Influence of Oil Tanker Subdivision on Probabilistic Oil Outflow Following Side Damage

We live in an era of growing environmental concern. In the future no activity will be considered acceptable unless it is reasonably safe for the environment. Since the late 60's, when growing demand for oil transportation by sea introduced the concept of VLCCs, oil tanker accidents received wide coverage by the media. In 1967 the stranding of the Torrey Canyon highlighted the immense threat of massive pollution that could result from accidental discharge of oil in the event of stranding, collision, etc. The Amoco Cadiz in 1978 and the Exxon Valdez in 1989 are some of the accidents which have since greatly added to the public's concern about that threat. Measures and tentative solutions to the technical aspect of accidental oil pollution from tankers have been put forward from time to time. The U.S. Oil Pollution Act 1990 (O.P.A.'90) and amendment 13F to Annex I of MARPOL 73/78 are the latest attempts to curb accidental oil spillage, by introducing new requirements for the construction of oil tankers. The requirements prescribe double hull construction as the reference in terms of pollution prevention, with MARPOL leaving the door open to novel design solutions that provide "equivalent protection" against oil pollution. The controversy that preceded and followed, particularly in the case of the unilateral enforcement of O.P.A.'90, prompted new research on the subject of oil tanker design assessment. A probabilistic method of assessment, already applied to passenger and cargo ship damage stability, was proposed for the comparison of oil outflow behaviour in collision and grounding accidents. Unfortunately, most of the literature concentrates on the debate over the equivalence of double-hull and mid-deck designs without worrying about improving one concept or the other. The probabilistic assessment is used as a "black-box" tool, fed with an input and yielding an output, without any explicit links between design parameters and oil outflow results. Furthermore, the choice of simplifying assumptions in the application of the probabilistic assessment, obliterates even more these links and limits the usefulness of the results. The present work starts with a short review and critique of the way the probabilistic method was used in oil outflow studies, highlighting in the process their important features and their weaknesses. An enhanced method is proposed, which has the merit of taking into account the random nature of the damage characteristics in the vertical dimension of the ship. Only side damage cases are addressed but the proposed method allows for bottom damage cases to be treated in exactly the same manner, thus providing a unified framework for the complete assessment of a tanker design. With the help of the enhanced probabilistic method, a series of systematic variations of the subdivision arrangement in three tanker design concepts is assessed. An attempt is made to chart the relationship between the choice of design parameter values and the environmental performance of the resulting tanker design. The results of the analysis are validated, when possible, with results from other oil outflow studies, while an explanation is proposed for results showing discrepancies. The analysis results are also used to draw basic guidelines for improving the subdivision arrangement of typical single-skin, double-hull and mid-deck tanker designs. The work shows in conclusion that in each of these concepts the subdivision arrangement can be optimised for a given level of environmental performance and that the probabilistic method of assessment is the way forward in the evolution from purely prescriptive construction regulations to goal-orientated design