Navigation safety control system development through navigator action prediction by data mining means

Taking into account current trends in the development of ergatic maritime transport systems, factors navigator’s influence on vessel control processes were determined. Within framework research hypothesis, to improve navigation safety, it is necessary apply predictive data mining models and automated control. The paper proposes a diagram system model for identifying navigator “human factor” during navigation. based principles decision trees, prediction by means applied, taking identifiers occurrence critical situation. Based results, method optimal situations was developed, which triggered at nodes tree, reduces likelihood impact proposed approaches tested laboratory “Development support systems”. use Navi Trainer 5,000 simulator (Wärtsilä Corporation, Finland) simulation safety have confirmed its effectiveness. As result testing, determined that activation allowed reducing 18–54 %. In 11 % cases, switched automatic mode and, as result, reduced risk emergencies. tools made possible neutralize negative reduce average maneuvering time 23