The review on applications of artificial neural networks is aimed at modelling, optimisation, control and regulation of chemical processes. Attention has been predominantly paid to the types of neural networks and to the choice of geometries. A substantial proportion of applications is focused on modelling and optimisation of refining and separation processes and on the control of chemical reactors. Feed-forward neural networks are used in most models. Recurrent neural networks are usually implemented in dynamic models depending on their capability of prediction of time series. A common feature of the reviewed methods is their efficiency in solving problems by neural networks.