A system for optimizing a stimulation operation includes an interface to equipment and sensors for performing the stimulation operation. The interface supplies control signals to the equipment and obtains measurement signals from the sensors. The system further includes a short-term optimizer that derives a current job state based at least in part on the measurement signals, and that further adjusts the control signals to optimize a short-term cost function. The short-term cost function includes a difference between the current job state and a desired job state derived from optimized values of a set of decision variables. The system further includes a long-term optimizer module that determines the optimized values based on a long-term cost function, the long-term cost function accounting for at least a long-term reward and a final state cost.