or

X4

rateproinecee sus ththis. AMon

recent growth in applications for direct part manufacture, or Rapid Manufacturing, has resulted in much

denfrom 3tractivning (e addiuring,w a laof man

With over 20 years of history, in its early years AM was mostly

also known as Rapid Prototyping (RP). These prototypes were most

vided a summarythe use of RM asr wishing to takeng they are con-

design,

Contents lists available at ScienceDirect

w.e

Int. J. Productio

Int. J. Production Economics 149 (2014) 194201technologies for production applications. It is inevitable that someE-mail address: sm270@ex.ac.uk (S. Mellor).commonly used as communication and inspection tools, producing management, organisation and implementation.

These issues are inter-related and this study centres on thethird of these areas, specically focusing implementation of AM

0925-5273/$ - see front matter & 2013 Elsevier B.V. All rights reserved.http://dx.doi.org/10.1016/j.ijpe.2013.07.008

n Corresponding author. Tel.: +44 7709994113.applied for the fabrication of conceptual and functional prototypes, manufacturing processes and materials,(3DP). Since the development of many of these technologies hasoccurred simultaneously, there are various similarities as well asdistinct differences between each one (Kulkarni et al., 2000).Reviews of the numerous AM technologies have been performedin previous works (Gibson et al., 2010; Groover, 2007; Hopkinsonet al., 2006).

which are for producing personalised products in(Strategic Directions, 2008). Ruffo et al. (2007) proof the pitfalls which exist for companies looking ata solution for current manufacturing problems oadvantage of this emergent technology, suggesticentrated in three specic areas:more widely used include stereolithography (SL), fused depositionmodelling (FDM), selective laser sintering (SLS) and 3D printing

emerging today; though its economic impact remains modest (Levyet al., 2003). There are few-large scale applications of RM, many of

the medical eld1. Introduction

Additive Manufacturing (AM) isjoining materials to make objectslayer upon layer, as opposed to subodologies, such as traditional machinyms found in the literature includtechniques, additive layer manufactand freeform fabrication. There is noogies which employ this methodresearch effort focusing on development of new processes and materials. This study focuses on theimplementation process of AM and is motivated by the lack of socio-technical studies in this area.It addresses the need for existing and potential future AM project managers to have an implementationframework to guide their efforts in adopting this new and potentially disruptive technology class to producehigh value products and generate new business opportunities. Based on a review of prior works and throughqualitative case study analysis, we construct and test a normative structural model of implementation factorsrelated to AM technology, supply chain, organisation, operations and strategy.

& 2013 Elsevier B.V. All rights reserved.

ed as the process ofD model data, usuallye manufacturing meth-ASTM Standard). Syno-tive processes, additivelayered manufacturing,rge number of technol-ufacture, some of the

several physical models in short time directly from computer solidmodels helped to shorten the production development steps(Santos et al., 2006). RP remains the dominant application ofpolymer AM processes and is well established in the market. Manyof the aforementioned technologies are limited to Rapid Prototyp-ing as they do not allow common engineering materials to beprocessed with sufcient mechanical properties (polymers, metals,ceramics, and composites thereof) (Kruth et al., 2007). The conceptof Rapid Manufacturing (RM) the production of end-use partsfrom additive manufacturing systems (Hague et al., 2004) isAdditive manufacturing: A framework f

Stephen Mellor n, Liang Hao, David ZhangCollege of Engineering, Mathematics and Physical Science, University of Exeter, Exeter E

a r t i c l e i n f o

Article history:Received 15 April 2012Accepted 9 July 2013Available online 23 July 2013

Keywords:Additive manufacturingRapid manufacturingTechnology implementationCase study

a b s t r a c t

As mass production has migswitch towards low volumeadded value. To competetechniques to provide the nlow volume production. Onmanufacture which involvemodel data. The benets ofand economic low volumesyears its dominant applicati

journal homepage: wwimplementation

4QF, United Kingdom

d to developing countries, European and US companies are forced to rapidlyduction of more innovative, customised and sustainable products with highthis turbulent environment, manufacturers have sought new fabricationssary tools to support the need for increased exibility and enable economicch emerging technique is Additive Manufacturing (AM). AM is a method ofe joining of materials, usually layer-upon-layer, to create objects from 3Dmethodology include new design freedom, removal of tooling requirements,consists of various technologies to process versatile materials, and for many

has been the manufacture of prototypes, or Rapid Prototyping. However, thevier.com/locate/ijpelsen Economics

of

ins

ove

e somesimilarities but have a number of distinctions. The rst AM system

Part removal and cleanup Post-processing of part

been applied to the production of consumer products, includinghigh value lighting goods and electronics. The aerospace sector hasalso found a number of applications, often driven by the possibilityof improving buy-to-y ratios (as some AM processes have highmaterial utilisation, most notable metal-based process) and redu-cing the weight of components through design optimisation(Petrovic et al., 2011). Other areas include, automotive, jewellery,architecture and defence applications.

2.2. New technology implementation: theoretical background

Skinner (1984) was one of the rst to propose that innovationin production technology can be used strategically as a powerfulcompetitive weapon, suggesting that it can bring to bear manyother strategic factors besides achieving low costs including,superior quality, shorter delivery cycles, lower inventories, lowerinvestments in equipment, shorter new product developmentcycles and new production economics. In Porters (1985) inuentialwork on competitive strategy he suggested technology is perhaps

S. Mellor et al. / Int. J. Production Economics 149 (2014) 194201 195 Application

Holmstrm et al. (2010) suggest the unique characteristics ofAM production lead to the following benets:

No tooling is needed signicantly reducing production ramp-up time and expense.

Small production batches are feasible and economical. Possibility to quickly change design. Allows product to be optimised for function (for example

optimised cooling channels). Allows economical custom products (batch of one). Possibility to reduce waste. Potential for simpler supply chains; shorter lead times, lower

inventories.to be commercialised was SL, whereby a concentrated beam ofultraviolet lamp is used to solidify a liquid photopolymer bytracing a two dimensional (2D) layer in the form of a contourand then an inll. Once the beam has completed a single layer thebuild platformwill then move downward in the z-axis, a new layerof photopolymer is distributed and the process is repeated untilthe nal layer is completed. Laser sintering and laser meltingprocesses work in a similar manner, whereby polymer or metallicpowders are selectively melted in 2D layers, through high powerlasers until a solid part is complete. Another popular process,particularly with hobbyists, is the FDM process. In this method,materials, usually polymer laments are extruded through aheated nozzle to print 2D layers successively, one on-top ofanother, until the part is complete. Whether through melting ofmetallic powders or through extrusion of polymer laments,all AM process share the additive principle of building compo-nents. It is possible to identify a number of key steps in the AMprocess sequence. Gibson et al. (2010) denes eight key steps inthe generic process of CAD to part:

Conceptualization and CAD Conversion to STL Transfer and manipulation of STL le on AM machine Machine setup BuildAs previously stated the numerous AM systems shar2.1. AM technologies and applications2. Backgroundand suggestions for future research.research framework is presented with a detailed description of theconstructs and supporting literature. The data collection process isthen described and the results of the framework test are described.Finally, the paper closes with conclusions, limitations of the studywitpaper is organised as follows. In the next section, a shortrview of AM technology and applications is presented alongh an introduction technology implementation theory. Then thethe

reqAM technologies due to their unique characteristics, resourceuirements, benets and tradeoffs and so on. The remainder oftose issues, rather this paper seeks to build on this, addingights into factors that are specic, or of particular importancethe

logies. However, it is not the aim of this study to rediscoverno

also important to the adoption of other manufacturing tech-are

the factors critical to the implementation of AM technologiesDesign customization.These benets have been captured in a variety of applicationsspanning a number of industries, and different stages of theproduct development life cycle. Examples include titanium aero-space parts where only 10% of the raw material is required whencompared to the original machined part (The Economist, 2012).Atzeni and Salmi (2012) showed the economics of additivemanufacturing for end-use parts through comparing the produc-tion of landing gear aircraft assemblies, through high pressure diecasting (HPDC) and laser sintering. The authors showed the costbenet at low to medium production volumes, illustrated in thebreakeven analysis shown in Fig. 1. The benets of AM have beencaptured in the production of race car gearboxes (The Economist,2012). AM facilitates the manufacture of smooth internal pathways, providing faster gear changes and reducing componentweight by 30%. Similarly, Cooper et al. (2012) illustrate thepotential for improved functionality in their study on formulaone technology, applying AM to hydraulic component manufac-ture gaining efciency of uid ow of 250%.

As previously stated the current dominant application for AMprocesses remains RP. Rapid Tooling (RT) also makes up some ofthe current AM activity which involves the fabrication of mouldsand dies. Regarding manufacturing applications of AM processes(RM), notable areas of success include the production of medicaldevices such as dental crowns and hearing aids, driven bycustomer requirements for individualised products and AM pro-cesses having the benet of design customization. RM has also

Fig. 1. Breakeven analysis performed by Atzeni and Salmi (2012) comparing HPDCand SLS processes.the most important single source of major market share changes

among competitors and is the prominent cause of the demise of anentrenched dominant rm.

Voss (1988) provided seminal work on proposing implementa-tion as a distinctive area of study of process innovations. The focuson AMT implementation in 1990s was a direct result of manysystems failing to meet their initial promise, with project man-agers unhappy with the system performance. In more recent years,the advent of ERP systems and RFID technology has generated aplethora of research articles on implementation as academics havesort to create process models and frameworks to assist managersin implementing new innovations successfully. Though AM as amanufacturing technology remains in comparatively low levels ofexploitation, with AM production representing only a very smallpercentage of global manufacturing, some authors suggest thebreakthrough is eminent. In order to take a pro-active approach toresearch we focus on developing an implementation frameworkbased on current innovators in this area in order to facilitate thisbreakthrough.

3. The research framework

put for-mework

proposes that both external forces and internal strategy drive

Products with increased functionality through designoptimisation.

Products of low volume.

The implementation of AM must be preceded by strategicalignment of the business, manufacturing and R&D strategy. Thetechnology benets must be linked to the capabilities required ofthe manufacturing unit, capabilities derived from the businessstrategy, viewed as the market-pull strategy to AM implementa-tion. However, it is also proposed in line with the currentresource-based view of the rm that investment in AM may beseen as a structural investment which will build new manufactur-ing capabilities, creating new business opportunities for theenterprise, the technology-push strategy.

3.2. Technological factors

In order for the adopting organisation to gain competitiveadvantage from the implementation of AM its ability to link thetechnology benets to the business strategy has been emphasised.The technology benets associated with AM technologies possessa number of benets, as previously stated in Section 2. However, as

S. Mellor et al. / Int. J. Production Economics 149 (2014) 194201196the consideration of AM as a method of manufacture and theapproach to AM implementation will be inuenced by factorswhich may be grouped in to ve constructs. These constructs arelaid out described in the following sections:

3.1. Strategic factors

The decision to invest in Additive Manufacturing technologiesmust be linked to the market and product characteristics. Highutilisation underpins any technology investment (Hill, 2000), if theprocess will not be highly utilised on one product it must meet themanufacturing and business needs of other products. Authors inthe eld of AM management have proposed a number of productcharacteristics which affect the types of products suitable to AMproduction. Generally, the product characteristics are as follows:

Products with a degree of customisation.wa

The conceptual framework for AM implementationsrd by the authors is illustrated in Fig. 2. The fraFig. 2. The proposed framework oshown in previous work by Sonntag (2003) it is equally importantthat the adopting rm also understands the trade-offs in usingnew manufacturing technology, for AM process (in general)material range remains low, machine and material costs remainhigh and process speeds are relatively slow. The lack of technicalstandards also presents a major barrier to adoption. Some of thesecharacteristics of AM are likely due to their relative immaturityand managers should be aware of this when deciding whetherto adopt.

There is also an inherent RP legacy with AM system which mayresult in a psychological barrier to adoption, as management onlysee the technology-class as being suitable for RP applications. Theadopting organisations ability to present benets of AM as amanufacturing process in a clear and balanced way will determinethe success of implementation.

3.3. Organisational factors

The size of an organisation has been identied to be critical tothe understanding of the process of implementation of newf AM implementation.

are considered explicitly with a view to inuencing the design. AMrequires users to match product with process and to understand

2003) have identied the substantial information gap regarding

S. Mellor et al. / Int. J. Production Economics 149 (2014) 194201 197new technology process capabilities. Therefore the workforceexperience and skill is also proposed to be a key factor in AMimplementation.

3.4. Operational factors

As proposed by Bailey (1993) a change in an organisationstechnology will inuence both its operational and administrativestructures to change. One area of operations which has beenproposed by many authors to be signicantly changed with theadoption of AM is product design. A number of authors havecommented on the impact of Additive Manufacturing on thedesign of products and designers themselves (Hague et al., 2003,2004; Mansour and Hague, 2003). The additive nature of AMprocesses means that this type of manufacture is unconstrained bymany of the limitations of conventional (subtractive or formative)processes (Hao et al., 2010). The unique characteristics of AMsystems require new design tools and practices to be developed,contrary to early promise made by some researchers there is nottotal geometric freedom and many consideration have been takeninto account when designing products for AM processes. It isproposed that the designers understanding of the new design foradditive manufacturing constraints will be an inuential factorin AM implementation approach.

Another area of operations which is likely to change signi-cantly with the adoption of AM is production planning and qualitycontrol. Research on AM process planning is still lacking, thoughCiurana and Riba (Munguia et al., 2008) investigated processingplanning strategies employed at 36 Additive Manufacturing cen-tres in Northern Spain. The authors used survey analysis, sup-ported by personal interviews with technicians, to identifystrategies used AM process planning including; part orientationstrategies, build volume strategies, layering strategies, supportmanufacturing technology. A number of scholars have suggestedsmall business cannot be considered scaled-down larger ones, andthe theories proved in large enterprises might not be suitable forsmall business (Federici, 2009; Schubert et al., 2007; Thong et al.,1996; Welsh and White, 1981). Therefore, the approach to imple-mentation for an SME is likely to be different to that in a largemultinational company. Linked to size, previous study into newmanufacturing technology implementation suggests that thestructure of an organisation is the key factor to successfullyimplementing manufacturing technology (Dalton et al., 1980;Dean et al., 1992; Belassi and Fadlalla, 1998; Abdul Ghani et al.,2002; Sun and Tian Cui, 2007; Saberi et al., 2010), and thatcompanies that adopt without rst re-designing organisationalstructures and processes encounter high difculties (Saberi et al.,2010; Millen and Sohal, 1998). Therefore, it is proposed forsuccessful implementation of AM technologies the decision toadopt will be accompanied by a change in jobs and tasks, and thusa change in work practices and structure.

Many authors have advocated that new technologies challengeestablished norms and strategic options. Linked to structure,organisational culture denes the complex set of knowledgestructures which organisation members use to perform tasksand generate social behaviour (Saberi et al., 2010). Hopkinsonet al. (2006) suggest possibly the largest but unknown impactcould be on company culture and how it changes to accommodaterapid manufacturing (RM). Using AM processes as a manufacturingtechnology requires designers and engineers to re-think design formanufacturing (DFM). DFM is any aspect of the design process inwhich the issues involved in manufacturing the designed objectgeneration and minimisation. However, as the authors concededthe true cost of AM system, implementation and operation. Inprevious studies on AM costing four key cost factors have beenidentied for additive processes: operation times, machine costs,labour costs and material costs. Ruffo et al. (2006) providedseminal work on the cost estimation of AM processes andextended this research for simultaneous production of mixedcomponents using laser sintering (Ruffo and Hague, 2007). How-ever, for other systems such as metal based processes thereremains a signicant gap in this knowledge.

In the context of AM no technology is currently capable ofcreating net shape parts, thus post-processing (such as supportremoval, heat treatment, etc.) are required, therefore the integra-tion of AM within a supportive production system is proposed tobe key to implementation success.

3.5. Supply chain factors

Additive Manufacturing implementation lies at the intersectionof two supply chains; rst it involves a supply chain from themachine vendors to the purchaser of the technology. Second, thepurchaser will then embed the technology in their respectivesupply chain and hence inuence their customers and suppliers.On the subject of new process technologies implementation,Bessant (1994) has argued that for the technology to deliver itsfull potential signicant organisational changes are required suchas the restructuring of relationships with suppliers towards morecollaborative forms. It is proposed that the implementation of AMtechnologies manufacture will require increased collaborationwith suppliers and customers. Vendor support during the imple-mentation process has long been recognised as a critical factor ofimplementation success. It has been shown that the level ofcomplexity of the technology innovation is directly related to thelevel of intensity of the user-supplier interaction processes (Zairi,1998). Therefore vendor support is proposed to be a key factor inAM implementation.

A characteristic of current AM industry worthy of note is thetendency for machine suppliers to be material suppliers (such asthe powders used) following implementation. This characteristic ispartly due to the immaturity of the technology (with a shortage ofmaterial suppliers) and also likely a strategy on the machinesupplier's part to protect future business. There will also bedecisions on where to locate manufacturing as the removal oftooling requirement may result in the possibility of distributingmanufacturing according to demand locations as in theory theonly inputs required for production are CAD data and rawmaterial.

4. Research methodology

Due to the exploratory nature of this research area, case studiesare used to investigate the AM implementation process and testthe research framework. Implementation research provides gui-dance to identify the most suitable interviewee for this researcharea, indentifying the four most critical constituencies as technol-ogy vendors, upper management, project engineers and planta limitation of this work is that the majority of the activity in thisstudy was RP and RT.

Hayes and Jaikumar (1991) suggested the use of unsuitable costaccounting methods, where implementation of new technologiesresults in costs incurred shifting from direct labour costs toessentially xed costs, often renders traditional methods focusingon less important factors unsuitable. A number of authors(Munguia et al., 2008; Grimm, 2004; Hopkinson and Dickens,operating and maintenance personnel (Zmud, 1984; Cooper and

S. Mellor et al. / Int. J. Production Economics 149 (2014) 194201198the implementation of DMLS at Company A. One of the maincontributing factors to implementation may be viewed as the CEOsperception of technology benets, at a stage where the technologywas very much in its infancy, the CEO recognised the potential forcompetitive advantage through innovation and technologyinvestment:

It was a belief as much as anything; there was no precedent tobase it on.I could see that metals was going to be massive in thefuture.I wanted to make sure that I set us up ready for when ithappened so that we were in a position of strength rather trying tocatch up.

Originally a specialist in purely SLS, another factor contributingto the decision to invest in DMLS at Company A may be viewed asthe changes in the RP sector. As the systems have matured, manycompanies have taken this capability in-house which has reducedthe amount companies request from specialist suppliers, such asCompany A. Though this change has had drastic effects in certainmarkets for the company, in many it has been less severe andalthough demand may have decreased, it has not disappeared.

Though the company has achieved signicant success thechallenge for case company is that in order to maintain thebusiness benets of the technology it needs to prove its produc-tion capability:

.the only way we are going to make any money on thissup

The company has a clear mission of becoming Europe's leadingplier of SLS and DMLS parts. A number of factors have led to5.1. AM strategy5.Zmud, 1990; Majchrzak, 1988; Hattrup and Kozlowski, 1993; Noori,1990). A single case was chosen for this study in order to get an in-depth view of AM implementation. The case company wasselected as it represents a rare case of a successful RM implemen-tation (in terms of business success), as it is a national leader insupplying both plastic and metal RM parts.

4.1. Company background

Company A is a leading supplier of SLS and Direct Metal LaserSintering (DMLS) products in Europe specialising in the productionof complex and functional metal rapid prototypes, aestheticmodels and low volume production components to industriesincluding Aerospace, Automotive, Dental, Medical, FMCG, Marine,Defence and Pharmaceutical. The company began the rst steps indeveloping a DMLS capability four years ago. The informant atCompany A was the company Chief Executive Ofcer (CEO) theenquiry was also supported through interviews with the machinevendor. The company was started as a specialist in polymer AMprocesses and focusing on producing prototypes for externalcustomers. Following implementation of metal-based processesthe focus has been on production applications due to the processcosts and the reduction in overhead costs for higher volumes. Thefocus on the implementation of the DMLS systems was made forfollowing reasons:

the interviewee was directly involved in the implementation ofDMLS technology and may be viewed as the project champion,

the interviewee regarded the metals based processes as key tofuture success, and nally,

when discussing production applications DMLS was the pro-cesses generally referred to as providing the major benets.

Results and discussionultimately is by doing production because the overheads forproduction work are radically lower than they are for prototyp-ingso then you have the problem, you don't want to doprototyping, but that's the only business around, you want to doproduction, but to do production and be good at it you've got toshow you're good at doing production. So you've got a realdichotomy there.

The strategy at Company A is therefore signicantly inuencedby its organisational antecedents, coming from a RP backgroundrather than a production background. The informant suggests that

it would have been better to be a machining company that adds itin rather than a rapid prototyping company that is trying tobecome a serious aerospace or whatever supplier

There are a number of explanations for this reasoning whichinuence the approach to AM implementation at Company A,including, the culture within RP (discussed in Section 5.3), thecompany not having an established customer base (as discussed inSection 5.5) and lacking production capability (particularly regard-ing the post-processing requirements in DMLS, discussed inSection 5.4). Their response to this apparent inefciency has beento become experts in design and application development in aparticular way. Though the company has built up a workshopfacility to support the DMLS system, through deliberately ndingand designing parts to suit the process they have been able to getround this inadequacy and as the informant suggests:

it is much more protable to design the part where you don't doanything to the part after you have nished, so you design it sothat it doesn't need anything doing to it and if you can do thatthen its far more protable.and far quicker.

Therefore, this represents the company's main in-house cap-ability. However, an important implication of this is that this doesrequire a lot of design thought, therefore regarding productivity,downstream process may be reduced, but upstream process isincreased. Regarding the current market for DMLS products, theinformant suggests that it is difcult to know where the next jobwill come from and how the market for DMLS products willchange over time. Therefore it remains a particularly turbulentenvironment, where both the technologies and the marketsare relatively new. The informant also states that with a fewexceptions, industry knowledge of the process has signicantlylagged behind that of the users. Among these exceptions is theaerospace industry, with requirements for lighter aircraft andmore efcient processes often inuenced by environmental legis-lation. Company A is targeting these manufacturers as they requireless education on process benets and the components are moresuited to AM production.

5.2. AM technology

The CEOs perception of AM benet is that they must beconsidered over the life cycle of the product, which itself presentsa challenge when attempting to educate customers on thesepotential opportunities. This is reected in the company's focuson Aerospace applications. In aircraft component manufacturer,the potential to reduce part weight through the design freedomunlocked when using AM processes can create mass savings overthe product life cycle. This is also reected in the company mainin-house capability, design and application development for AMprocess. The CEO suggests:

.in terms of material usage.particularly with the metals it'svery good but you have to go through a high energy process ofturning it into powder in the rst place, do you really gain a

benet there? It's marginal to be fair. The actual machines are they

5.3. Organisational changethat they do not have an established customer base for productionmetallic components. Compared to a machining company who is

S. Mellor et al. / Int. J. Production Economics 149 (2014) 194201 199The company is an SME and is purely self-funded which resultsin a lack of capital for technological investment and R&D activities,creating a barrier to increasing capacity and developing produc-tion applications. One approach to solving these issues is using RPto fund RM R&D activities. For Company A the experience and skillshortage is in conventional metal working and aerospace processow. The role of the production manager for the DMLS imple-mentation was identied as an area where this is most visible,with the CEO describing the role as being much more specializedthan that of the SLS side of the company. Therefore this may beviewed as an organisational characteristic of RP companies movingto RM, with experience and skill gap in production applications.

The company's experience in RP has also affected the organisa-tional culture, as the knowledge structures which organisationmembers use to perform tasks and generate social behaviour willhave been invented, discovered or developed in an RP environ-ment. The CEO suggests that in an RP environment the focus is toget parts out quickly and at a decent quality, speed is likely to bean important organisational strength when dealing with changeand the demand for quick turnaround is key to success. Whereasin the manufacture it is likely that cost and quality control becomecritical for success. The company has a centralised organic struc-ture with the CEO being the key decision maker which hasbenets in this turbulent environment for speed of responsehowever they are vulnerable to individual misjudgement. Theworkforce structure is composed of design engineers and engi-neers, with a production manager. The structure has changedsignicantly, starting as a replication of the original plastic RP sidestructure. The CEO suggests the organisational structure is some-thing they will likely work out as they go along, as there are noreally efcient at building stuff now? NoIn terms of efciency ofuse of energy they are not very good at all..but the real benet iswhen you look at the fact that you can produce lighter parts.That is really where it comes into its own and the fact you canthen turn things round faster, you can design things that are closerto doing the job that you wanted to do in the rst place but youhad to compromise.

The most signicant trade-off for the case company was identiedas the machine cost, hindering the potential for increasing in-houseDMLS capacity. Other tradeoffs identied by the CEO include highprocess costs, largely due to the slow speed at which parts areproduced, resulting in high product costs reducing the potentialmarket size of DMLS. However, as the informant states these arelikely to change as the technology matures:

The tradeoffs are constraints in terms of design for the metalparts because of the need for supports.currently limited range ofmaterials but that will change.it's expensive, that will change afair bit. In fact it will probably become an irrelevance at somepoint in twenty, forty years time or whatever

Though the company focuses much of its activity on aerospaceapplications development, the certication periods for this sectorhas meant the company have been forced to look at other sectorsover this period of certication. This is in part due to the fact thatthere are few technical standards for AM processes, resulting inlong certication periods for safety critical parts in commercialaircraft applications.precedents to work from.trusted as a parts supplier, the company must therefore spendsignicant amount of resources on attracting new customers.On the subject of machine suppliers vendor restrictive practicesincluding machine suppliers controlling the powders which can beprocessed and locking down machine parameters were discussed.These practices reduce the material range available to the organi-sation, likely reducing the potential products therefore marketsthe company can serve. Powder control by the supplier also meansthat material cost remains high, as there is a lack of competition.Second, restrictive practices by suppliers include locking downprocess parameters creating annoyance for high end users. Thisis likely to hinder the R&D practices of the company as thecompany is unable to experiment with process parameters opti-misation tasks. This issue also creates a reliance on the machinesuppliers R&D activity, as the systems become closed to operatoradjustment. The CEO suggests it is surprising how little themachine vendors look to learn from the experience of CompanyA regarding machine development. In particular, he points outhow some vendors are more responsive to making them produc-tion capable than others.

In this case location of manufacture remains one of centralisedproduction, as the option of locating production according todemand is reliant on an established customer base, or at least anunderstanding of the demand for products according to location.Also, the requirements for post-processing and supporting equip-ment (CNC, etc.) restrict the exibility of the manufacturingsystem in terms of location. This is likely to be the case for sometime until further machine improvements are in place whichreduce the requirements for post-processing of components and5.4. Systems of operations

The company spends a huge percentage of customer facingtime educating customers. This education is likely to be aroundprocess capabilities and constraints, where design for processconsiderations must be taken into account in order to capturethe technology benets. From this case study the quoting processis identied to be a time and resource consuming operation.Whereas in RP, quotation is less complex as the process chainitself is signicantly less complex (discussed below). Furthermorein RP there is no need for part re-design and cost is not necessarilyan order winner at one-offs and small batch volumes, wheredelivery speed may be more important. For RM, process chaincomplexity, design for process and cost reduction must all beconsidered at the quotation stage.

The process chain in RM applications was also highlighted as anarea of signicant change during the case study. The length andcomplexity of the process chain in RM applications are drasticallyincreased (when compared to RP) as heat treatment, nishing andmeasuring processes are required for quality production parts. Inthis environment the CEO suggests it becomes a whole workshopcoordination. This increase in complexity is partly due to thecharacteristics of the DMLS process, requiring support removaland other post-processing activities. Second, this is due to thequality control requirements for production parts when comparedto prototypes. As previously stated, these downstream processesmay be reduced through quality design for process and optimisedprocess planning strategies.

5.5. The AM supply chain

The case company's background in Rapid Prototyping meansAM moves closer to net-shape manufacture.

S. Mellor et al. / Int. J. Production Economics 149 (2014) 1942012006. Conclusions and future work

The case company has shown to provide support for theresearch framework, its constructs and provided insights into therelationships between the variables. During the analysis stage carewas taken to identify the reasons for the factors identied; toidentify those that may be common characteristic to AM imple-mentation in a certain environment and therefore a source of apotentially more generic solution (improving external validity).Some of the management and implementation challenges for theRP convertor have been discussed along with potential solutionsand opportunities. It is important to acknowledge the nature ofthe case study presented in this paper and the specic scenariounder study. The scenario investigated in this paper is that of acompany coming from a background in prototyping and imple-menting AM as a new manufacturing process for production ofnew products. It is expected that the inuence and importance ofthe framework factors will be determined by the scenario understudy. For example case, the challenge of changing an RP cultureand building a reputation as a production company are likely to beless inuential in a company coming from a background intraditional machining and established in an aerospace supplychain. In such a case, the challenges with understanding newdesign for additive manufacture constraints and changing a tradi-tional production culture would likely have greater inuence onimplementation success. Therefore, future work may look tocompare the approaches in these different scenarios, and poten-tially map these approaches using the factors presented in thisframework.

Limitations of the study include the fact that the frameworkwas tested using a single case study. Yin (2003) suggests single-case designs may be viewed as vulnerable and Voss et al. (2002)also advise that single case research limits the generalisability ofthe conclusions, models and theory developed. Although theserisks exist in multi-case research they are somewhat mitigated andtherefore to improve the generalisability of the framework futurework may be focused on further case studies of AM implementa-tion in different organisational contexts and supply chain scenar-ios. As the number of implementers increases the variety of caseswill be open to researchers. Though there is unlikely to be onecorrect approach to implementing AM processes, this study hasprovided an insight into the challenges with AM implementationand has proposed a framework to assist managers in implement-ing this potentially disruptive technology class.

References

ASTM Standard, Standard Terminology for Additive Manufacturing Technologies,vol. 10.04.

Abdul Ghani, K., Jayabalan, V., Sugumar, M., 2002. Impact of advanced manufactur-ing technology on organisational structure. Journal of High Technology Man-agement Research 13, 157175.

Atzeni, E., Salmi, A., 2012. Economics of additive manufacturing for end-usablemetal parts. The International Journal of Advanced Manufacturing Technology62 (9), 11471155.

Bailey, J., 1993. Managing People and Technological Change. Pitman, London.Belassi, W., Fadlalla, A., 1998. An integrative framework for FMS diffusion. Omega

26 (6), 699713.Bessant, J., 1994. Towards total integrated manufacturing. International Journal of

Production Economics 34 (3), 237251.Cooper, D.E., Stanford, M., Kibble, K.A., Gibbons, G.J., 2012. Additive manufacturing

for product improvement at Red Bull Technology. Materials & Design 41 (0),226230.

Cooper, R.B., Zmud, R.W., 1990. Information technology implementation research: atechnological diffusion approach. Management Science 36 (2), 123139.

Dalton, D.R., Todor, W.D., Spendolini, M.J., Fielding, Gordon J., Porter, L.W., 1980.Organization structure and performance: a critical review. The Academy ofManagement Review 5 (1), 4964.

Dean, J.W., Yoon, S.J., Susman, G.I., 1992. Advanced manufacturing technology andorganization structure: empowerment or subordination? Organization Science

3 (2), 203229.Federici, T., 2009. Factors inuencing ERP outcomes in SMEs: a post-introductionassessment. Journal of Enterprise Information Management 22 (1/2), 8198.

Gibson, I., Rosen, D.W., Stucker, B., 2010. Additive Manufacturing TechnologiesRapid Prototyping to Direct Digital Manufacturing. Springer, New York.

Grimm T., 2004, User's Guide to Rapid Prototyping. Society of ManufacturingEngineeers, Dearborn Mich.

Groover M., 2007. Fundamentals of Modern Manufacturing, 3rd ed. Wiley,Hoboken NJ.

Hague, R., Campbell, I., Dickens, P., 2003. Implications on design of rapid manu-facturing. Proceedings of the Institution of Mechanical Engineers, Part C:Journal of Mechanical Engineering Science 217 (1).

Hague, R., Mansour, S., Saleh, N., 2004. Material and design considerations for rapidmanufacturing. International Journal of Production Research 42 (22),46914708.

Hao, L., Mellor, S., Seaman, O., Henderson, J., Sewell, N., Sloan, M., 2010. Materialcharacterisation and process development for chocolate additive layer manu-facturing. Virtual and Physical Prototyping 5 (2), 5764.

Hattrup, K., Kozlowski, S.W.J., 1993. An across-organization analysis of the imple-mentation of advanced manufacturing technologies. The Journal of HighTechnology Management Research 4 (2), 175196. (Autumn).

Hayes, R.H., Jaikumar, R., 1991. Requirements for successful implementation of newmanufacturing technologies. Journal of Engineering and Technology Manage-ment 7 (34), 169175.

Hill, T., 2000. Manufacturing Strategy: Text and Cases, 2nd ed.. Palgrave,Basingstoke.

Holmstrm, J., Partanen, J., Tuomi, J., Walter, M., 2010. Rapid manufacturing in thespare parts supply chain: alternative approaches to capacity deployment.Journal of Manufacturing Technology Management 21 (6), 687697.

Hopkinson, N., Dickens, P., 2003. Analysis of rapid manufacturing using layermanufacturing processes for production. Journal of Mechanical EngineeringScience 217 (C1), 3139.

Hopkinson, N., Hague, R, Dickens, P., 2006. Rapid Manufacturing: An IndustrialRevolution for the Digital Age. John Wiley, Chichester England.

Kruth, J.-P., Levy, G., Klocke, F., Childs, T.H.C., 2007. Consolidation phenomena inlaser and powder-bed based layered manufacturing. CIRP Annals Manufactur-ing Technology 56 (2), 730759.

Kulkarni, P., Marsan, A., Dutta, D., 2000. A review of process planning techniques inlayered manufacturing. Rapid Prototyping Journal 6 (1), 1835.

Levy, G.N., Schindel, R., Kruth, J.P., 2003. Rapid manufacturing and rapid toolingwith layer manufacturing (LM) technologies, state-of-the-art and future per-spectives,. CIRP Annals Manufacturing Technology 52 (2), 589609.

Majchrzak, A., 1988. The human side of factory automation: managerial and humanresource strategies for making automation succeed, 1st ed.. Jossey-Bass Pub-lishers, San Francisco.

Mansour, S., Hague, R., 2003. Impact of rapid manufacturing on design formanufacture for injection moulding. Proceedings of the Institution of Mechan-ical Engineers, Part B: Journal of Engineering Manufacture 217 (4), 453461.

Millen, R., Sohal, A.S., 1998. Planning processes for advanced manufacturingtechnology by large American manufacturers. Technovation 18 (12), 741750.

Munguia, J., de Ciurana, J., Riba, C., 2008. Pursuing successful rapid manufacturing:a users' best-practices approach. Rapid Prototyping Journal 14 (3), 173179.

Noori, H., 1990. Managing the dynamics of new technology: issues in manufactur-ing management. Prentice Hall, Englewood Cliffs N.J.

Petrovic, V., Gonzalez, J.Vicente Haro, Jord Ferrando, O., Delgado Gordillo, J.,Puchades, J. Ramn Blasco, Grian, L. Portols, 2011. Additive layered manu-facturing: sectors of industrial application shown through case studies. Inter-national Journal of Production Research 49 (4), 10611079.

Porter, M.E., 1985. Technology and competitive advantage. Journal of BusinessStrategy 5 (3), 6078.

Ruffo, M., Hague, R.J.M., 2007. Cost estimation for rapid manufacturing simulta-neous production of mixed components using laser sintering. Proceedings ofthe Institution of Mechanical Engineers, Part B: Journal of EngineeringManufacture 221 (11), 15851591.

Ruffo, M., Tuck, C., Hague, R., 2006. Cost estimation for rapid manufacturing - lasersintering production for low to medium volumes. Proceedings of the Institutionof Mechanical Engineers, Part B: Journal of Engineering Manufacture 220 (9),14171427.

Ruffo, M., Tuck, C., Hague, R., 2007. Make or buy analysis for rapid manufacturing.Rapid Prototyping Journal 13 (1), 2329.

Saberi, S., Yusuff, R.M., Zulkii, N., Ahmad, M.M.H.M., 2010. Effective factors onadvanced manufacturing technology implementation performance: a review.Journal of Applied Sciences 10 (13), 12291242.

Santos, E.C., Shiomi, M., Osakada, K., Laoui, T., 2006. Rapid manufacturing of metalcomponents by laser forming. International Journal of Machine Tools andManufacture 46 (1213), 14591468.

Schubert P., Fisher, J., Uwe, L., 2007, ICT and innovation in small companies. In:Proceeding of the 15th European Conference on Information Systems, St.Gallen, June 0709, pp. 12261239.

Skinner, W., 1984. Operations technology: blind spot in strategic management.Interfaces 14 (1), 116125.

Sonntag, V., 2003. The role of manufacturing strategy in adapting to technologicalchange. Integrated Manufacturing Systems 14 (4), 312323.

Strategic Directions, 2008. The next frontier: New technologies will enable cost-effective customization. Strategic Directions 24 (8).

Sun Xiao-lin, Tian Ye-zhuang, Cui Guo-gang, The empirical study on the impact of

advanced manufacturing technology on organisational structure and human

resources management. In: International Conference on Management Scienceand Engineering, 2007. ICMSE 2007, 2007, pp. 15481553.

The Economist, April 2012, Solid print: making things with a 3D printer changes therules of manufacturing, Economist, London.

Thong, J.Y.L., Yap, C.-S., Raman, K.S., 1996. Top management support, externalexpertise and information systems implementation in small businesses. Infor-mation Systems Research 7 (2), 248267.

Voss, C., Tsikriktsis, N., Frohlich, M., 2002. Case research in operations management.International Journal of Operations & Production Management 22 (2), 195219.

Voss, C.A., 1988. Implementation: a key issue in manufacturing technology: theneed for a eld of study. Research Policy 17 (2), 5563.

Welsh, J.A., White, J.F., 1981. A small business is not a little big business. HarvardBusiness Review 59 (4), 1832.

Yin, R., 2003. Case Study Research: Design and Methods, 3rd ed.. Sage Publications,Thousand Oaks, CA.

Zairi, M., 1998. Supplier partnerships for effective advanced manufacturing tech-nology implementation: a proposed model. Integrated Manufacturing Systems9 (2), 109119.

Zmud, R.W., 1984. An examination of push-pull theory applied to processinnovation in knowledge work. Management Science 30 (6), 727738.

S. Mellor et al. / Int. J. Production Economics 149 (2014) 194201 201