Predicting Next Purchase Day and Order Volume for Customers in a Supply Chain
TINE SA serves over 20,000 business customers who place orders directly with the company. Each customer exhibits unique purchasing behaviors influenced by various factors, including seasonality and product shelf-life constraints. These behaviors are further shaped by individual seasonality patterns (such as stable or variable seasonal dates, geographically specific holidays, and annual events) and distinct warehouse management strategies (such as stockpiling or maintaining a consistent order policy).
This thesis aims to develop a predictive model (or a set of models) to accurately forecast the next purchase date and order volume for each customer. By understanding and predicting these patterns, TINE SA can optimize inventory management and improve supply chain efficiency. We assume that to analyze and predict customer purchasing behaviors it could be useful to use some of advanced machine learning techniques, such as clustering algorithms, random forest algorithms and neural networks. The model is expected to take into account various influencing factors, including seasonality, product shelf-life, and individual customer ordering patterns.
Goal
Develop a predictive model (or a set of models) to accurately forecast the next purchase date and order volume for each customer.
Learning outcome
- A comprehensive model that predicts the next purchase date and order volume for each customer.
- Insights into the unique purchasing behaviors of TINE SA's diverse customer base.
Qualifications
- Statistical Learning and AI methods
Supervisors
- Pedro Lind
Collaboration partners
- The supervision will be done together with data scientists in Tine (https://www.tine.no/)
