An approach to designing the systems design process of naval ships

The design of complex Engineered to Order (ETO) products can span many months, if not years and must manage risks arising both from uncertainty in behavioural predictions of the physical system realisation and reliability of the evolving design process. The aim is to develop a novel approach to optimise engineering design processes for ETO products, taking into account combined uncertainty predictions for both product and process behaviours. A generic approach will be developed to consider how these two areas of concern affect overall accuracy in predicting resulting design performance. This will be done through the use of deep learning analytics on large industrial datasets, to trace the introduction and propagation of uncertainty from multiple sources and inform how to optimise the employment of engineering analysis within the design process. Specific engineered cases will be used to evaluate the approach and its feasibility.