The peak distribution of working surface profile measured on the machine generates the peculiar change pattern corresponding to the type of change in working surface condition of grinding wheel. In this paper, based on the results of pattern recognition of peak distribution, the quantitative evaluation method of progress of change in working surface condition is proposed. Furthermore, the proposed method is examined for its validity due to apply it to the actual grinding processes. By the results, following are found out. The pattern recognition of peak distribution determines the evaluation value distribution, which includes peak and valley generated in different regions corresponding to the change in working surface condition. The peak height and valley depth indicate the progress of change in working surface condition, and then those values properly become the evaluation index of the condition change. The change in the index is confirmed to be corresponding to the progress of change in working surface condition with actual grinding process. Furthermore, the evaluation results based on the index make it possible to properly explain the changes in ground surface roughness and grinding force.