Please use this identifier to cite or link to this item:
http://hdl.handle.net/10609/150787
Title: | Previsión de demanda mediante técnicas de machine learning |
Author: | Calvo Martucci, Sebastián |
Tutor: | Polo, Lorena |
Others: | Benito Altamirano, Ismael |
Abstract: | Stock management is crucial for a company’s operational performance, facing challenges ranging from supplier issues to unexpected changes in demand and market promotions. This project tackles this complexity by creating a predictive model to estimate demand more accurately and facilitate stock management. To achieve this goal, the project begins with the justification and contextualization, establishing objectives and the working methodology. Then, existing literature on proposed solutions is reviewed to provide a foundation. Additionally, a technique for estimating inventory calculations is investigated. Following this, data from a company that sells automotive products is analyzed, involving data exploration and cleaning to ensure quality. Correlations between variables are examined, features are created, and clustering techniques are applied. Once the exploration and preprocessing stages are completed, model selection proceeds, detailing the accuracy indicators and hyperparameters used. Subsequently, the accuracy results from the different selected models are evaluated and ranked. Finally, the economic impact of the improved algorithm is studied, and the project’s conclusions are presented, including lessons learned, achievement of objectives, and potential future work directions. |
Keywords: | stock management machine learning predictive algorithms predictive model |
Document type: | info:eu-repo/semantics/masterThesis |
Issue Date: | 20-Jun-2024 |
Publication license: | http://creativecommons.org/licenses/by-nc-nd/3.0/es/ |
Appears in Collections: | Bachelor thesis, research projects, etc. |
Files in This Item:
File | Description | Size | Format | |
---|---|---|---|---|
sebascmTFM0624memoria.pdf | Memoria del TFM | 2,27 MB | Adobe PDF | View/Open |
Share:
This item is licensed under aCreative Commons License