The aim of this study is to improve the modeling of emerging road freight technologies in the Swedish national freight model Samgods and to develop methods for generating and analyzing large ensembles of scenarios that account for uncertainties related to the development and impacts of such technologies. The focus is on modeling of battery electric and driverless trucks. A case study is conducted to explore the impacts of input uncertainty related to the development of battery electric and driverless trucks on the current base forecast for 2045. In the base forecast, it is assumed that a decarbonization of the road freight transport system will be achieved through large-scale adoption of battery electric trucks complemented by conventional trucks operating with 100% biofuels.
A set of tools has been developed which includes the following: Extension of the logistics module in Samgods (Logmod) to include three parallel groups of trucks that can complement and compete with each other. Model representations of key aspects of truck electrification and automated driving such as: routing that accounts for driving range and charging infrastructure locations, impacts on vehicle characteristics and operating costs, and operational constraints for driverless trucks depending on their automation level.
The modeling of battery electric and automated trucks is achieved through a combination of input models that calculate scenario specific vehicle parameters, exogenous datasets of scenarios for public charging locations and routing tools to calculate route-specific charging patterns and their associated time and monetary costs and to account for route-specific feasibility of driverless truck operations. The results include impacts in terms of key system indicators including modal split, the share of transport work by each truck technology type, system costs, biofuel and electricity use. Tools for scenario discovery are used to identify scenario conditions that generate substantial deviations in outcomes compared to the base forecast.