Tesi etd-02122025-180814 |
Link copiato negli appunti
Tipo di tesi
Tesi di laurea magistrale
Autore
HAKIM, SARA
URN
etd-02122025-180814
Titolo
Sales Forecasting for Tecnoinox: a Machine Learning approach for professional kitchen equipment
Dipartimento
INFORMATICA
Corso di studi
DATA SCIENCE AND BUSINESS INFORMATICS
Relatori
relatore Prof. Trasarti, Roberto
Parole chiave
- Class imbalance
- Forecasting
- Machine Learning
- Time series
Data inizio appello
28/02/2025
Consultabilità
Non consultabile
Data di rilascio
28/02/2095
Riassunto
In an increasingly competitive market landscape, sales forecasting has become essential for companies striving to maintain a strategic edge. This thesis focuses on forecasting sales for Tecnoinox, a leading Italian manufacturer of professional kitchen equipment. Founded in 1984, Tecnoinox has expanded globally, integrating Industry 4.0 technologies and emphasizing sustainability in its operations. By leveraging historical sales data from 2018 to 2023, this study addresses challenges posed by the division of data into product lines and families.
A customized segmentation of these categories was developed to optimize forecasting models for both well-represented and sparse datasets. Special attention was given to irregular product lines characterized by high variability, which required tailored approaches to improve prediction accuracy. Various machine learning, statistical, and deep learning models are compared to evaluate their performance.
The findings provide insights into the effectiveness of different models and offer practical recommendations for enhancing sales strategies and operational efficiency. By anticipating sales demand, Tecnoinox can better align production and marketing efforts, strengthening its position in the global market.
A customized segmentation of these categories was developed to optimize forecasting models for both well-represented and sparse datasets. Special attention was given to irregular product lines characterized by high variability, which required tailored approaches to improve prediction accuracy. Various machine learning, statistical, and deep learning models are compared to evaluate their performance.
The findings provide insights into the effectiveness of different models and offer practical recommendations for enhancing sales strategies and operational efficiency. By anticipating sales demand, Tecnoinox can better align production and marketing efforts, strengthening its position in the global market.
File
Nome file | Dimensione |
---|---|
Tesi non consultabile. |