Page 295 - Practical Design Ships and Floating Structures

270

of naval architects working concurrently on different aspects of the project under development, co-
ordination being offered by discussions, which usually result in implementation of design adjustments.
It is argued, that it is essential to follow a methodology for effective design co-ordination, both for
issues of concurrency and decision-making.
With this general frame in mind, it is argued that a “Design for Safety” paradigm could provide the
vehicle for safety to constitute a focal issue for an improved ship design process, Vassalos et al. (2000).
This paper deals with a specific issue within this vast subject, namely the provision of a procedure for
the efficient integration of damage survivability considerations within the ship design process.

2 OPTIMISATION FOR SAFETY IN SHIP DESIGN
Design optimisation techniques have been developed with the aim to formalise the search for optimal
solutions, to identify possible solutions beyond prescnt state-of-the-art and effectively to reduce
development time and associated costs. The general approach is to structure an optimisation function
(and a set of design variables), which is considered against a set of optimisation criteria and design
constraints and to produce a solution based on mathematical programming techniques. Obviously, due
to the fact that the problem at hand is reduced to a function or a set of variables, there is a need for the
mathematical formulation to represent the problem as accurately as possible in order to eliminate the
chance of solving a fraction of the original problem. With rapidly advancing computer technology,
computers are becoming more powerful, and correspondingly, there is an increase in the size and
complexity of the problems being tackled, as well as the fact that new optimisation techniques can be
developed and put into practice. Optimisation methods, coupled with modem tools of computer-aided
design are also being used to enhance the creative process of conceptual and detailed design of systems.

Within the broad field of ship design, a large variety of design optimisation techniques have been
successfully applied to solve either the general design problem (for example, determination of main
dimensions and characteristics) or to provide optimal solutions to specific areas of ship design (for
example, lines development or resistance). In the specific area of the assessment of safety in ship
design there is a rather limited amount of work published. In the following, two areas that the
proposed procedure is deemed appropriate are briefly described.

2.1 Probabiktic Rules-Based Optimal Design
The objective is the development of a procedure for optimal compartmentation and internal arrangements
of passenger Ro-Ro ships adopting the probabilistic concept of damage stability. It proves necessary to
evaluate first the robustness of a probabilistic rules-based design procedure in a range of scenarios and its
sensitivity to the main design parameters involved in the assessment process, which will lead to the
definition of suitable constraints. After setting up a framework of local and global optimisation
techniques regarding ship compartmentation for enhanced damage stability characteristics, a formal
integration within an overall computer-aided design procedure can be performed.
2.2 Quantifutive Risk Assessment (QRA) -Based Optimal Design

An appropriately structured QRA framework provides the means for a unified measure of safety,
represents an absolute measure for the risk levels of a vessel and quantifies the effect of risk reduction
measures. To integrate such a framework within the ship design process efficiently, it is necessary to
consider utilisation of appropriate design optimisation tools that perform the trade-offs and take into
account identified safety-critical design features and established criteria. This way, an effective and
efficient balance between risk reduction measures and cost benefits can be achieved. As a
consequence of the optimisation procedure, the effects of optimal design features will be identified and