The productivity in ship hull construction largely depends on the block division. In order to achieve an efficient block division plan, an optimization system is developed by integrating a product model module with a genetic algorithm module, with the help of a graph model module which acts as an intermediate agent. As for the generation of block division plans, ‘Nodal Cut Set Method’ is proposed in which various plans are generated based on the chromosome, using graph theory and logical operations. Plans are evaluated using the product model data. The validity of the proposed system is shown by executing some optimizations.
Block construction method is adopted in modern shipyards for better efficiency and productivity of hull construction. The block construction process consists of many stages, for example sub assembly, assembly, grant assembly and dock stage. The efficiency of production activities in these stages largely depends on block division plan. The decision on block division depends on many factors like production facilities, ship structural arrangement and outfitting, labor skills, quality related factors like weld deformation that may occur during block assembly and so on. All these factors should be considered simultaneously. Therefore the generation of proper block division plans is carried out by expert designers with long term experience in shipbuilding. But such expert designers are very few in numbers and it is required to develop some systems to support block division.
In order to achieve an optimum block division plan, it is required to carry out following two steps:
Generation of various block division plans.
Evaluation of block division plans.
There are so many factors which have to be considered in the evaluation of block division plan. Interface weld between blocks, size and weight of blocks, quality related factors like weld deformation which may be occurred during block assembly and so on are considered as some of the typical evaluation factors, Kumar et al (2005).