Accurately predicting the delivery time for vehicles often proves to be difficult. The complexity inherent in supply chains and additional factors influencing them in general, make it near impossible to derive meaningful calculations and predictions by hand. The automotive industry faced many challenges to overcome over the past few years. COVID-19 not only limited resources but caused all kinds of disruptions along the supply chain. The War in Ukraine further added to that instability and both events caused longer delivery times for vehicles, parts, and even raw materials.

This Bachelor Thesis gives an overview over the client’s supply chain and a literature review analyses the factors possibly affecting it and how they interact with delivery times. Additionally, a basic understanding for Machine Learning algorithms have been explained and additionally, three Machine Learning models have been developed using RStudio with the goal of simplifying predictions for delivery times, while also being more accurate. A Linear Regression model, a Decision Tree model, and a Random Forest model have been developed.

In conclusion this thesis gives a crucial insight on how supply chain disruptions, a lack of transparency or politics in business can affect delivery times and their predictions respectively. This thesis also stressed the importance of data driven decision making, while keeping in mind that not all data is valuable. The final models can all be further developed or also deployed and therefore be put into practice at the client company.