The trisecting, acting, and outcome comprise the three core parts of the TAO model for three-way decisions. “Acting” formulates suitable movement strategies for each region based on the results of “trisecting” so that objects can move from unfavorable regions toward favorable regions. Regarding the study of “acting”, there are mainly two movement strategies: object-based movement strategy and region-based movement strategy. However, these two methods only make movement strategies based on existing objects, so they may not be applicable to unknown new objects, namely, the generalization ability is not strong. Therefore, to develop a more generalized movement strategy that can be applied to both existing objects and new objects, we proposed a movement strategy based on interval-value. First, using interval-value to represent each flexible attribute value, a formal representation of the interval-value based movement strategy is proposed. Second, to generate movement rules from interval-value based movement strategies, we present four methods to generate four types of specific movement rules: optimistic movement rules, pessimistic movement rules, the highest frequency of global interval-value movement rules, and the highest frequency of local interval-value movement rules. Finally, the proposed model is efficient, as demonstrated by experimental data.