Scheduling algorithms are usually based on the assumption that we know the exact durations $d_i$ of all the tasks that we want to schedule. In real life, the durations $d_i$ may vary. In rare cases when we know the probabilities of different durations, we can apply stochastic scheduling methods. Most often, however, we do not know the probabilities, we only know the upper bound $d^+_i$ and the lower bound $d^-_i$ for the duration $d_i$; in other words, we know an interval $[d^-_i,d^+_i]$ of possible values of duration $d_i$. In these situations, we want a schedule that satisfies the given constraints for all possible values of durations $d_i$ from the given intervals. An algorithm for producing such a schedule is given in this paper. As an example, this algorithm is applied to a railway station.