We propose an incremental search method for making planning decisions in real time. As an example, we present the problem of scheduling jobs in a factory as a real-time decision problem, and model the real-time constraints as strict limits on the amount of computation that can be performed before a scheduling decision must be made. Our approach is to use incremental heuristic search techniques to simulate and evaluate the effects of future decisions, and thereby generate each individual scheduling decision as needed. We present four incremental search algorithms that can be used to make real-time planning decisions. We also present preliminary results for the two-machine flowshop scheduling problem which indicate that this is a promising method for real-time planning.