The objective of this study is to model land development in the 17 sub-districts in Phuket province of Thailand from 2000 to 2009. Logistic regression was used to monitor changes in land-use over this period and predict future changes. The ROC curve was used to measure the performance of the model. Land-use from a previous survey in 2000 and sub-district identity were included as determinants. The area of developed land increased by 4557 ha over the study period. In 2000, agricultural land was more likely to become developed in 2009 and developed land was more likely to remain developed land in 2009. Land development occurred mostly in Chalong and Talat Nua sub-districts. The area under the ROC curve was 0.83, indicating a reasonably good fit of the model. Prediction of such changes may be used to provide useful information for decision makers and planners.