An Investigation of Offline Reinforcement Learning in Factorisable Action Spaces