Today, the automotive industry is faced with more pressure to realize higher quality products, lower production costs and shorter lead times. The key to the success of organizations is the effective integration of design and other downstream applications such as machining, process planning, analysis, assembly, inspection etc. This paper reviews the current research in feature recognition and neural network applications in feature-based environment. The central issues associated with feature-based environment and drawbacks of the current approaches are addressed. The future research directions and proposals to overcome the shortcomings are presented.