Cost Forecasting for Extended Warranty contracts Application to a real case study in the automotive sector

In the era of Big Data many companies are relying on data-driven decisions to contribute to the achievement of business targets. One of the most challenging yet interesting challenges a company faces is the ability to adopt and develop increasingly competitive and innovative methodologies. The role of new technologies is to extract valuable knowledge and derive data-driven insights to enhance business performance. This thesis project deals with a topic of relevance to the automotive industry: Predictive Analytics, which has become a cornerstone of modern business strategy, enabling companies to anticipate future trends, optimize operations and make data-driven decisions. Specifically, the aim of the project is the development and implementation of predictive models in order to accurately forecast the end-of-life costs of Extended Warranty contracts, for vehicles used in agricultural and construction sectors. Forecast analysis supports and improves the management of warranty contracts, by reducing maintenance costs and allocating funds for ongoing contracts effectively. To make predictive analytics successful, we applied a rigorous data science methodology, known as KDD Process, which includes data cleaning, preparation, transformation and feature engineering. Model evaluation and validation techniques ensure the reliability and robustness of predictions. The project provides a comprehensive overview of Machine Learning algorithms, heart of predictive analytics, which are able to find patterns in data and to achieve the underlying objective of the project. We developed several Machine Learning models (Linear Regression, Random Forest, XGBoost, K-Means), features and hyperparameters selection techniques (Correlation Analysis, Recursive Feature Elimination, domain expert support and Cross Validation), in order to reach the best combination model-purpose. The entire methodology is applied to a real case study, chosen according to the needs of the companies involved: Accenture, a consulting partner, and CNH, worldleading company in the design, production and sale of agricultural machinery and construction equipment.