This paper demonstrates the implementation of Logical Analysis of Data (LAD) methodology in the field of prognostics in Condition Based Maintenance (CBM). In this paper the LAD classification methodology, based on Sensitive Discriminating and Equipartitioning methods for data binarization, Mixed Integer Linear Programming (MILP) and Hybrid Greedy methods for pattern generation, is used. Using the generated patterns, two methods of calculating the survival function are introduced. The methodology is applied on Prognostics and Health Management Challenge dataset, which is a condition monitoring dataset provided by NASA Ames Prognostics Data Repository. The results obtained by using LAD methodology, are compared with that obtained by using the Proportional Hazards Model (PHM).