Rapid Manufacturing for Premium Aircraft Seating T. Steward1,2, R. Bibb1, H. Millward1 and S. Tong2 1. The National Centre for Product Design & Development Research (PDR), University of Wales Institute Cardiff (UWIC), Cardiff CF5 2YB, UK. 2. Contour Premium Aircraft Seating (CPAS), Cwmbran NP44 3HQ, UK. [email protected] [email protected] [email protected] [email protected]

Abstract Premium aircraft seating designs are often bespoke for every customer (the airline). Product differentiation, innovation and prestige are very important to the airlines and the services they offer, and they are willing to pay a high premium for this. Within an airline’s premium aircraft cabin, there are generally only a few seating units between which there is a large amount of variation due to their unique position in the tapering aircraft cabin. All of these configurations need to be designed, manufactured and delivered within very short timescales. Therefore, there needs to be a great deal of confidence that the product meets the demanding requirements of the airlines, the aircraft manufacturer, the certification testing that dictate the product’s entry into service as well as ensuring economic production. The complexity of seating design and the variation between the different configurations can dictate the manufacture of hundreds of types of low volume components. For these production volumes, the economies of manufacturing suggest that reducing tooling requirements can notably reduce part cost. This high value, low volume production is a typical application for the use of the Rapid Manufacturing (RM) technologies. The potential to reduce lead times and eliminate tooling whilst producing high tolerance parts could have significant commercial benefits for the business. This paper uses case-study material to explore the implementation of RM technologies for the manufacture of premium aircraft seating products, going on to look at the implications for further use of Rapid Prototyping (RP) in the new product development process. It describes the challenges facing the more widespread use of these technologies in this sector of the aerospace industry and the efforts that Contour Premium Aircraft Seating (CPAS) and the RM industry are undertaking in order to address them. Further recommendations on what might be done to increase the implementation on RM and RP in this industry are also made.

Keywords Aircraft Seating, Rapid Prototyping, Rapid Manufacturing, New Product Development

INTRODUCTION Contour Premium Aircraft Seating (CPAS) CPAS is part of the Premium Aircraft Interiors (PAIG) group that compromises a number of companies involved in the design and manufacture of aircraft interior components such as seating, furniture, galleys and lavatories. CPAS deals in aircraft seating, particularly for the premium markets. These can be generally grouped into ‘First Class’ and ‘Premium Business Class’ markets for which CPAS currently holds 50% and 20% of the market share respectively. These provided CPAS with an £80 million turnover in 2006. The ‘Super First Class’ market encompasses fully enclosed and highly prestigious seating suites including a fully lie flat seat/bed and the associated furniture and features. The ‘Premium Business Class’ market encompasses similar lie flat seat/bed functionality and quality but on a smaller scale. CPAS’ past and current customers include British Airways, Virgin Atlantic, Lufthansa, Cathay Pacific and Air France. The design organisation in CPAS is split between two sites in Camberley, Surrey and Cwmbran, South Wales, where the manufacturing and assembly facilities are also based. The manufacturing facility in Cwmbran provides most of the sheet metal and vacuum forming components and the majority of machined components for the products, whilst composite components are outsourced. Rapid Manufacturing Rapid Prototyping (RP) refers to a group of technologies that build components layer by layer directly from CAD models. Rapid Manufacturing (RM) describes the process of using the RP processes in the direct manufacture of components and is the main focus of this paper, though Batch Manufacturing processes that use the RP technologies indirectly for the inexpensive tooling of medium production runs is also addressed. The advantages of RM include reduced tooling costs, reduced lead times, increased design flexibility and geometric freedom allowing economic custom manufacturing where part geometry is not limited by efficient use of billets or manufacturing processes. This provides opportunities to reduce part count and weight whilst improving design integrity through avoiding the inevitable design compromises in order to facilitate manufacture and assembly. These Design-For-Manufacture (DFM) considerations are described in more detail by Hague et al. [1] and can have a great influence on part design and the effort involved in the later stages of the design process. The high value and low volume nature of manufacture of premium aircraft seating components is a typical application for RM and the opportunities associated with RM are particularly advantageous and could potentially transform the approach to product design in this industry. Rapid Prototyping Tools, such as prototyping, encourage design iterations and development testing as well as effective communication and collaboration during the early fluid stages of the design process. They can be vital in the optimisation of the design. Through quick turnaround, they reduce the risks associated with new designs in meeting all its requirements upon its initial release and avoid significant issues being identified during the manufacture, assembly and early service life of the product that could necessitate design changes.

DESIGN AND MANUFACTURING REQUIREMENTS Customer Requirements Premium aircraft seating designs are often bespoke for every customer, as perceived product differentiation is very important and the customers are willing to pay a very high premium for ‘First Class’ products. This is due to the importance of perceived product differentiation, the design’s conformance to the airline’s brand and service concepts. The majority of ‘First Class’ products are designed and built to a specific customer brief that often includes concept work produced by external design consultancies. The customers are consequently involved in the decision-making during the design process and are particularly interested in the final signoff of the design and build. They also impose stringent requirements for aesthetics, reliability and ease of maintenance. The overall development time is already constrained by the market to approximately nine months for derivative products and up to 18 months for custom products. Weight is becoming an increasingly important consideration in the design of the seats. They have to be carried for the life of the aircraft, contributing to fuel consumption and reducing payload, a significant issue for the airlines’ business. Strength requirements of suite components are a significant contributor to the total suite weight, but for some features, such as tables, a certain amount of weight has to be maintained for aesthetic requirements and perceived quality. Designing for Aircraft Within a business or first class cabin on an aircraft there are usually only a between four and eight seating units. Due to the non-uniform geometry of the cabin, the layout of the seats, other furniture and utilities within the cabin as well as the variation between the different aircraft models in an airline’s fleet, there can be significant variation between seating units requiring a number of configurations to be designed and built (Figure 1a). The features expected of a ‘First Class’ suite are becoming increasingly elaborate with additional furniture, entertainment systems and the associated wiring, control boxes, actuation systems and mechanisms leading to part counts approaching 5000. Furthermore, the seats need to successfully integrate into the aircraft and comply with the aircraft manufacturer’s specifications. Often concepts supplied by external consultancies fail to include provision for all the features that are required for safety and integration requirements dictated by the certification testing and aircraft manufacturer. Late additions of these features reduce the scope of integrating them into the original styling of the product and can compromise design intent. Certification and Aircraft Manufacturer Requirements Certification requirements are dictated by the aviation authorities, such as the Federal Aviation Administration (FAA) [2], and constitute a series of tests and criteria that the seat needs to pass before it can be entered into service. Every new non-metallic material and combination of materials needs to be tested for flammability performance including, heat release, smoke density and smoke toxicity. These need to be carried out before any final engineering Bill Of Materials (BOM) is released. There are also a series of static and dynamic tests that are carried out once the design is complete and released for production, which represent abuse loads and crash conditions. Should there be any failures at this point, changes are costly and difficult to implement without.

Often initial concepts fail to include provision for safety and certification needs as well as the integration requirements dictated by the airframe manufacturer, such as Airbus. The late addition of these essential features as well as other changes in response to testing results can on average have a count of twice the number of parts in the product by the time the design is truly frozen. Furthermore, late changes can reduce the scope of integrating any visual alternations into the original product styling, compromising the design intent through having to make changes within a mature configuration and can affect the initial delivery of the product.

RAPID MANUFACTURING IN PREMIUM AIRCRAFT SEATING Suite Commodity Breakdown In a typical ‘First Class’ suite (shown in Figure 1b), the main features of the suite can be broken down into the seat, the furniture and the plinth. Furniture includes items such as side fixtures, wardrobes and suite dividers. The furniture and the seat are mounted on the plinth that provides the structural interface between the suite and the aircraft. Due to the high number of suite configurations, many components that are produced in low volumes or in a number of variations as there is a limit to the commonality that be carried over. With less than 10 suites in an aircraft and perhaps 20–50 aircraft across an airline’s fleet, maximum production quantities are in the mid-hundreds.

Figure 1 – (a) Typical First Cabin Layout; (b) Typical First Class Suite (© CPAS)

.

Figure 2 – Suite Cost Breakdown

Figure 3 – Tooling Cost Breakdown

Figures 2 and 3 show the relative material cost and tooling cost for a typical ‘Super First Class’ suite. Composites constitute over half of the suite’s cost and the majority of tooling costs. Metalwork, including Aluminium, Stainless Steel and sheet metal components, constitutes 14% of the total suite cost and over half of the seat cost. Within a typical seat, metalwork constitutes the majority of cost and weight, specifically for structural aluminium alloy or steel components. The remainder of the structure, particularly in the furniture and plinth, consists of composite materials, usually trimmed with textiles, vacuum formed plastics and cosmetic features. The main methods of manufacture are conventional machining, sheet metal forming or vacuum forming, most of which have minimal tooling costs. Table 1 shows the breakdown of the commodities into the application and the specific manufacturing process along with a suggestion for a direct RM (or Batch Manufacturing) alternative. Commodity Composites Metalwork

Plastics

Application Structural furniture components Primary structural seat components Secondary structural seat components Trim Trim

Existing Process Composite moulding

Trim

Injection moulding

Rapid Manufacturing (Rapid Tooling)

Machining/extrusion Sheet metal Machining/extrusion Vacuum forming

Direct Selective Sintering

Laser

Selective Laser Melting Stereolithography Selective Laser Sintering Fused Deposition Modelling (Vacuum Casting)

Table 1 - Manufacturing Applications Typical composite components in aircraft seating are relatively large and constitute over half of the suite cost and the majority of the tooling cost. Furthermore, these tools require long lead times and due to the low production volumes, process capabilities are difficult to control meaning that final assembly and finishing of the composites is a manual and highly skilled process. This commodity therefore provides the largest potential for RM in terms of tooling reductions and increasing process capability. Although there are no direct RM processes that can produce the composites of this nature, it is perhaps in the context of rapid tooling that RP technologies can be applied. Composite components are produced externally so RM applications in this area are not being considered in this paper.

Rapid Manufacturing Case Studies Metalwork: Door Latch and Seat Button The majority of metalwork within a typical suite is mainly primary or secondary structural components produced by machining and sheet metal respectively. There are no tooling costs for these processes although there is an investment and lead-time associated with CNC machine programming. The rapid manufacture of a batch of door latches (shown in Figure 4a) was commissioned when the original supplier could not meet the deadline for the first batch of production parts. They were produced using the Rapid Metal Casting (RMC) process, a Batch Manufacturing process, from Aluminium Alloy. The design was unchanged from being designed for machining, which caused some issues with the casting process, particularly the location of sprues relative to presentation surfaces and there were issues matching the RMC material with adjacent machined and extruded components. Although RMC had the potential to offer significant cost savings over machined components, the door latch examples show that the process cannot currently compete in terms of cosmetic appearance and repeatability for final production against machining processes but could be used appropriately for prototyping. The seat button (shown in Figure 4b) was a new, approximately 3 cm2, cosmetic part, conceptually designed to be built from 3160L Stainless Steel by the Selective Laser Melting (SLM) technique as an option to replace a generic button shroud with a feature more in keeping with the styling of the seat. Its complicated and styled geometry would have necessitated a casting process with the secondary tooling and etching operations unjustifiably adding to the components and would have been unfeasible for the production quantities of 30 units in total. Using RM meant that these features could all be included in the SLM build with only one final finishing operation to finalise the presentation surfaces. A one-off prototype to assess the design and the process was produced initially and integrated into a mock-up of the seat. Being functionally and aesthetically representative of the final product, a number of design changes were initiated including reducing the overall size of the part and moving the location of the attachment hole so as to better captivate the button relative to the soft trim surface that it was being installed in.

Figure 4 – RM examples: (a) Door Latch; (b) Seat Button; (c) Drip Trap (© CPAS)

Plastics: Drip Tray Plastic components tend to have high cosmetic requirements and although requiring tooling, the cost of vacuum forming tooling, used in the majority of cases, is reasonably low. Other processes, such as injection moulding, are rarely used due to the cost of tooling relative to the low production quantities. The drip tray was a 2 cm2, 1 mm-thick vacuum formed component used to shield an electrical connection from accidental spillage in a seat that was already in production (shown in Figure 4c). This small component was originally designed to be Polycarbonate and produced by vacuum forming but, due to its size, it proved difficult to produce efficiently using this technique. A total of only 300 parts was required for subsequent delivery and alternative production options were investigated including a Nylon based Polyamide component produced using the Selective Laser Sintering (SLS) technique and a Polycarbonate one produced using the Rapid Injection Moulding (RIM) process and compared with traditional injection moulding and the Batch Manufacturing process of Vacuum Casting (VC) (Figure 5). The SLS process was chosen as it proved slightly more cost effective for these quantities and avoided the tooling costs and lead-time of RIM whilst maintaining design flexibility. Further changes to the design utilised this freedom of geometry to include small locating features.

Figure 5 - Drip Tray Break-Even Analysis

Case Study Observations Component, Material and Process Suitability The main observation from the application of RM in these case studies is that the economic use of RM is currently most feasible for small, non-cosmetic and non-structural components. For metalwork, this applies to highly styled parts that can often be expensive to machine due to the complicated geometry dictated by styling. Although RM allows the production of lowvolume and geometrically unconstrained parts, there are still additional finishing operations required to finalise the part. For larger metal parts, the build time and envelope limitation of the current RM machines does not offer significant advantage over conventional machining

and specifically for structural parts, there is not enough understanding of build parameters to ensure repeatability and consistency or to enable up-front engineering analysis. For plastic materials, small components are difficult to produce by vacuum forming and they are produced in quantities that do not justify other tool-dependant processes such as injection moulding making RM particularly applicable. However, similarly to metal components, viability is dictated by any additional finishing operations. Hauge et al. [3] describe the limitations of the material and process information for limited RM materials that are currently available, particularly the reliability of supplier data sheets. For larger components, RM will have to compete both in terms of time and cost to vacuum forming. Flammability A specific requirement of the aircraft interiors industry is material flammability, specifically those dictated by the Airworthiness Authorities [2] and Airbus [4]. Similar to long-term environmental performance and aging of RM plastics, there has previously been limited information regarding their flammability performance. CPAS’s Technical Office has conducted a series of flammability tests on a 2-part urethane material used for Batch Manufacturing process of low-pressure injection moulding using silicon tooling known as Vacuum Casting, a particularly applicable process for production quantities usually encountered. Despite passing the vertical burn test, the material failed the smoke and toxicity, smoke density and heat release tests leading to the conclusion that it would be ‘inadequate for general use with an aircraft interior’ and limiting its use to ‘small items (under 12” square surface area) [and only] for [seats being installed on] Boeing Aircraft’ [5]. Although the possibility of producing larger plastic components by RM would be advantageous, this does not conflict with the current economic and process advantages associated with small plastic components that have been mentioned previously.

FURTHER CHALLENGES FOR RAPID MANUFACTURING AND PROTOTYPING Materials The biggest challenge facing the more widespread use of RM and RP is the material properties of the materials that are available as well as the information and understanding available on these materials and the process variables. Although the RM industry is continually developing more functional materials and plastics specifically to meet the aerospace industry’s flammability requirements, there is still little justification for users of RM to invest time and resources to experiment with new processes and develop further understanding on RM, especially in the context of a highly-regulated, customer-driven manufacturing organisation. Furthermore, there does appear to be a gap of readily available, high strength, low weight and predictable RM materials and processes that could compete with tooling intensive manufacture of composite components. Engineering Definition A component is fully defined through a combination of nominal geometric information and engineering definition such as dimensions and tolerances. Within CPAS’s engineering documentation procedures, the role of the CAD model is not controlled and used purely to record nominal geometric information of the part it is representing and there is no scope

within the CAD systems to include any engineering definition within the model file. This therefore requires the creation of 2D drawings to act as the design master communicate all the parameters of the design introducing a potential step for the duplication and corruption of data. As a result, the model file, the basis of design definition for the RM processes is not formally controlled, may not be up-to-date with the current design or fully define the part in terms of tolerance information and features such as threads leading to supply chains issues. Furthermore, models are only directly available to engineers with expensive CAD licences leading to inefficient and time-consuming data transfer. Comprehensive 3D CAD models are an important part of the RM process. Although most design and manufacturing companies have now made the inevitable migration to 3D CAD systems, previous systems have not allowed or facilitated the ability to compressively record all engineering definition within the model file so that it can be communicated efficiently. Although these capabilities are now available on more recent CAD systems, the real world migration from current systems is significantly lagging due to the capital investment and the sheer quantity of legacy data present in organisations. CPAS is in the process of upgrading its CAD system so that it will eventually allow the annotation and definition of tolerance information within the model files, though it may be some time before this function is fully educated, utilised and suppliers are fully compatible with it. If, however, RM is going to have a more prominent role within the manufacturing industry, is there the capability to intelligently interpret the information in excess of the nominal geometry that is vital when dealing with product assemblies? As with RM, the CAD model is the basis for the RP processes. Without the CAD capabilities to fully define the component within a model, for complex parts, RP rarely is commissioned before the component has been laid out in a 2D drawing to avoid the risk of losing definition. It is important however, that prototyping occurs before any definition has been laid out in 2D drawings to maintain its fluidity and maximise the efficiency of design iterations as a result of the prototyping. Rapid Prototyping Material and Process Selection Although there are some highly functional RP materials available, they are not representative of the final production material or process properties. This introduces additional considerations to the justification of RP, often necessitating a compromise of delivery time, cost, quantity and functionality. Further considerations include build envelope, quantity and the compromise of accuracy and surface finish with build time. Where post-processing operations are required, operator skill is also a consideration and can add considerably to the delivery time. In fact, for large parts, the build time might even be in excess of conventional machining. Unlike traditional manufacturing methods, RP is not supported with a large amount of accumulated historic information. These RP decisions are perhaps best left up to the external RP bureau though this limits the associativity between the final prototypes and the exact validation requirement. Virtual Prototyping Virtual prototyping tools include photo realistic rendering, kinematic and clash and tolerance analysis and model interrogation. The application of these tools can potentially improve the quality, capability and speed of the engineering analysis, particularly relative to physical prototyping and testing. The feedback as a result of the use of these virtual prototyping tools is much more rapid than with physical prototypes, meeting high levels of simulated fit and

function analysis and visualisation if not tactility. Furthermore, they offer the potential to share not only geometric representations, but also annotated engineering definition with nonCAD users. It is recognised that Best-In-Class manufacturers, whilst still putting emphasis on the benefits of physical prototypes, commission on average 30% fewer prototypes than less successful companies within their product development process by favouring virtual prototyping methods [6]. Rapid Manufacturing enabling Rapid Prototyping Prototypes, specifically Rapid Prototypes, are usually produced on different machines with different process capabilities to those intended for the final method of manufacture. As a consequence, an RP part may be representative of the nominal design but not necessarily production tolerances, limiting their application for accurate fit testing or the assessment of production mechanisms. Sourcing prototype parts through RP techniques means that they are not competing with prioritised production capacity so, do not compromise delivery. This does however mean that a valuable opportunity to develop the manufacturing process for these parts and consequently validate the design is lost. This is the same for outsourced conventionally produced prototype parts. If the RM processes are developed to the extent that the majority of parts can be economically produced using these methods, the scope for RP will be dramatically increased and have a significant influence of the product design and development process as a production representative prototype will be readily and quickly available.

CONCLUSIONS The manufacture of premium aircraft seating is a highly regulated and challenging industry. Despite the obvious advantages of RM and RP, CPAS has had limited success in implementing these processes within the engineering and manufacturing departments. This is due to a number of well-recognised limitations of the RM materials and processes and internal company obstacles: 

Currently, parts for which it is most economically feasible to produce using RM are small, non-structural and non-cosmetic parts with additional finishing requirements still proving to be a limiting factor.



For the current nature of suite designs, it is impossible for RM to compete on a large scale with conventional and established manufacturing processes, specifically due to the extent of process understanding and material properties but also in terms of economics.



Specifically for the aerospace industry, flammability requirements largely dictate what plastic materials can be used in products. There is however scope for small plastic components which ties-in with economic considerations.



There is a lag in the migration of industry to full RM enabling CAD systems providing comprehensive engineering definition, model file control and transfer.



Computer-enabled simulation and analysis tools can perform much more encompassing analysis on production representative models much quicker and accurately than on physical prototypes.



There is a disassociated relationship between the available RP materials and process properties and those for final production. If final production is achieved through RM, then the possibilities for prototyping will also be significantly increased whilst greatly streamlining the design and manufacturing process.

A number of researchers have stated that true RM is still up to 10 years away. This is an accurate reflection of the findings of this paper as currently, RM applications are limited to small and non-structural plastic and metal parts. When these limitations are overcome however, there is the potential for the acceleration of the product development process for complex mechanical products.

REFERENCES 1.

Hopkinson, N., Hague, R. and Dickens, P. (2006) Rapid Manufacturing: An Industrial Revolution for the Digital Age, John Wiley & Sons Ltd.

2.

Gardlin, J. (2003) Airworthiness Standard 14CFR 25.853 (a & d), Federal Aviation Administration, FAA Airframe and Cabin Safety Branch, 1601 Lind Ave. SW, Washington 98055, USA

3.

Hague, R., Mansour, S. and Selah, N. (2004) Material and design considerations for rapid manufacturing, International Journal of Production Research, Vol. 42, pp. 46914708

4.

Turanski, P. (2003) Fireworthiness Requirements Pressurized Section of Fuselage ADB 0031, Issue E, Materials and Processes, Airbus, Airbus Documentation Office, 31707 Blagnac CEDEX, France

5.

Carr, P. (2005) Flammability Report AXZON-A-2062, Technical Office, Contour Premium Aircraft Seating, Cwmbran, NP44 3HQ

6.

Brown, J. (2006) Simulation-Driven Benchmark – Getting It Right First Time, The Aberdeen Group, accessed online from; www.aberdeen.com; 09/01/2007