Inteligencia de negocios aplicada a la gestión de ventas en una empresa de consumo masivo como herramienta de ayuda para la toma de decisiones
This research focuses on optimizing sales management by applying data analysis models in the business context. Good sales management is essential in the business environment, as it encompasses planning, coordinating and monitoring sales activities. However, the ability to meet customer expectations...
Salvato in:
| Autore principale: | |
|---|---|
| Natura: | masterThesis |
| Lingua: | spa |
| Pubblicazione: |
2024
|
| Soggetti: | |
| Accesso online: | https://repositorio.uteq.edu.ec/handle/43000/7843 |
| Tags: |
Aggiungi Tag
Nessun Tag, puoi essere il primo ad aggiungerne!!
|
Aggiungi Tag | Riassunto: | This research focuses on optimizing sales management by applying data analysis models in the business context. Good sales management is essential in the business environment, as it encompasses planning, coordinating and monitoring sales activities. However, the ability to meet customer expectations faces several challenges. These include the difficulty in anticipating future customer needs, identifying opportunities to improve sales, and properly segmenting customers to detect those who might be at risk of churn. The research addresses these issues through several approaches. First, time series models are employed to forecast product demand. Second, association algorithms are used to identify cross-selling opportunities. Finally, customer segmentation is performed to detect those at risk of churn, facilitating the implementation of preventive measures. For product demand forecasting, several time series models were applied: Prophet, Exponential Smoothing, SARIMA, Random Forest, and XGBoost. The MAPE metric was used to assess the accuracy of the predictions. Although XGBoost showed the best average accuracy, Random Forest turned out to be the most consistent model and was frequently selected as the best fit for most products. Furthermore, association rule analysis using the FP-Growth algorithm proved to be more efficient than Apriori, highlighting its ability to generate personalized recommendations and increase revenue per customer. Customer segmentation using hierarchical clustering techniques revealed significant patterns that facilitate the identification of customers at risk of churn and the implementation of more effective retention strategies. Thus, this work contributes knowledge to the business sector in Ecuador, by offering tools and strategies that can provide a competitive advantage in a dynamic and competitive environment. |
|---|