This paper, dedicated to controlled generalized batches Petri nets without discrete nodes, presents a method for computing a control trajectory for reaching a steady state from a given initial marking. A steady state is characterized by a state in which the marking and the firing flow vector are constant. By controlling the firing flow vector of transitions, the proposed control strategy is an event-based one and relies on an algorithm solving a linear programming problem. This new control strategy, called in this paper as maximal flow based ON/OFF control, exploits the maximal firing flows and reduces the delay of the transient behavior for reaching the steady state. A practical communication system is provided to illustrate the relevance of such a control strategy for Cyber-Physical Systems.