A decision support system designed to enhance human-machine interaction in transportation scheduling is proposed. We aim to integrate human factors and ergonomics from the beginning of the design phase and to propose a system fitted with enough flexibility to be able to deal with the characteristics of a dynamic context such as transportation scheduling. In this interdisciplinary approach, a link is done between problem solving methods (operations research techniques and data classification algorithms) and human-machine interaction (solving control modes). A set of scheduler-oriented algorithms favouring human-machine cooperation for problem solving is proposed. Some of these algorithms have been efficiently tested on instances of the literature. Finally, an original framework aiming to assist scheduler in constraint relaxation when the problem becomes infeasible is proposed and evaluated.