LeaseLabs is active in the used car market and car leasing. They have a lot of knowledge in this field and wanted to combine this with a more data-driven approach. Notilyze has enabled Leaslabs to gain more data-driven insights, allowing them to work more efficiently and accurately.
Determine lease price
The focus of this project was to gain insight into which factors play a role and to what extent in the valuation of cars on the second-hand market. In addition, we analyzed the depreciation of cars over time for a number of reasons. First, it can be valuable to see how certain types of cars depreciate over time to determine when to buy and sell. Secondly, the depreciation of cars is an important part of the lease price and it is therefore very relevant to make good estimates.
We combined data on used car sales with publicly available data on car registration from the Dutch government (RDW). An analysis of the relationship between car price and various explanatory variables provided insight into which factors play a role in determining prices. Mileage, power, original price, vehicle age, among other things, play a role, which confirms our intuition.
After this analysis, we delved deeper into the data and developed various machine learning models that fit the data. Examples include linear regressions and a variety of tree-based models such as decision trees, random forests and gradient boosting. We then selected the best performing model, which estimates the value of a car on the used market based on several variables.
In addition, with this model we can provide insight into the depreciation of the car. By applying some business rules, potentially interesting vehicles can be selected for further research - or to monitor in the future - based on their depreciation history.
Using the model to estimate car prices in the used market, Notilyze developed several API services for LeaseLabs in SAS Intelligent Decisioning. One API simply provides an estimated price based on known car features, while another uses the model and other business rules to determine lease prices. In the latter, car functions are also entered and the service provides the lease price with an overview of the components.
With our analyses and developed APIs, LeaseLabs' decisions are now not only based on expertise and gut feeling, but now also on data! A good example of real-time Analytics as a Service.
Stationsplein 45 A4.004
3013 AK Rotterdam
+31 10 798 62 95