Optimisation of fed-batch processes can be described as a constrained non-linear end-point dynamic optimisation problem. Although iterative dynamic programming (IDP) is feasible, it is usually more time-consuming and very difficult to apply to on-line optimisation because of solving the non-linear differential-algebraic equations of a process model at each iteration. The replacement of a rigorous mechanistic model by a neural network model takes the advantage of high speed processing, since simulation with a neural network model involves only a few non-iterative algebraic calculations. A modified algorithm, called iterative dynamic programming for discrete-time system, is proposed for neural network model based on-line re-optimisation control of fed-batch processes. The proposed scheme is illustrated using simulation studies of an ethanol fermentation process.