logo SBA

ETD

Archivio digitale delle tesi discusse presso l’Università di Pisa

Tesi etd-01232024-180945


Tipo di tesi
Tesi di laurea magistrale
Autore
CAVALIERI, TOMMASO
URN
etd-01232024-180945
Titolo
Data-Driven Approaches to NPI Demand Forecasting: A Case Study of Luxottica's Wholesale Clients
Dipartimento
INFORMATICA
Corso di studi
DATA SCIENCE AND BUSINESS INFORMATICS
Relatori
relatore Prof. Venturini, Rossano
tutor Dott. Motta, Davide
Parole chiave
  • data-driven decision-making
  • demand planning
  • forecasting
  • new product introduction
  • predictive analytics
  • supply chain management
Data inizio appello
23/02/2024
Consultabilità
Non consultabile
Data di rilascio
23/02/2094
Riassunto
In the rapidly evolving landscape of global markets, where consumer preferences shift with unprecedented speed, the capability to accurately forecast demand for New Product Introductions (NPI) has become a critical determinant of business competitiveness. This thesis delves into the strategic implementation of advanced predictive analytics to revolutionize demand planning at Luxottica, a leader in the global eyewear industry, by presenting a case study of its wholesale clients. The study meticulously evaluates the efficacy of advanced data science techniques, i.e. time series forecasting, regression and classification models, against traditional forecasting methods, which predominantly rely on qualitative insights and intuitive judgment. The analysis underlines the models' robustness in reducing forecast errors, thereby enhancing inventory management and production planning, key points for maintaining an efficient supply chain and reach customer satisfaction. By running a simulation on a period spanning over the last year and a half, forecast error at lag 1 showed an outstanding decrease when applying more innovative procedures, from around 24\% to less than 7\%. Furthermore, the research articulates the broader business implications of integrating data-driven forecasting models, including organizational optimization and the potential for substantial cost reductions, highlighting the strategic value of adopting advanced analytics in global supply chain operations. By bridging the gap between academic theoretical models and practical applications in a real-world business setting, this work contributes significantly to the literature on supply chain management. It also provides valuable insights for enhancing operational efficiency and market responsiveness, offering a blueprint for operational excellence and strategic advantage in the competitive global marketplace. Looking forward, the enrichment of data resources for a more data-oriented design of Luxottica's supply chain would be fundamental to properly support the expansion of real-time data analytics and the exploration of machine learning algorithms, whose adaptability promise to further elevate the precision of demand forecasting.
File