In order to help manager to make termination decision for the ongoing R&D project timely and correctly, the authors review and compare the theory and methodology of termination decision from the literature. Furthermore, we discuss the considerable importance to identify attributes of an ongoing R&D project. To identify attributes of an ongoing R&D project, the authors examine characteristics and scopes of neural network. Thus, the feasibility of neural network, which is possibly used in R&D project termination decisions, is further explored. On the basis of the theory and methodology of neural network, this study attempts to develop a mode identification method to make R&D project termination decisions in an effective manner. Finally, the validity of the proposed method is verified by an example. The result shows that the method is both simple and practical.