Differing in an important aspect from other process reasoning and decision methods based on shape similarity and semantic similarity, the proposed methods considers sufficiently the effect of the selection rule of a feature operation chain on process design activity. Mining the selection rule of the feature operation chain and its correction result are regarded as the supporting evidence of design-making, which ensure the effectiveness of process design. First of all, the process realization requirements are grouped, and the selection rule is mined by conducting statistics of the feature operation chain that matching specific requirements. Then the theoretical selection probability of the matched feature operation chain is analyzed, and the actual selection rule is corrected by using the obtained theoretical selection probability. Finally, the correction selection probability is held as evidence, and then the D-S evidence theory is used for decision-making of the feature operation chain. Experimental results show the effectiveness of proposed method.