When applying transport policies or business strategies to a trans- portation system, different effects, such as environmental and economic effects, typically appear. Simulation can be used to predict such effects for a given pol- icy or strategy, and we argue that to fully capture such effects there is a need to model the underlying decisions and actions in the transport chain. The aim of this paper is to identify and discuss some important transport chain responsibilities and roles that need to be captured in a simulation model in order to study these types of effects. The roles concern basic functionality related to using and providing services, and the responsibilities include decisions and actions that concern ordering of products and transports, and planning of how to use resources and infrastructure. We contribute with a framework, which on an abstract level describes responsibilities, roles and interactions that are important to consider in an agent-based transport chain simulation model. We argue that our framework can be used as assistance when developing an agent-based transport simulation model. To illustrate how the framework can be realized, we briefly describe an agent-based simulation tool called TAPAS.