Abstract: Calculating the score and accuracy of the picture fuzzy value is a key step in the multiple attribute decision making problem. The important tools of this step are the score function and accuracy function. The existing functions are faced with two problems: on the one hand, for some obviously different picture fuzzy value, the calculation results of the existing functions are contrary to the basic theory; on the other hand, it is very cumbersome to use the score function and precision function together in the multiple attribute decision making. In order to solve the above problems, a novel score function whose calculation results always conform to the basic definition without the aid of precision function is proposed. At the same time, the properties of the novel score function are discussed and proved. In addition, the multiple attribute decision making method using the novel score function is proposed and compared with the existing multiple attribute decision making method using the existing function. The experimental results show that the method using the new score function can still obtain the correct decision results. Finally, it is proved that the proposed score function has higher computational accuracy than the existing function, and can maintain its effectiveness when applied to multiple attribute decision making problems.