The NP-complete, deterministic scheduling problem for a single server system is addressed. Given a set of n tasks along with the precedence constraints among them, their timing requirements, setup costs, and their completion deadlines, a neuromorphic model is used to construct a nonpreemptive optimal processing schedule so that the total completion time, total tardiness, and the number of tardy jobs are minimized. This model exhibits faster convergence than techniques based on gradient projection methods.
