Understanding Factors Affecting Fuel Consumption of Vehicles Through Explainable Boosting Machines

A significant economic cost for many companies that operate with fleets of diesel and petrol vehicles is related to fuel consumption. Consumption can be reduced by acting over some factors, like driving behaviour style. Improving these factors can reduce the fuel usage of a vehicle without changing other aspects, such as planned routes or stops. This mitigates economic costs while reducing emissions associated to fuel consumption. In this paper we show how Explainable Artificial Intelligence (XAI) is useful for quantifying the impact that fuel factors have on the consumption of a vehicle fleet. We use Explainable Boosting Machines (EBM), trained over different features in order to both model and explain the relationship between them and fuel consumption, and then assess quality of the explanations using prior domain knowledge. We work with real-world industry datasets that represent different types of vehicles, from passenger cars to heavy-duty trucks.