Can a “physics based” approach capture safety of a ship’s operation? (Part 1 of 2)

A ship is a good example of a socio-technical system where its behavior cannot be understood without a model description that covers both social and technical aspects.

Engineering approaches to improve safety are developed under the assumption that there is a link between the technical solutions implemented and the safety level during operation. There is also a link between how the ship is operated and the safety level during operation. However, this second link is often hidden to engineers because traditional engineering approaches and tools typically do not describe how risk decisions taken on-board affect safety. As discussed within the intact stability community operational guidance or limitations are an important aspect of a holistic safety approach for intact stability. However, such operational measures also introduce new uncertainties.

The work in regard to the second generation intact stability criteria is based on three alternative assessment procedures: Level 1 vulnerability assessment, Level 2 vulnerability assessment; and Direct stability assessment. Compliance with Level 1, 2 or the Direct stability assessment fulfils the requirements of the intact stability criteria. It is also proposed that alternatively, ship-specific operational limitations or operational guidance can be developed for conditions failing to fulfil the criteria.

The work within the second generation in-tact stability criteria so far has focused on “physic-based analysis” and “passive” safety measures described by the level 1 and 2 assessments. However, the operational environment and the operation itself is not static, this may lead to that safe passive design measures need to be far reaching in order to exclude unsafe operations.

When investigating 36 intact stability incidents they add up to more than 408 fatalities. The median number of persons on-board is 14 and the median number of fatalities per accident is 3 (13 and 6 respectively if the ship capsized or sunk). In all but 11 cases the ship was lost as a result of the accident. The incidents can all most often be contributed to a combination of causes and for many of the accidents the cause is uncertain.

 

The MV Finnbirch before she sank in 2006. Was it due to design or operation? Photo: The Swedish Maritime Administration / Helicopter Lifeguard 997, 2006.

Many of the incidents, approximately 20 out of 36, are cases were the operational condition and ship state was not according to design. For example, vessels that are over loaded and/or operated in heavy weather with hatches open potentially in combination with forces from fishing gear. Cargo shift is also common. These conditions lead to a poor recoverability after large heel angles. Therefore, in the investigated incidents there are many aspects that a physics based approach that only consider the operational conditions to a limited number of standard situations will not capture.

For example, does the difference in potential control over the ship’s loading condition produce different conditions for safety work, different reliability of the passive safety designed into the craft, and different reliability as well as different need for operational safety measures. However, knowledge on safe operations, based on knowledge about the vessel’s limitations and weaknesses (edge awareness) could increase the reliability of the crew decisions taken on-board in relation to intact stability especially for ships and vessels that relatively often operate beyond the operational conditions defined during the design. Therefore, operational safety measures can be an effective approach to reach acceptable levels of safety, especially for operations with large uncertainties.

Link to more info on the original article:

Liwång, H., and Rosén, A., 2018,"A framework for investigating the potential for operational measures inrelation to intact stability", 13th International Conference on theStability of Ships and Ocean Vehicles (STAB2018), Kobe, pp. 488-499.