现实生活中遇到的许多问题都具有不确定性，使得在对系统进行决策评估时，指标值难以精确化。在此情形下，人们常常对指标值给出一个区间，到目前为止， 尚未有人研究区间数灰靶决策。首先定义了区间数、m维区间数的距离及其距离性质，并证明了当区间数为实数时，区间数距离就是实数距离的推广；提出了区间数规范化方法，在此基础上, 建立了基于区间数的灰靶决策模型，从而把灰靶决策模型由实数序列拓展到区间数序列，使灰靶决策理论得到发展，同时为扩大灰靶决策的应用领域提供了理论根据。最后以实例验证了该模型的有效性与实用性。

There exists uncertainty in most problems in man´s practices and the uncertainty hinders decisionmakers from defining accurately indices values when they evaluate the system. For this problem, present decision-makers often define an interval number to denote this index value, but none has studied the interval number decision model in grey target decision. In this paper, the interval number, the distance of m dimension interval number and its characteristics are defined firstly, and the fact that the distance of interval number generalizes the distance of real number is proved, then, the method to normalize the interval number is developed. Based on the above, the multi-attribute decision model of grey target based on interval number is derived, which extends the grey target decision model from real number sequence to interval number sequence and richens the grey theories in grey target decision. Last, the validity and practicability are illustrated with an example.