As its function to connect people between islands at more competitive prices, sea transportation become one of the most vital facilities for some archipelagic countries, like Indonesia. This is the reason why the number of units of sea transportation in Indonesia keeps on increasing every year. In this industry, maintenance is one of the most important activities in shipping industry as it can determine the eligibility of the ship. However, this activity is not offset by the capacity of the national shipyard, makes the estimation of ship maintenance duration as a very important. This research uses one of data mining method, namely CART (Classification and Regression Tree) to estimate the duration of maintenance that is limited to dock works or which is known as dry docking. By using the volume of dock works as an input to estimate the duration, there are 4 classes of dry docking duration obtained with the different linear model and job criteria for each class. These linear modelscan then be used to estimate the duration of dry docking based on its job criteria.