Numerical Method for Cost-Weight Optimization of Stringer-Skin Panels

Abstract

The presented work addresses the need to integrate cost into the early product definition process as an engineering parameter. The methodology developed is generic and fundamental in developing causal predictions of manufacturing cost that are driven by the design parameters which give rise to the main elements of that cost relative to process capability. The manufacturing cost modelling is original and relies on a genetic-causal method of 1) classifying the generic cost elements that are linked to particular genetic identifiers relating to materials, processes and form; 2) developing causal parametric relations that link those genetic identifiers to a parent manufacturing cost. The application studied is a fuselage panel that is typical to commercial transport regional jets. Consequently, a semi-empirical numerical analysis using ESDU reference data was coupled to model the structural integrity of thin-walled structures with regard to material failure and buckling: skin, stringer, flexural and inter-rivet. The optimisation process focuses on Direct Operating Cost (DOC) as a function of acquisition cost and fuel burn. It was found that the ratio of acquisition cost to fuel burn was typically 4:3 and that there was a 10% improvement on the DOC for the minimal DOC condition over the minimal weight condition; due to the manufacturing cost saving from having a reduced number of larger- area stringers and a slightly thicker skin than preferred by the minimal weight condition. It is also noteworthy that the minimal manufacturing cost condition was slightly better than the minimal weight condition, which highlights the key finding: the traditional minimal weight condition is a dated and sub-optimal approach to airframe structural design.