The software maintenance work is presently much more expensive than what it should be and takes more time to implement. It consumes a significant amount of resources in software development. Object-oriented technique has become a defacto standard in the software development community. Therefore, it is necessary to design object based software system keeping maintainability in mind. In this work data clustering technique's K-means clustering is used to identify maintainable classes at design level. We also show the significant level of goodness of clusters using chi-square test.