Abstarct: As one of the main processes of evaluation, exploration, and development of oil and gas reservoirs, the drilling industry has always faced many complications and challenges. Non-productive times are one of the most important issues that due to the wide geographical scope of the drilling operation and its nature, due to unexpected events, it causes a delay in the well construction process. Non-productive times in drilling operations can have different reasons and origins and impose additional costs on this industry. Among the reasons and events that lead to the creation of non-productive times are the events and phenomena in which the role of geomechanical factors is evident in their occurrence. Therefore, identifying, predicting and controlling these events is very important in order to reduce costs, increase productivity and create added value. Hence, the implementation of the preventive analysis, analysis and evaluation process, along with the use of the latest tools and techniques of the day, can reduce the operational risks in a favorable way. Therefore, in this research, with the aim of evaluating the impact of geomechanical factors on non-productive times in the drilling process, in the first step, geomechanical events were identified and extracted through daily drilling reports. Therefore, in this research, with the aim of evaluating the impact of geomechanical factors on non-productive times in the drilling process, after identifying the events of geomechanical origin in wells 83 and 96 of Mansouri field through daily drilling reports, in order to create a proper view of the mechanical condition of the wells, a one-dimensional model Geomechanics were prepared. Then the data related to each event, including geomechanical, petrophysical and drilling data, were compiled in a data bank for each. Further, by using artificial intelligence tools and neural network design baxsed on algorithms for diagnosing and predicting drilling problems, a model was obtained that can optimally identify and predict geomechanical events with more than 80% accuracy and more than 90% detectability. This is very important in adopting preventive and control measures and thus reducing non-productive times.