A proposal of software maintainability model using code smell measurement

Maintenance is a very important phase in a software's life cycle. More than half of the effort and cost of software is in the maintenance phase. Code smell is one of the problems that need to be dealt with in this phase. Code smell can be caused by bad design and implementation choice. In the end, it has the potential to degrade code maintainability. This paper proposes a new software maintainability model using measurement of code smell. Code smell has been used to indicate problems in software code which could be the results of bad design and may lead to bad maintenance. Our proposed model is then validated against the density of design patterns in 6 open source applications. The design patterns have been known to give positive impact on maintainability. Our result proves a strong relationship with correlation value of 0.51 between our model to the design patterns density. It shows that our model can become an alternative to measure a maintainability quality of an application.