Feature technology is considered an indispensable tool for computer integrated manufacturing (CIM) which is a focal research area in the manufacturing industry. Although various methods have been proposed, there are still problems hindering its practical applications. This paper presents an intelligent manufacturing feature extraction method employing artificial neural network (ANN) techniques, design by features and heuristic algorithms. After a brief study of the current state, the paper presents three main sections related to the proposed method, including the adoption of design by features, a heuristic algorithm of interacting feature identification and a novel ANN-based feature recognition method. The implementation and testing results are discussed. Finally, conclusions and future work are summarised.