Make or buy analysis for rapid manufacturing
The Authors
Massimiliano Ruffo, Wolfson School, Loughborough University, Loughborough, UK
Chris Tuck, Wolfson School, Loughborough University, Loughborough, UK
Richard Hague, Wolfson School, Loughborough University, Loughborough, UK
Abstract
Purpose – The purpose of this paper is to outline how rapid manufacturing (RM) could influence the decision-making process for managers involved in make or buy decisions.
Design/methodology/approach – A literature review on make or buy issues has been carried out and the results of which have been distilled into a number of qualitative considerations. These considerations have been formed into three possible make or buy scenarios: the firm has no experience of rapid prototyping (RP) or RM; the firm already has an RP department; and the firm already has an RM function. In order to analyse the decision further a quantitative approach has been taken, mainly adapted to the last scenario but applicable also to the second scenario. Here, manufacturing cost data has been directly compared with price information from two current RP bureaus. The differences between RM cost and RP price have been studied.
Findings – Strategically, the points analysed were in favour of the make option. Economically, the lack of dedicated RM bureaus and the consequent use of RP costing has further pushed the make or buy decision in favour of make.
Originality/value – There is a lack of work on the implementation of RM as a mainstream manufacturing process. Existing knowledge has begun to look at the use and costs of RM, however, this paper highlights the lack of dedicated RM providers.
Article Type:
Research paper
Keyword(s):
Rapid prototypes; Make or buy decisions.
Journal:
Rapid Prototyping Journal
Volume:
13
Number:
1
Year:
2007
pp:
23-29
Copyright ©
Emerald Group Publishing Limited
ISSN:
1355-2546
Introduction
Competition among automotive brands has become increasingly fierce in recent times (Koo Kim, 1995). In addition, consumers are becoming more discerning and consequently more demanding (Holweg and Greenwood, 2005). Methods of increasing sales have been to offer increased variation of product and getting new models to market faster. Consequently, volumes of production per model are becoming smaller, leading to a move lower volume and niche car production. However, new production and management methods are necessary to support this divergence. Rapid manufacturing (RM) (Hopkinson et al., 2005) is an emergent manufacturing technique that has developed from rapid prototyping (RP) processes and methods (Jacobs, 1996). The modus operandi of RM produces a physical prototype from a three dimensional Computer Aided Design (3D-CAD) model. RM has many differences over conventional manufacturing methods such as destructive machining (milling, grinding, etc.), or forming (injection moulding, casting, etc.). The main distinction is an absence for the need of tools. RM produces components in an additive manner by slicing the 3D-CAD model into a series of discrete layers. These layers are then reconstituted by the RM machine to produce a physical representation of the 3D-CAD model.
Typical RM machines include stereolithography (Harris et al., 2003), laser sintering (Nutt, 1991), fused deposition modelling (Pham and Gault, 1998) and 3D printing (Sachs et al., 1992). The impact of this manufacturing technique is twofold, firstly as tools are not required to produce the physical part, the geometrical complexity can be increased and thus design possibilities increase accordingly (Hague et al., 2004). The second impact is on production management and the costs of production (Hopkinson and Dickens, 2003). Tooling is no longer required and subsequently these costs do not require amortisation across a large number of components in order to make the tooling investment cost effective. Additionally, the time to market for a product can be reduced as lead times for tooling (that can be several months) are removed.
For companies looking at the use of RM as a solution for current manufacturing problems or wishing to take advantage of this emergent technology many pitfalls exist. These pitfalls can be concentrated into three specific areas:
- Manufacturing processes and materials (Custom-Fit, 2005); and
- Design (MADO, 2005).
- Management, organisation and implementation (ManRM, 2005).
All these aspects are subject to on-going research to solve current issues and provide a smooth transition for the technology for full exploitation by industry. This paper brings to attention a critical decision facing companies and institutions considering RM as a production solution. The paper describes the make or buy situation for RM and compares make or buy costs for automotive parts.
RM has the ability to affect all parts of the manufacturing chain. The change from traditional to additive production methods will have implications for the following:
- supply chain management;
- logistics and distribution;
- outsourcing;
- operations; and
- design and product development.
It is not the intention to discuss all these issues here but to concentrate on issues of the outsourcing decision for RM. In many ways the decision to outsource the RM decision runs parallel to those for traditional manufacturing methods and as such a review of current make or buy decision-making studies is necessary.
Literature review
The management of an outsourced process is fundamental for the future growth of a firm (Fine and Whitney, 1996). Murthi (2002) affirms that making the wrong decision for outsourcing can result in cost overruns, project delays or a solution that does not fit business needs. Historically purchasing by organisations had been done on the basis of obtaining a best price, taking into account other factors such as quality and delivery. However, in many cases a significant number of factors such as delivery reliability, technical capability and the financial stability of the supplier were not taken into consideration. Few companies took a strategic view of their make or buy decisions, with many companies deciding to buy rather than make based on a short-term reasons of cost reduction and capacity (Humphreys et al., 2000; Canez et al., 2000). Traditionally, economists viewed vertical integration or vertical control as an attempt to earn monopoly charges by gaining control of input markets or distribution channels (Klein, 2004). McCarthy and Anagnostou (2004) summarise the reasons for outsourcing, which are:
- Exploit external suppliers investments, innovations and capabilities.
- Reduce operating costs, whilst increasing focus on core competencies.
McCarthy and Anagnostou (2004) also discuss the theory of transaction costs. Transaction costs integrate the economic arguments with the managerial and organisational aspects of the business. Transaction costs provide the full cost of providing a product or service.
Klein (2004) concludes his “lessons from empirical studies” affirming that while early empirical work on transaction cost determinations of vertical integration tended to focus on black-and-white distinctions between make or buy, researchers increasingly recognize that a wide variety of contractual and organisational options are available; there are many shades of grey. The literature on make-or-buy hybrids has grown dramatically in the last ten years, while there are fewer studies of mundane issues such as outsourcing versus in-house production.
Benefits and pitfalls of outsourcing
Goodridge (2005), has given opinion based on practical experience of outsourcing an entire product, he advocates subcontracting but only after some expensive mistakes. The advantage of subcontracting includes:
- Reduction of production expenses.
- Lock-in profit; often the real costs of production do not run to plan, while the cost of outsourcing is clear from the outset.
- Never have to say no to a customer; it is an expansion of a product line without spending money on new equipment.
However, drawbacks do exist; subcontracting may cost the customer because it may carry extra costs; so choice of suppliers is important. In addition, outsourcing can put the production schedule into disarray. Anderson and Anderson (2000) expose the negatives connected with sub-contracting. They identify three main problems:
- The diffusion of secret information.
- The direct dependence on suppliers; they can cause delays or other problems.
- Losing the knowledge to integrate what has been outsourced, expanding in the long-term, costs of integration.
Methods to avoid these problems include:
- Taking a long-term view.
- Never outsource core capabilities.
- Consider partial outsourcing of other critical capabilities.
- If outsourcing is a critical capability, use two or more suppliers, in order to maintain price competition. However, this will also increase the opportunity of technology diffusion.
- Joint ventures.
Anderson and Anderson (2000) are not the only ones against the “buy” choice. Anderson and Parker (2002) insist that outsourcing decisions can create a path-dependent outsourcing trap in which a firm experiences higher long-term costs following an immediate cost benefit. There is support for this open idea, that there is no single “best” outsourcing policy (De Jesus, 2001). At times, going the ready made route may make sense while at other times a custom built solution will be better; more often a hybrid solution that uses both will be the best solution.
Different outsourcing strategies
Motivations for outsourcing differ; two different approaches within the automotive sector are illustrated by American and Japanese automobile firms. Fine and Whitney (1996) discuss the differences between two automotive manufacturers. General Motors (US) focus their corporate energies on product design and utilise suppliers for production. In reverse, Japanese car companies (Toyota in particular) centre their strategy on organisational infrastructure (Just In Time, Material Resource Planning, etc.), buying the majority of the components externally. They realise it is difficult to learn how to do everything well and they want to maintain control of as much of the process design and production chain as possible.
But which is the right strategy? Ultimately, the product is what the customers purchase, but manufacturing skills shows up in many areas that customers notice in one way or another, such as quality of fit, introduction of new models, time to delivery, product durability and reliability. As Venkatesan (1992) affirms, US companies in many cases were found to be over-investing in commodity parts and had been neglecting the development of proprietary components, which could have created sources of competitive advantage. There are three main reasons why a company would seek to use suppliers for its business (Fine and Whitney, 1996):
- Capability. The company cannot make the item.
- Manufacturing competitiveness. The supplier has lower cost, faster availability, etc.
- Technology. The supplier's version of the item has higher performance.
On the other hand, two important reasons for not seeking a supplier may be termed “strategic”:
- Competitive knowledge. The item or the skills in producing it is crucial to the success of the company.
- Customer visibility/market differentiation. Clear brand differentiation.
Geoffrey Parker in interviews with Fine and Whitney (1996) distinguished the two main reasons why a company would seek dependency on suppliers, condensing the previous list into only two categories:
- Dependency for capacity.
- Dependency for knowledge.
It is clear that American and Japanese Automotive companies chose to be dependent, respectively, for knowledge and for capacity.
Walker and Weber (1984) studied automobile component procurement and found that uncertainty about production volume raises the probability that a component is made in-house, but “technological uncertainty” measured as the frequency of changes in product specification and the probability of technological improvements, has little effect.
Effect of RM on the outsource decision
As has already been discussed, outsourcing often occurs due to lack of capacity or knowledge of processes within the firm. Clear triggers for outsourcing exist such as (Walker and Weber, 1984):
- price competition;
- lack of capacity;
- skills shortage;
- need to increase responsiveness;
- need to increase quality; and
- reduce time to market.
These triggers change depending on the case in question; however, they cannot be viewed in isolation as a single trigger is likely to affect others in the list. One critical trigger that can often drive the outsource decision is that of cost. Cost to the consumer in a competitive market is set by the external environment of which a company has little control over.
For those companies who have expertise in RP/RM the make or buy decision will be different to those who currently manufacture only by traditional methods. For the traditional enterprise the development of RM as a new manufacturing technique does mean that the skills shortage aspect of outsourcing will be prevalent, in other words companies may not have the necessary skills in-house to perform RM in the short-term. Additionally, for in-house manufacturing via RM, new equipment (machine and support systems) will be necessary. For this reason several make or buy scenarios will exist, due to either a capacity or knowledge gap. As RM becomes common, the knowledge gap is likely to be less of an issue for firms and the capacity gap will prevail in decision making. Different scenarios for outsourcing are necessary to cover the different states of RM knowledge and capacity within firms. Three different scenarios will be defined; these are:
- no experience of RP or RM;
- already has an RP department; and
- already has an RM function.
In terms of knowledge and capacity needs, scenario 1 clearly has a knowledge dependency, while the third scenario has a capacity dependency. Scenario 2 lies between these two extremes (as shown in Figure 1).
In terms of current UK manufacturing organisations, especially for small to medium enterprises, scenario 1 is most relevant. However, there is a fundamental difference with this RM scenario and current work presented in the literature. There is a dilemma before the make or buy decision is started; what manufacturing process should be used?
Scenario 1
This company has traditional production capabilities and no expertise in the RP/RM area. As such the decision process is complicated by the choice of production method. A representation of the decisions open to the company and associated levels of risk with each decision are shown in Figure 2.
In this case the company can make the following decisions shown in Figure 2, it must be emphasised that the move away from traditional production or outsourcing represents the knowledge gap previously discussed and as such RM represents greater risk to the company. A list of the possible options for production (reported in increasing risk):
- manufacture in-house with existing equipment;
- outsource production to traditional manufacturer;
- outsource the product to an RM supplier; and
- manufacture in-house through purchase of RM equipment.
Scenario 2
Scenario 2 represents an organisation with current RP technology and as such knowledge of RP/RM methods. This means that the current equipment may be used for production (if compatible with product specification) otherwise investment in RM machinery or outsourcing RM production is necessary. Figure 3 shows a schematic of the possible decision.
Risk and therefore improbability of choice increase from right to left and top to bottom, for example, the least attractive choice is likely to be that of purchasing new RM equipment. However, other strategic factors such as increased capacity and flexibility may also influence this choice and as such promote the RM case. Conversely, these strategic thoughts may also be true for an organisation going through structural change from a manufacturer to an assembly operation and so the buy option could be more acceptable.
Scenario 3
This case exists where there is RM capacity available to the company. The company has the ability to manufacture in-house or to purchase from RM suppliers. This decision could be based on a cost trigger. Figure 4 shows the two choices available.
In reality the outsourcing decision will not be heavily based on a knowledge dependency and therefore is more likely to be due to capacity. Humphreys et al. (2000) advises when to invest in the make option: this option may be desirable in a case where the technologies involved in the activity are in the embryonic stage and therefore may provide considerable scope for future growth. What technologies match the description more than RM?
Make or buy comparison
Using cost as the key trigger for the make or buy decision key data is require both on the costs of internal manufacture and external purchase costs from suppliers. Using scenario 3 as an example of the decision choices (i.e. that RM is a capable and economic means of production) a comparison of the costs of production and those of two external suppliers will be made. Costs of production are based on a costing model similar to that presented by Hopkinson and Dickens (2003) where fixed and variable costs are attributed to the amount of time taken for the RM machine to complete a build of parts. It is not the intention to detail the components of the costing model here but to use these results to compare and contrast those from current RP bureaus (supplier of RP parts). The costs of in-house production are based on a costing model developed at the Loughborough University and validated by two automotive project partners, which both use RP/RM technologies. Fixed and variable costs are listed in Table I.
Where administration overhead includes costs incurred due to running the enterprise, administrative staff, office space and equipment. Production overhead includes costs incurred due to production, energy and floor space – rent, rates, ancillaries, heat and light. In addition, to these costs capital equipment depreciation has been set at eight years depreciation for PC equipment set to five years. The RM equipment used for production is actually a current RP system; 3D systems Vanguard laser sintering machine. Production of two components, namely, a rotary knob and a timing belt cover has been performed and compared with quotations from 2 RP service bureaus, Bureau 1 and Bureau 2. Images of these components can be found in Figure 5.
The two images shown in Figure 5 have the following geometrical data, shown in Table II.
All production quotes and internal costs are based on laser sintered components manufactured in Duraform PA or its equivalent (PA220). Different volumes of production have been used; 1, 10, 50, 100, 250 and 500 parts. The comparison of internal costs of production (make) versus bureau price (buy) has been performed for these three parts at the specified production volumes.
Results – rotary knob
The comparison of make or buy values for the rotary knob are shown in Figure 6.
The initial high costs for production of single and low volumes of parts occur due to set up and variable costs being amortised over a small amount of parts. As is shown these costs quickly drop off to provide a stable part cost above a volume of approximately 100 parts.
Timing belt cover
The costs of production and purchase for the timing belt cover are shown in Figure 7.
The costs associated with make or buy are again presented with the now familiar higher costs of production for a single component, costs become stable at a volume of ten components for the make option and Bureau 2. However, the make figure from Bureau 1 is stable at all volumes (from a single part to 500).
Discussion
The nature of RM production is based on additive, parallel production, i.e. parts can be produced in parallel with each other as long as there is sufficient space in the build envelope of the machine. In other words different parts can be built simultaneously on a single machine. This begins to explain the drop in price/cost for the parts with increasing volume as the costs of production are split between the multiple parts in the build envelope. The stable area of the three graphs is where the build volume has been filled to capacity and as such the addition of parts is no longer possible. The volumes at these stable values occur are shown in Table III.
As RM production is additive in nature more parts (or larger parts) in the build envelope take longer to build. Thus, for large parts the time for manufacture increases and the costs of this production are amortised over fewer parts and therefore are more expensive than smaller components.
The relationship between volume and cost/price for all three options (make and supplier prices) follow similar trends as such it is possible to conclude that the make cost model is similar to that being used by the suppliers and therefore valid. However, the graphs shown in Figure 6 and, 7 all show (for the stable cost/price) a significant saving for make when compared to buy.
Cost saving: make vs buy
The cost savings of make vs buy are shown below in Table IV, the percentage saving is shown for the make option over the two components used in the comparison and the two suppliers used.
The gap of costs of production vs supplier's prices can be explained in two ways:
- profit margin; and
- extra costs of the organisation.
The cost of production shown for make does not include any margin for profit. Secondly, the supplier organisation will have extra overheads to account for in their sales (for example, management salaries) and for a make or buy decision to be applicable to a specific company these extra costs will need to be included in the make figure as an overhead.
However, the empirical studies of make or buy shown in this paper firmly point to the make option being preferable where cost is the primary driver. In addition, it must be stated that the costs of supplier's do not include logistics costs and these would need to be added as well as costs of stock holding warehousing, etc. The costs of stock do not necessarily apply to the make option as the flexibility afforded by RM means that manufacturing can be scheduled so that parts are produced as and when required.
Conclusions
In conclusion, this paper has outlined some of the key criteria when considering the make or buy decision. These decisions are based on a number of factors, cost, capacity, knowledge, response and quality. These factors are applicable to the embryonic RM process. However, the flexibility of the RM process means that any part can be produced (with suitable 3D-CAD file). This changes the focus of the analysis thus some of the factors previously mentioned. For example, the quality of the final part will be dependent on the CAD model and not necessarily on the production process. RM gives the opportunity to make varied parts at the same time and as such the capacity of the RM machine can be utilised by numerous different parts.
In terms of make or buy, the comparison of the two parts show unanimously that the make option was preferable in terms of cost. Indeed, savings greater than 50 per cent over prices from service providers was evident. As the machines and materials are likely to be the same if not identical then there should also be no degradation in quality. What is clear is that purely based on costs the make option appears to be preferable. In addition, the make option also scores in terms of convenience as no logistics are required for delivery and scheduling can mean that stock holding is reduced/eliminated. The costs of purchasing from the Bureaus shows that the RM concept has not infiltrated these organisations yet, indeed the prices quoted by the Bureaus are more akin to one off prototypes. This study has shown that if capacity of RM processes can be filled, the costs of parts should radically decrease. With the increased use of RM this should lead to reductions in costs (due to economies of scale) for the purchasing of both processes and materials.
Figure 1Supplier dependency in different scenarios
Figure 2Manufacturing and outsource choices for a company with no RM experience
Figure 3Possible choices for a company with RP experience
Figure 4RM organisation make or buy?
Figure 5Images of parts used in make or buy analysis
Figure 6Manufacture and purchase costs for rotary knob
Figure 7Costs of production and purchase for timing belt cover
Table IFixed and variable costs for RM production
Table IIGeometrical data for two components used in make or buy analysis
Table IIIVolume where costs/price stabilise
Table IVSaving for make vs buy
References
Anderson, E., Anderson, M. (2000), "Are your decision today creating your future competitors? Avoiding the outsourcing trap", The Systems Thinker, Pegasus Communication, Waltham, MA, September, .
Anderson, G., Parker, G. (2002), "The effect of learning on the make/buy decision", Production and Operations Management, Vol. 11 No.3, pp.313.
Canez, L., Platts, K., Probert, D. (2000), "Developing a framework for make-or-buy decisions", International Journal of Operations & Production Management, Vol. 20 No.11, pp.1313-30.
Custom-Fit (2005), "Framework 6 project", Custom-Fit, Edmonton, Project No. 507437-2, .
De Jesus, A. (2001), "Making the build-or-buy decision: a thorough analysis before you decide", Planning and Management, Tech News (supported by IBM), New York, NY, available at: www.uwnyc.org/technews/pf_v5_, .
Fine, C., Whitney, D. (1996), Is the Make or Buy Decision and Core Competence?, MIT Center for Technology, Policy and Industrial Development, Cambridge, MA, February, .
Goodridge, M. (2005), "Sub-contracting the make or buy decision", Screen Printing Magazine, available at: www.screenweb.com/garment/cont/subcontacting0700.html, .
Hague, R., Mansour, S., Saleh, N. (2004), "Material and design considerations for rapid manufacturing", International Journal of Production Research, Vol. 42 No.22, pp.4691-708.
Harris, R.A., Hague, R.J.M., Dickens, P.M. (2003), "Crystallinity control in parts produced from stereolithography injection mould tooling", Proceedings of the Institution of Mechanical Engineers part L - Journal of Materials-design and Applications, Vol. 217 No.4, pp.269-76.
Holweg, M., Greenwood, A. (2005), "Product variety life cycles and rate of innovation – trends in the UK automotive industry", available at: www.3daycar.com (accessed 29 April)., .
Hopkinson, N., Dickens, P. (2003), "Analysis of rapid manufacturing – using layer manufacturing processes for production", Proceedings of Institution of Mechanical Engineers Part C Journal of Mechanical Engineering Science, Vol. 217 No.1, pp.31-9.
Hopkinson, N., Hague, R., Dickens, P. (2005), Rapid Manufacturing – An Industrial Revolution for the Digital Age, Wiley, New York, NY, .
Humphreys, P., Lo, V., McIvor, R. (2000), "A decision support framework for strategic purchasing", Journal of Materials Processing Technology, Vol. 107 pp.353-62.
Jacobs, P.F. (1996), Stereolithography and Other RP&M Techniques from Rapid Prototyping to Rapid Tooling, SME, Dearborn, MI, .
Klein, P.G. (2004), "The make-or-buy decision: lessons from empirical studies", University of Missouri, Columbia, MO, CORI Working Paper No. 2004-07, .
Koo Kim, C. (1995), "Brand popularity and country image in global competition: managerial implication", Journal of Product and Brand Development, Vol. 4 No.5, pp.21-33.
McCarthy, I., Anagnostou, A. (2004), "The impact of outsourcing on the transaction costs and boundaries of manufacturing", International Journal of Production Economics, Vol. 88 No.1, pp.61-71.
MADO (2005), "Materials analysis and design optimisation", EPSRC, Swindon, Project No. IMCRC75, Engineering Physical Science Research Council Project, .
ManRM (2005), "Management, organisation and implementation of rapid manufacturing", Foresight Vehicle, London, Department of Trade and Industry Foresight Vehicle Project, .
Murthi, S. (2002), "Build versus buy - making the right decision", Free Information Builder's Whitepaper, available at: www.developer.com/java/other/article.php/1488331 (accessed 31 January), .
Nutt, K. (1991), "The selective laser sintering process", Photonics Spectra, Vol. 25 No.9, pp.102-4.
Pham, D.T., Gault, R.S. (1998), "A comparison of rapid prototyping technologies", International Journal of Machine Tools & Manufacture, Vol. 38 No.10/11, pp.1257-87.
Sachs, E., Cima, M., Williams, P., Brancazio, D., Cornie, J. (1992), "3-dimensional printing – rapid tooling and prototypes directly from a CAD model", Journal of Engineering for Industry, Vol. 114 pp.481-8.
Venkatesan, R. (1992), "Strategic outsourcing: to make or not to make", Harvard Business Review, Vol. 70 No.6, pp.98-107.
Walker, G., Weber, D. (1984), "A transaction cost approach to make or buy decisions", Administrative Science Quarterly, Vol. 29 pp.113-41.
Corresponding author