Evaluation of the Physical Risk of Ship Grounding

Evaluation of the Physical Risk of Ship Grounding

and

Submitted to the Department of Ocean Engineering

on January 17, 1996 in Partial Fulfillment of the Requirements for the

Degree of Master of Science in Ocean Systems Management

An investigation of the potential to formulate a model for the prediction of the physical risk of navigational groundings during transits of vessels into and out of ports was undertaken using historical data. A database that included information on the circumstances surrounding groundings in five U.S. ports (Boston, Houston/Galveston, New York/New Jersey, Tampa, and San Francisco) for the period between 1981 and 1995, was assembled and analyzed. Time series of the underlying grounding rates for the five ports of study were developed using U.S. Coast Guard casualty data and U.S. Army Corps of Engineers transit data. The grounding risk contributing potential of five parameters was examined to determine their usefulness as explanatory variables in the development of the physical risk model. These parameters are: wind speed, visibility, human factor, depth uncertainty and ship characteristics.

Even though the data acquired for the purpose was neither perfect nor complete, it was possible to extract useful information on historical groundings. The results suggest that wind speed and visibility conditions appear to be useful parameters in explaining the occurrence of groundings. In general, it was found that historical groundings are associated with higher average wind speed and lower average visibility than safe transits. Human factors are considered as one of the most risk contributing factors, responsible for 80% of maritime casualties. The information available were not adequate for a quantitative assessment of their grounding risk contributing potential. Additional information must be obtained. Uncertainty in hydrographic surveys, on which nautical charts are based, does not appear to contribute significantly to groundings in the particular set of study ports. Ship characteristics appear to provide another set of useful explanatory parameters. In general, barge trains were more likely to ground than other ships.

An approach is proposed towards the development of the mathematical model for the prediction of the probability of grounding (physical risk). The potential for mutual exploitation between the physical risk model and the International Safety Management Code is briefly discussed. Some conclusions and suggestions for the subsequent effort are presented.

Thesis Co-Supervisors:	Nicholas M. Patrikalakis	Hauke L. Kite-Powell
Title:	Professor of Ocean Engineering	Research Specialist (WHOI)