Ensuring the quality of important machining features with low production cost is one of the main objectives of process planning. Owing to the difficulty in identifying the relative importance of features, most automatic process planning systems manipulate all features equally. This often leads to the inability of the system to generate valid process sequences when more features and constraints are involved. This paper reports a fuzzy model that uses the concept of 'feature manufacturability' to evaluate feature priorities and identify important features. The model is created by means of the construction of parametric fuzzy membership functions and fuzzy objective functions. These functions, based on neural networks methodologies, enable the evaluation of the complexity of features in a part description model and the manufacturing capability in an environment description model simultaneously. After feature prioritization, operation sequencing of the important features can be carried out first within a much smaller search space and the operations of the less important features can be arranged easily due to reduced constraints.