Achieving a Predictive Design for Manufacture Capability in Composites by Integrating Manufacturing Knowledge and Design Intent

The use of advanced composites in commercial aircraft structures has significantly increased in recent years through products such as the Airbus A350XWB, where they make up 52% by weight of the structure. But the transition over from metals has actually been slower than anticipated, despite the advanced composites promise of offering lower weight components that are capable of exhibiting high strength-to-weight ratios & high stiffness. This slow uptake is primarily due to the high cost of manufacturing; and is now more of a concern, as when designing a new aircraft the mechanical properties are not the only aspect taken into consideration. Composites must be cost-competitive.

Historically, civil aircraft design and manufacturing was largely conducted in-house and relied heavily on manual intervention, especially during assembly. This reliance on manual operations came as a result of long development times and ongoing aircraft design iterations, which together rendered the mass production of aircraft costly and infeasible. In the past decade a transformation has occurred as aircraft manufacturers see higher sales and uptake; and are increasingly subcontracting parts and systems to suppliers. Boeing for example increased their outsourcing from 35-50% for the 737 program to 70% for the 787 program. Whilst this has provided cost saving opportunities for the Original Equipment Manufacturer (OEM), it adds pressure to an already restricted supply chain to deliver parts that are not only made to specification but governed by shortening times and cost reductions. It has also demonstrated supply chain inefficiencies in the global industry, and that the need for cost-effective manufacturing methods tailored for smaller and medium sized suppliers has become more evident.

For these companies, the cost of rearranging the work space and of purchasing new equipment is quite restrictive, especially if manufacturing small batches of components, as they may not reach their break-even point. In order to meet the projected growth & demand for the aerospace industry, and guard against projected skills shortages, manufacturing techniques need to be developed to allow for greater efficiency, affordability, and greater consistency in quality. The ultimate goal is to produce high quality components right first time, consisting of the proper dimensions and performance properties that are not only reproducible, but economically viable.

Composites usage is particularly dominant in secondary structures or sandwich panels. The complex geometries associated with these restrict the use of automation and so hand layup dominates the manufacturing process. It involves forming a pre-impregnated cloth over a geometry into as near-net shape as possible, using shear as the main in-plane deformation mode. Difficulty in manufacture arises from geometrical clashes (imposed by structural and aerodynamic performance of the aerofoil, resulting in tight dimensional tolerances); audit trails (imposed by the OEM), and; their low-cost development (company imposed). These coalesce such that the majority of the total manufacturing cost for aircraft composite components resides in secondary structures, dependent on inefficient design and manufacturing processes based on tacit skills and understanding.

To break this vicious cycle for price-critical parts, either low-cost manufacturing methods or designs for manufacturability need to be implemented. This research targets the latter, developing a new toolset capable of informing for intelligent design processing that considers manufacturing capabilities earlier, and delivering the design intent to manufacturing as functional unambiguous workflow instructions enabling right first time yields. A new process towards composites DfM will be developed through this research, and in enabling gathered information to be exploited in simple formats, a user-based knowledge system will be achieved.

The approach taken in this research - of closing the loop in composites DfM by better integration of manufacturing knowledge and design intent, and in delivering a virtual tool capable of predictive capacity in design and better informing of manufacture - offers manufacturers of secondary structure composite products the potential to achieve their ultimate goal. That of producing high quality components right first time, consisting of the proper dimensions and performance properties that are not only reproducible, but low in cost; whilst still using the hand layup process. Despite this research targeting well-founded installed techniques and capabilities (if inefficient), this approach is innovative and positively disruptive in two ways:

1. It will offer a new understanding in the relationship of design complexity and manufacturing capability through the formulation of a capability index, based on a geometry's surface area the effect of coupling features together. No such index is known to exist in composites hand layup & DfM; though examples are available for other materials and products (such as the Lucas method in DfA). Combining this index with observed and measured key variables of the laminator, the material, and mould interactions will allow for performance predictions to be made
2. By developing a virtual tool better able to describe how and why a laminator forms a material to a surface, a predictive capacity for composites design and manufacture will be delivered, integrating an exploitable Knowledge Base System (KBS). In combination they will deliver a unique service, capable of streamlining conceptual design and offering a simple yet powerful suitability analysis, whilst automatically delivering manufacturing workflow. There is no toolset able to achieve either of these tasks at the detail envisioned, but commercial tools for selecting processes do exist at a top level (such as that provided by Granta Design)

The PI aims to build a new level of detailed understanding in composites manufacture, so that the next steps in optimising layup can be informed (and guard against the risks outlined by Bainbridge in 1983 that are still relevant today). It is anticipated that the following groups will benefit from the proposed research:

a. Aerospace Industry (knowledge): OEMs will benefit by parts of improved reliability & quality consistently being achieved in the supply chain, of reduced incidences of concessions or defective/scrapped parts, of build & cost timescales being met, shorter lead times in design and manufacturing, and better understanding in the impact of future product design complexity. Increased OEM/tier collaboration and further enhancing of SC21 is also feasible
b. UK manufacturing (people): the research will position the UK as a global leader in composites DfM. It will underpin decisions for UK companies competing for manufacturing work, and could enable the on-shoring of parts manufacture previously lost to lower cost economies
c. Skills and training/employment (people & society): the research will lead to enhanced capabilities in composites design and manufacturing, leading to increased skills and opening up opportunities for job creation. A spin-out or start-up as a consultancy or engineering company to exploit the research is also feasible
d. Simulation providers (economy): the research could inspire simulation companies to develop new portfolio items for the composites market, and collaborate, to maximise the research outputs. Improved simulation performance will also reduce waste streams
e. Exploitation in education (society): the research will be communicated to undergraduate & taught postgraduate students at the earliest opportunity, combining with traditional materials to refresh course content and demonstrate new lines of enquiry. It will be used in undergraduate & postgraduate research projects; and demonstrated through the planned website plus University Open Days