and for evaluating the impact of certain decisions using the resulting model, improving with this the decisions making on the real system -- Discrete simulation has been taught in the simulation course of the Operations Management and Logistics program at EAFIT University -- However this course only has academic exercises for the modeling processes -- This motivated to the realization of a project for documenting a business scenario on which students can build a simulation supported on a real situation -- This would allow students an approach to the modeling of real industry problems -- The productivity analysis of a production line of the Bakery Company Novapan in Medellin city was elected as the business scenario -- The simulation was made through the discreet simulation software Promodel ® -- This Paper presents the main aspects of the analyzed business scenario, the methodology, the discrete simulation model, some simulation results, the educational proposal and some learned lessons during the development of this research