The Alloy tool-set has been widely used in software design and analysis. However, Alloy Analyzer that automates a bounded exhaustive analysis relies on expensive SAT solving. Recent studies have shown that users often perform consecutive analysis with slightly different models, which opens up the possibility of using incremental analysis for speeding up the interaction. In this paper, we aim to develop an incremental analysis to reduce the number SAT invocations, thus speeding up the evolving Alloy analysis. Firstly, we proposed Regression Command Selection based on dependency analysis, similar to those used by Regression Test Selection, to select the commands only affected by users’ changes. Secondly, we applied Solution Reuse on the selected commands to further reduce the actual command execution (which invokes SAT solving). Our experimental results show that our approach significantly outperforms Alloy Analyzer in the analysis of evolving Alloy models.