In this paper, we advocate the use of reactive robots in industrial process control and production management. It is explained why reactive robots are well-suited to modern industrial applications that necessitate a high degree of autonomy and reactivity to unforeseen events and an illustrative example is studied in depth. A reactive robot whose role is to regulate the demands of a set of service units for tool use is described and different behavioral models for it studied and compared. Extensive simulation studies have revealed that a behavioral model relying on individual independent motivations of the robot is sufficient for it to exhibit the “optimal” behavior in this case. The process is modeled as a dynamical system that tends by itself to a steady state and that is perturbed by the robot. In turn, the robot's motivations have a steady state of their own which is contradictory with the process' steady state and hence is perturbed by its presence. As a consequence, the robot and the process are two coupled dynamical systems that perturb each other as each one tries to arrive to its steady state. For the designer, modeling of the overall problem in this way, has as a consequence that the motivational state space of the robot may be designed in advance given the process characteristics and this design should be based on a worst-case analysis. The mythical distinction between reactivity and planning is also rediscussed and the notion of operationality as opposed to optimality is explored.