The goal of this study is to provide a model for maritime accidents which focuses on human factor related safety concerns and permits the economic evaluation of countermeasures. In order to halt the trend of maritime accidents it is necessary to identify the human related factors that are associated with them and address a method of analysis to isolate these factors and evaluate their economic impact. The United States Coast Guard (USCG) investigates and records data on accidents for vessels in U.S. waters and U.S. flag vessels all over the world. Other organizations which collect accident data are Lloyds of London, the P&I Club and some marine insurance companies. In order to understand the magnitude of human factor related accidents the USCG casualty database CASMAIN, which provides consistently gathered data for a period of more than ten years from (1981 to 1991), is analyzed. This database contains close to 66,133 records on 42,367 accidents. Accident analysis in the past was restricted to statistical reporting of the nature of casualty and its primary causes. Analysis rarely add ressed such issues as contributing causes and the variety of human factors causes and still further those contributing factors which led to failure of the human being from performing at an that more than 80% of the accidents were a result, whether direct or indirect, of human error. Such issues as workload, fatigue and stress in the transportation arena are being considered only now and other factors such as the effects of circadian rhythms and alertness switches are still in the realm of research. To better understand the human factors causes and their rationale a functional model of critical tasks and resources to carry them out is outlined. The human factors causes of accidents are also detrimental to the personnel resources required for a safe ship operation. If the risks posed by these factors can be assessed we would be in a better position to ensure that the tasks are performed properly and thus realize a safer and more efficient maritime system. Past historical accident data from the USCG CASMAI N database was then developed and organized to provide numerical estimates of human error frequencies for use in the proposed model. The resulting model would permit assessing costs and alternatives and result in a methodology of better addressing marine transportation safety with an emphasis on its prime creator, user and benefiter:-The Human. / Master of Science