Is 3D printing a case of disruptive innovation?

(1)

Is 3D printing a case of disruptive

innovation?

Author: Balázs Kisgergely

Student number: 10604154

Institution: University of Amsterdam

Program: Msc. in Business Studies - Entrepreneurship & Innovation

Supervisor: dhr. dr. G.T. (Tsvi) Vinig

2

nd

Supervisor: dhr. drs. A.C.C. Gruijters

Date: 17 May, 2015

(2)

Statement of originality

This document is written by Balázs Kisgergely who declares to take full responsibility for the contents of this document.

I declare that the text and the work presented in this document is original and that no sources other than those mentioned in the text and its references have been used in creating it.

The Faculty of Economics and Business is responsible solely for the supervision of completion of the work, not for the contents.

(3)

Abstract

The Disruptive Innovation Theory, first introduced by Clayton Christensen, has created a significant impact on management practices, triggered rich academic debate and contributions from several authors. Through their research and examples, various industries such as the personal computer (PC), mobile phones and automotive have received tremendous attention.

However, the application of the theory for ex ante predictions in still rudimentary. The main contribution of this paper is a creation of a complementary framework to test the specific characteristics of disruptive innovations. An analysis of 3D printing as a technology and an industry is provided, focusing on the benefits and drawbacks of the process as well as its current and possible future applications. Subsequently, the aforementioned framework is applied and contrasted with the technology. To create a clear distinction, an analysis on the differentiation of radical and disruptive innovation; and the market effects of sustainable, radical and disruptive innovation is provided.

In the course of research, 3D printing was proven to be in possession of all the key attributes defined in the framework, therefore it could be considered a disruptive innovation.

(4)

Introduction

The process of 3D printing exists since the 1980s, but through recent developments it has been gaining traction and is currently considered an emerging technology that could change design and manufacturing in the near future. Currently, the main area where the process is utilized is rapid prototyping. 3D printing allows its users to create prototypes significantly faster and cheaper than traditional methods (plastic injection molding, clay). Furthermore, changing the prototypes does not incur any additional costs other than printing the artifact again. The technology is already a disruptive force in prototyping, but center of attention is rapidly shifting towards the new “holy grail”: manufacturing. 3D printing is currently competitive with traditional manufacturing methods on relatively small production runs, limited by size and the availability of materials. These limitations are on course to decrease significantly, rapidly closing the gap possibly in the near future. Given the process continues or even accelerates, the implications are limitless: 3D printing would make product development and manufacturing of final products significantly faster, more efficient and affordable; it would change industries and possibly the economic and political landscape of the World. Naturally, such high expectations imply that there is significant hype surrounding the technology, especially in the last few years.

During this research I will explore advantages and disadvantages of 3D printing, focusing on the areas that require improvement to facilitate growth. I intend to show current applications in different industries and explore where it could become a prominent technology in the near future. Using a framework based on the thorough study of academic literature, I will analyze if 3D printing possesses the specific characteristics of a disruptive innovation. Furthermore, focusing on the near future, I will show the foreseeable effect of the technology on our society.

Research question

This research is conducted to gain more insight into 3D printing as a technology and its effects on product development, various industries and ultimately on our economy. This study will seek an answer to the following research question:

(7)

In order to develop a clear understanding of the subject and to answer the research question, the following sub-questions will be addressed in the course of this study:

What are the specific attributes of 3D printing that could make it a disruptive innovation?

In which fields does it require further improvement to achieve mainstream adoption?

What are the (niche) markets where 3D printing may gain ground?

Academic relevance

Schumpeter is considered the patron saint of innovation economics, especially with his classic 1942 book Capitalism, Socialism and Democracy. Today, in the 21st century, when technology is advancing probably at the fastest rate in the history of mankind, innovation is the core of each successful company and economy.

Numerous authors have researched innovation, the process of innovation (Knudsen & Swedberg), disruptive innovations (Christensen, Govindarajan & Kopalle, Schmidt & Druehl) and the dimension of innovation (Henderson & Clark) and the industry cycles (Klepper, Williamson, Miller). Through their research and examples, various industries such as the personal computer (PC), mobile phones and automotive have received tremendous attention.

On the other hand, these theories have not been applied to the process of 3D printing. The main contribution of this paper and the research gap it intends to fill is to apply these theories to the development and technology of 3D printing. I aim to find the specific characteristics and initial capabilities that make it superior even at its current state than alternative technologies at their best and the specific markets that can utilize these capabilities. Furthermore, as I mentioned, there is significant hype surrounding 3D printing. I intend to see beyond the hype and through the rigorous use of literature and empirical examples, find out what is well founded, draw valid assumptions and give indication for the near future.

(8)

Practical relevance

Novel processes and 3D printed products have been receiving increasing coverage in the news, but most of our society still does not have a clear understanding of what the technology implies. This research provides an up-to-date overview of the industry, technology and applications of 3D printing.

Even though 3D printing is already a multi-billion dollar industry, the market for 3D printed products and proprietary applications are still at their infancy. A large number of forecasts predict that this will be a new frontier of innovation exhibiting exponential growth in the near future. Being aware of the inherent capabilities, limits and the progress of the technology may influence entrepreneurs, companies, business decisions and the quality of innovation. As they are the driving forces behind our economy, their decisions will ultimately influence us all. This research does not deliver clear solutions, but it hopes to shed light on characteristics and areas where 3D printing is creating disruption and innovation.

Structure of the thesis

The structure of the thesis is as follows. First, the review of core literature will be provided, focusing on the development in the innovation literature, disruptive innovation theory, technology life-cycles, firm advantages and barriers of entry as well as 3D printing as a technology itself. Based on the literature, a framework will be created in order to answer the research question. Then, a methodology chapter will follow, where the data collection and analysis method will be explained.

In the next section the research findings will be discussed. First, the advantages and disadvantages of 3D printing will be introduced, as well as the current applications the technology. Second, an overview of the industry will be provided. Next is an analysis of the main findings of this research, where the framework will be tested to find out if 3D printing is a case of disruptive innovation. For further clarification, radical and disruptive innovation will be differentiated. In the discussion the answer to the main research question will be evaluated and the market effects of sustainable, radical and disruptive innovation will be clarified. Following an outlook for the future, conclusions will be drawn, limitation of the study will be explained and suggestions for future research will be provided.

(9)

Review of core literature

Innovation

One of the first authors who contributed greatly to the literature on innovation was Schumpeter. The author defines innovation as the fundamental impulse that sets and keeps the capitalist engine in motion in his groundbreaking book the Business Cycles:

“Technological change in the production of commodities already in use, the opening up of new markets or of new sources of supply, Taylor-ization of work, improved handling of material, the setting up of new business organizations such as department stores—in short, any "doing things differently" in the realm of economic life”

Schumpeter also distinguished invention as discovery of something new and innovation as putting what has been learned to work commercially.

Knudsen & Swedberg (2009) builds on Schumpeter’s work to define the process of innovation: “An order exists in some part of the economy, firm or an industry. An entrepreneur emerges and suggests a different way of doing things which, when successful, leads to high profit (entrepreneurial profit in Schumpeter’s terminology)” This part is described as the “competing up” stage reaches its plateau with temporary monopoly of the innovator. Subsequently, the profit of the innovator attracts other actors follow, after them still others in increasing number, in the path of innovation, which becomes progressively smoothed for successors by accumulating experience and vanishing obstacles. This “competing down”, as Schumpeter calls it, also leads to profit, although less than what the original innovator can claim. The novel pattern spreads further, as more and more imitators appear and try to get into the game. At this stage a new order for best practices in some area of the economy has begun to emerge. The process can be separated into two distinct phases, during the first of which the system, under the impulse of entrepreneurial activity, draws away from an equilibrium position, and during the second of which it draws toward another equilibrium position. Furthermore, the theory predicts that profit will be the highest when a new combination successfully breaks with the old order, and that profit will fall as the new order gradually establishes itself.

(10)

Schumpeter calls this process the creative destruction “that incessantly revolutionizes the economic structure from within, incessantly destroying the old one, incessantly creating a new one.” In Schumpeter's vision of capitalism, innovative entry by entrepreneurs was the disruptive force that sustained economic growth, even as it destroyed the value of established companies and laborers that enjoyed some degree of monopoly power derived from previous technological, organizational, regulatory, and economic paradigms.

In terms of economics, “physical marginal productivity of every factor must, in the absence of innovation, monotonically decrease. Innovation breaks off any such "curve" and replaces it by another, which, again except for indivisibility, displays higher increments of product throughout, although, of course, it also decreases monotonically. Or if we take the Ricardian law of decreasing returns and generalize it to cover industry as well, we can say that innovation interrupts its action, which again means that it replaces the law that had so far described the effects of additional doses of resources by another one. In both cases transition is made by a jump from the old to the new curve, which now applies throughout and not only beyond that output which had been produced before by the old method” (Schumpeter, 1939).

Regarding costs, the old total or marginal cost curve is destroyed and a new one put in its place each time there is an innovation. Whenever a given quantity of output costs less to produce than the same or a smaller quantity did cost or would have cost before, we may be sure, if prices of factors have not fallen, that there has been innovation somewhere (Schumpeter, 1939).

Furthermore, Schumpeter defines three significant attributes of innovations. First, innovations do not remain isolated events, nor they are evenly distributed in time; on the contrary they tend to cluster, to come about in bunches, simply because first some, and then most, firms follow in the wake of successful innovation. Second, innovations are not at any time distributed over the whole economic system at random, but tend to concentrate in certain sectors and their surroundings. When some innovation has been successfully carried into effect, the next wave is much more likely to start in the same or a neighboring field than anywhere else. Third, major innovations hardly ever emerge in their final form or cover in one throw the whole field that will ultimately be their own.

(11)

The next step in the evolution of innovation literature is Foster’s (1986) and Christensen’s (1992, 1997) works on the technology S-curve. The technology S-curve is often used to predict whether an emerging technology is likely to supplant an established one. “The operative trigger is the slope of the curve of the established technology. If the curve has passed its point of inflection, so that its second derivative is negative (the technology is improving at a decreasing rate), then a new technology may emerge to supplant the established one” (Christensen, 1997)

Mainstream studies divide technological innovations into two types, but use different terminologies for different stages in history. There are two general classes of technologies: (1) revolutionary, discontinuous, breakthrough, radical, emergent or step-function technologies; (2) evolutionary, continuous, incremental or ‘nuts and bolts’ technologies (Yu & Hang, 2010). Each category tends to fare well on its on field, but their application is still limited.

Subsequent notable work on innovation can be attributed to Henderson & Clark (1990), who make the distinction along two new dimensions: knowledge of the components and knowledge of the linkage between them, called architectural knowledge.

Radical and incremental innovations are the extreme points along both dimensions. “Radical innovation establishes a new dominant design and, hence, a new set of core design concepts embodied in components that are linked together in a new architecture. Incremental innovation refines and extends an established design. Improvement occurs in individual components, but the underlying core design

(12)

There are two further types of innovation in the model, innovation that changes only the core design concepts of a technology and innovation that changes only the relationships between them. The former is a modular innovation, which is an innovation that changes a core design concept without changing the product's architecture. On the other hand, architectural innovation changes a product's architecture but leaves the components, and the core design concepts that they embody, unchanged.

The previous theories were still found inadequate to explain anomalies such as the hard disk drive industry in the 1980’s, which triggered Christensen (1997) to develop the Disruptive Innovation Theory, differentiating innovations as either sustaining or disruptive. The former improve the performance of established products, along the dimensions of performance that mainstream customers in major markets have historically valued.

On the other hand a “disruptive technology or disruptive innovation is an innovation that helps create a new market and value network, and eventually goes on to disrupt an existing market and value network”. It explains a phenomenon by which an innovation transforms an existing market or sector by introducing simplicity, convenience, accessibility, reliability and affordability where complication and high cost are the status quo. Initially, a disruptive innovation is formed in a niche market that may appear unattractive or inconsequential to industry incumbents, but eventually the new product or idea completely redefines the industry.

“This happens because the pace of technological progress in products frequently exceeds the rate of performance improvement that mainstream customers demand or can absorb. As a consequence, products whose features and functionality closely match market needs today often follow a trajectory of improvement by which they overshoot mainstream market needs tomorrow. And products that seriously underperform today, relative to customer expectations in mainstream markets, may become directly performance-competitive tomorrow” (Christensen, 1997).

Subsequently, Christensen refined his theory to classify two kinds of disruptive innovations, namely new-market disruptions and low-end disruptions (Christensen & Raynor 2003). New-market disruptions compete with “non-consumption” because new-market disruptive products are so much more affordable to own and simpler to

(13)

use that they enable a whole new population of people to begin owning and using the product, and to do so in a more convenient setting. On the other hand disruptions that take root at the low end of the original or mainstream value network are called low-end disruptions.

Several authors have contributed to define disruptive innovations in detail. Govindarajan and Kopalle (2006) introduce low-end and high-end disruptions, provide a more general view of disruptiveness of innovations, and explores beyond the case of low price/low performance, through examples such as the cellular phone. Schmidt and Druehl’s (2008) complementary framework suggests, that when an innovation diffuses from the low end upward toward the high end, a pattern called low-end encroachment, the incumbent may be tempted to overlook its potential impact. The authors refine new-market disruption into two types, fringe-market low-end encroachment and detached-market low-low-end encroachment; and differentiate immediate low-end encroachment.

The predictive use of the Disruptive Innovation theory has been argued and challenged by many scholars (Barney 1997, Daneels 2004, Tellis 2006). Schmidt (2004) has made efforts to address the predictive use of the theory and provides tools to firms to assess whether a market is ripe for disruption. If the part-worth curves for traditional attributes show overshoot of customer needs while they are correspondingly underserved in secondary attributes, then a new lower-cost product might conceivably be successful by emphasizing the secondary attributes while sacrificing performance on the traditional attributes. If the new product’s performance could be improved over time with regard to these traditional attributes, eventually the new product could be expected to move up market, reflecting low-end encroachment. Christensen has suggested that incumbent companies are less likely to develop disruptive innovations as they focus on their current customers’ needs and thus sustaining innovation. Elaborating on this concept, Govindarajan and Kopalle (2006) make ex ante predictions about the type of firms likely to develop disruptive innovations. If Firm A’s ‘willingness to cannibalize’ is higher than that of Firm B, Firm A would appear more likely than Firm B to develop disruptive innovations (Druehl and Schmidt 2008; Govindarajan and Kopalle 2006).

(14)

According to Schumpeter, “a system - any system, economic or other - that at every given point of time fully utilizes its possibilities to the best advantage may yet in the long run be inferior to a system that does so at no given point of time, because the latter’s failure to do so may be a condition for the level or speed of long-run performance.” In other words, a technology might be dominant at any given time, but if it has reached its full potential or if it’s advancing at a decreasing rate, an opportunity arises for other technologies to converge and eventually surpass it. If an attribute, or some attributes of a technology are fundamentally superior to those of the incumbent technology at its best, but other dimension valued by the mainstream market are not yet developed, then it’s only a matter of time and development until a radical and disruptive innovation.

The exploration and analysis of these decisive parameters will be a key to decide if 3D printing is truly a disruptive innovation that will change and create new industries and markets.

Life-cycles and competitive strategy

In order to understand how a technology develops and how we can differentiate specific stages, we shall look at industry and technology (life-)cycles. Research scholars widely depict industry cycles in three stages. : An initial, exploratory, embryotic stage; an intermediate or growth development stage; and a mature stage (Williamson, 1975; Klepper, 1997).

In the initial, exploratory or embryonic stage, market volume is low, uncertainty is high, the product design is primitive, unspecialized machinery is used to manufacture the product and it is marketed through a variety of exploratory techniques. Volume is typically low and a high degree of uncertainty characterizes business experience. Many firms enter and competition based on product innovation is intense.

In the second, intermediate or growth stage, market definition is sharpened, output grows rapidly in response to newly recognized applications and unsatisfied market demands, the design of the product begins to stabilize, product innovation declines, and the production process becomes more refined as specialized machinery is substituted for labor. Entry slows and a shakeout of producers occurs. Klepper and Miller (1995) uses the shakeout phase as the stage where the greatest number of exits

(15)

from the industry occurs, which typically takes place well after the number of new entrants into the industry has declined. The combination of the drop in the number of new entrants along with the high number of exiting firms during the shakeout phase leads to a decline in the total number of firms during the mature and declining stage of the life cycle.

Stage three, the mature stage, corresponds to a mature market. Markets may continue to grow, but do so at a more regular and predictable rate, the number of new entries declines further, market shares stabilize, innovations are less significant, and management, marketing, and manufacturing techniques become more refined (Williamson, 1975; Klepper, 1997).

Klepper (1992) has identified three distinct patterns of innovative activity with respect to a product’s life cycle. The first is that innovative activity tends to be the greatest during the earliest phases of the life cycle. Second, during the early and growth stages, “the most recent entrants account for a disproportionate share of the major product innovations that are introduced”. Finally, as the stage of the life cycle evolves towards maturity, there is a distinct and pronounced shift in the locus of innovative activity, away from new entrants and towards established enterprises.

Firm advantages and barriers to entry

So far we have an understanding of how the innovation process works and how it can be divided into specific stages. In the followings, we shall explore which firms tend to have advantages at different innovation processes and how entry barriers influence companies at these stages.

Academic literature provides countless examples of innovations that were deemed irrelevant by incumbent firms, but eventually radically changed industries. “Established firms tend to be good at improving what they have long been good at doing, and that entrant firms seem better suited for exploiting radically new technologies, often because they import the technology into one industry from another, where they had already developed and practiced it.” (Christensen, 1997) Furthermore, the maturity and size of the firm can be of significance as well. “In the initial stages of development of new product industries, uncertainty about the new technology is

(16)

particular advantages in R&D” (Klepper, 1997). The absence of size advantages is coupled with large prospective returns for successful innovations, which attract many entrants. As the field on which the innovations are carried out expands, research becomes more specific and subdivided into well-defined tasks, yielding scale economies in R&D to specialized firms. The accumulated know-how of a firm also rises over time, increasing the R&D required by entrants to duplicate the knowledge of incumbents. Both of these factors raise entry barriers. On the other hand, spillovers of knowledge beneficial to rivals can be both significant and difficult to take into account through markets unaided by trademark, copyright, or patent (Demsetz, 1982). The decline of profit margins and the increase in competition forces firms to compete on efficiency as well. Eventually knowledge gets codified, decreasing R&D entry barriers, but profit margins are sufficiently compressed to make entry unattractive and competition shifts from technology to efficiency and price.

According to Porter, barriers to entry are “advantages that incumbents have relative to new entrants”. There are seven major sources: supply-side economies of scale, demand side benefits of scale, customer switching cost, capital requirements, incumbency advantages independent of size, unequal access to distribution channels and restrictive government policy (Porter, 2008). Studies have shown, that some market barriers are clearly more important than others, cost advantages (supply-side economies of scale) and capital requirements were found to be the most relevant (Karakaya et al, 1989). Supply-side economies of scale is when firms that produce at larger volumes enjoy lower costs per unit because they can spread fixed costs over more units, employ more efficient technology, or command better terms from suppliers. Capital requirements are the need to invest large financial resources in order to compete can deter new entrants. 3D printing might actually have the largest effect on these two barriers. It can decrease capital requirements, as the costs of development and testing are decreased. Furthermore, supply-side economies of scale can be mitigated as well, if the manufacturing of a product is done using the technology.

Literature on 3D printing

Traditional manufacturing – manufacturing comes from the French word for “made by hand” - has fueled the industrial revolution that has enabled our world today, yet it

(17)

contains inherent limitations that brings about the need for new approaches. This etymological origin is no longer appropriate to describe the state of today’s modern manufacturing technologies. Casting, forming, molding, and machining are complex processes that involve tooling, machinery, computers, and robots (Campbell et al, 2011).

In manufacturing, and machining in particular, “subtractive” refers to more traditional methods. The term subtractive manufacturing is a retronym developed in recent years to distinguish it from newer additive manufacturing techniques. Similar to a child cutting a folded piece of paper to create a snowflake, these technologies are “subtractive” techniques, in which objects are created through the subtraction of material from an object. Final products are limited by the capabilities of the tools used in the manufacturing processes (Campbell et al, 2011). Although fabrication has included methods that are essentially "additive" for centuries (such as joining plates, sheets, forgings, and rolled work via riveting, screwing, forge welding, or newer kinds of welding), it did not include the information technology component of model-based definition. Machining (generating exact shapes with high precision) has typically been subtractive, from filing and turning to milling, drilling and grinding. (ASTM F2792-10, 2010)

The term additive manufacturing (AM) refers to technologies that create objects through sequential layering.

Additive Manufacturing is the “process of joining materials to make objects from 3D model data, usually layer upon layer, as opposed to subtractive manufacturing methodologies.” [Standard Terminology for Additive Manufacturing Technologies, ASTM F2792-10, June 2010.]

Additively manufactured objects can be utilized in several fields throughout product development and life-cycle, from pre-production (i.e. rapid prototyping) to full-scale production (i.e. rapid manufacturing), in addition to tooling applications and post-production customization.

The term “3D printing” itself is commonly referred to various additive manufacturing technologies in literature. Generally, additive manufacturing is used for industrial

(18)

applications and 3D printing for the consumer market, but the lines often intertwine. As there is no clear distinction, both terms will be used in this paper interchangeably. The technology has been around for decades; Charles Hull of 3D Systems Corp created the first working 3D printer in 1984. A large number of additive manufacturing processes are available, differentiated by the way the layers are deposited to create parts and the materials that can be used.

Some methods melt or soften material to produce the layers, e.g. selective laser melting (SLM) or direct metal laser sintering (DMLS), selective laser sintering (SLS), fused deposition modeling (FDM), while others use UV (ultraviolet) laser to harden a photosensitive polymer, e.g. stereolithography (SLA). With laminated object manufacturing (LOM), thin layers are cut to shape and joined together (e.g. paper, polymer, metal). Recent developments in the synthesis of end-use products allow for increasing numbers of materials to be used simultaneously, similar to an inkjet printer with various colors, but with different materials such as plastics, metals or ceramics. Each method has its own advantages and drawbacks the main considerations in choosing a machine are generally speed, cost of the 3D printer, cost of the printed prototype, cost and choice of materials and color capabilities. (Wohlers, 2005)

The additive manufacturing process itself begins with a 3D model of an object, usually created by computer-aided design (CAD) software or scan of an existing artifact. Specialized software slices this model into cross-sectional layers, creating a computer file that is sent to the AM machine. The AM machine then creates the object by forming each layer via the selective placement (or forming) of material (feedstock). Feedstock refers to stock material that is fed into a printer (Pearce et al. 2010), whether in the form of plastic; resins; super alloys, such as nickel-based chromium and cobalt chromium; stainless steel; titanium; polymers or ceramics. The AM machine solidifies material feedstock layer by layer with micro-millimeter detail. Because of the visibility of the layering in many cases of 3D printing, some form of minor ʻfinishingʼ is required for the object to be aesthetically satisfactory to likely consumers/users.

Before computer-aided manufacturing the machining or milling of an object from a block of material in a tooled factory setting involved a subtractive ʻcuttingʼ process, first by using blades and eventually lasers. In 3D printing all material can be

(19)

accounted for, including the ʻscaffoldingʼ for objects that cannot support themselves, which is printed alongside the object itself so reducing human intervention in the production process. (Birtchnell et al., 2013)

Due to the novelty of the technology and the fast pace of development, the literature on 3D printing is not yet well developed. The technology is thought to be in the “Apple I stage”; indicating the similarity to personal computing’s development in the early 80’s. As I will explain on the following pages, there are several trajectories and opportunities that are conceivable for the future.

Disruptive Innovation Framework

An examination of literature on the Disruptive Innovation Theory by (i) Abernathy and Clark (1985), (ii) Christensen and Bower (1996), (iii) Christensen (1997), (iv) Adner (2002), (v) Charitou and Markides (2003), (vi) Christensen & Raynor (2003), (vii) Gilbert (2003) and (viii) Govindarajan & Kopalle (2006) suggests that disruptive innovations have the following four characteristics in common:

• The innovation underperforms on the attributes mainstream customers value (i, ii, iii, iv, vi, viii)

• The new features offered by the innovation are not valued by the mainstream customers (i, ii, iii, v, vi, vii, viii)

• The innovation initially attracts an emerging, or an insignificant, niche market, thus limiting the profit potential for incumbents. The disruptive product may offer a higher per-unit margin, the perceived lower market size makes the profit potential appear limited. (v, vi, vii, viii)

• Over time, further developments improve the innovation’s performance on the attributes mainstream customers value to a level where the innovation begins to attract more of these customers. (ii, iii, iv, v, vi, viii)

(20)

Using Christensen’s (2003) original framework as the bases and analyzing the relationships among these four characteristics, I identified four separate phases in the development of disruptive innovations. Initially, a disruptive innovation is triggered by two distinct conditions. It underperforms existing technologies on the attributes that mainstream customers value whereas the new features that it offers are not valued by these customers. On the other hand, these new features are valuable enough for some niche segments where the low performance in main attributes has less importance. Consequently, the innovation takes off in these niche segments. As the technology develops and its performance improves along the attributes that the mainstream customers value, more and more industries start experimenting with it. Increasing use and investment into its development accelerates the pace of improvement and the innovation becomes appealing for mainstream use. Through the development phase, the innovation overcomes the initial shortcomings where it lagged behind existing technologies, while the new features it offers makes it superior to them. Eventually the new innovation disrupts prevailing value networks, replacing existing technologies in more and more fields and industries.

It is essential to differentiate disruptive and radical innovations. The radicalness of innovations refers to the extent an innovation is based on a substantially new technology relative to existing practice, while the disruptiveness of innovations refers to the extent an emerging customer segment, and not the mainstream customer

(21)

segment, sees value in the innovation at the time of introduction, which over time disrupts the products mainstream customers. The radicalness is a technology-based dimension of innovations, and the disruptiveness is a market-based dimension (Govindarajan & Kopalle, 2006).

Disruptive innovations can involve either radical technologies (high end) or incremental technologies (low end). Disk drives or low-cost, no-frills airlines are examples of disruptive innovations that are less radical in nature, while digital cameras relative to analog cameras, cellular phones relative to wired phones, iPod relative to Walkman, and electronic calculators relative to slide rules are examples of disruptive innovations that are more radical (Christensen, 1995). The innovation diffusion literature suggests that technologically radical innovations primarily appeal to the early-adopter category at the time of product introduction and over time appeal more to the mainstream market (Rogers, 2003).

Govindarajan & Kopalle (2006) suggests four key aspects that distinguish disruptive innovations from those that are radical but not disruptive.

• First, the niche segments that initially find disruptive innovations attractive can be either the more price-sensitive segment (in the case of low-end disruptions) or the less price-sensitive segment (in the case of high- end disruptions); however, the early-adopter segment, which finds the radical (but not disruptive) innovations initially attractive, is less price sensitive (Rogers, 2003).

• Second, disruptive innovations underperform on the dimensions mainstream customers value, whereas radical innovations that are not disruptive perform well on the dimensions mainstream customers value. Customers do not initially adopt the radical innovation because of, for example, complexity, compatibility, and risk (Rogers, 2003).

• Third, disruptive innovations introduce new dimensions of benefits the niche segment, but not the mainstream customers, finds more attractive. On the other hand, radical innovations that are not disruptive focus on the same dimensions of benefits mainstream customers value (Christensen, 1997).

(22)

• Fourth, although disruptive innovations pose a dilemma to the incumbents, the introduction of non-disruptive radical innovations do not pose a dilemma for incumbents, because such firms know early adopters will eventually spread the benefits by word of mouth to the rest of the market.

While by now research scholars have reached a general agreement on the basic characteristics of disruptive innovations, the differentiation of distinct types of disruptions is still debated. Christensen originally differentiated new-market and low-end disruptive innovations, which was further developed in his later work (Christensen et al, 2004).

Group Identifier What could happen

Signals

Nonconsumers People who lack the

ability, wealth or access to conveniently and easily accomplish an important job themselves. They typically hire someone to do the job for them or cobble together a less-than-adequate solution

New-market disruptive innovation

Product/service that helps people do more conveniently what they are already trying to get done

Explosive rate of growth in a new market or a new context of use

Undershot consumers

Consumers who consume a product but are frustrated with its limitations, they display a willingness to pay more for enhancements along the dimensions most important to them Sustaining up-market innovation (radical and incremental)

New, improved products and services introduced to existing customers Integrated companies thrive, specialist companies struggle Overshot consumers

Customers who stop paying for further

improvements in

performance that historically merited attractive price premiums

Low-end disruptive innovation

New business model emerges to serve least-demanding customers Displacing innovation Emergence of specialist company targeting mainstream customers Downward migration of required skills

Emergence of rules and standards – widely propagated statements of what causes what

Migration of provider closer to end customer Table 1: Types of disruptive innovations (Source: Christensen C., Anthony S., Roth E. 2004)

(23)

Research Methodology

In order to answer the research question “Is 3D printing a case of disruptive

innovation?” both qualitative and quantitative research were considered, but due to

the design of the research, the novelty of the topic and the availability of data that could be obtained, the former was chosen. The research is exploratory - which is by definition “a study research that aims to seek new insights, ask new question and to assess topics in a new light” (Saunders & Lewis 2012) - as it was found most suitable to address the novelty of the topic.

Data collection

Four methods of gathering information are generally used for qualitative research: participation in the setting; direct observation; in depth interviews; and analysis of documents and materials (Marshall & Rossman, 1998). All of these four methods were used in the course of this research.

First, a five-month internship was carried out at 3D Hubs B.V., an Amsterdam-based startup company aiming to create a worldwide network of 3D printers that will make distributed manufacturing a reality. The company has the largest network of 3D printers in the world with more than 16,000 locations in 150 countries and is considered one of the mayor worldwide 3D printing service providers. During the course of a 5-month internship, the researcher acquired hands-on experience with several 3D printing processes, use cases as well as data and knowledge about the industry.

In order to avoid opinion bias, the aim for the interviews was to recruit interviewees with different backgrounds and with different relations to 3D printing. The interviewees included designers, 3D-printer owners, 3D print shop owners, service providers and producers of 3D printed products. Some of the interviews were supplemented with internal data sources.

The interviews consisted of two separate parts. The first part was always specific to the person being interviewed1. As the interviewees had different backgrounds, worked !!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!

(24)

in different industries, this part was aimed to understand their connection to 3D printing and where they or their companies stand in the value chain.

The second part was a semi-structured interview. In a semi-structured interview the participant is asked about several topics using predetermined questions, but the order of the questions asked may vary. Moreover, in order to create discussion, the semi-structured format allows the interviewer to eliminate, alter or ask additional questions if appropriate. The advantage of semi-structured interviews is that this design allows the researcher not only to elicit foreseeable information, but also unexpected types of information (Seaman, 1999). Moreover, the results are comparable and consistent; hence basically the same type of information is gained from each participant, while the researcher possesses enough flexibility and freedom during the interviews.

Altogether 9 semi-structured interviews were conducted. They all had an open character, lasted between 45 and 90 minutes and were audio-recorded. The purpose was to get clear answers and create discussion in order to explore the topic as extensively as possible.

For collect data from direct observation, the researcher visited the 3D Printed Canal House, which is a house currently under construction in Amsterdam using a 3D printer. I viewed the planning and the printing process as well as made inquiries to the architects for more details. Second, the 3D printed results of a public transport design project for the city of Budapest were observed and its designers interviewed2.

Triangulation in data collection is aimed to increase the credibility of the results through multiple expert interviews and the usage of secondary data sources (Tracy, 2010). In the course of research, the primary data was supplemented with the analysis of documents and materials.

Data analysis

The data consisted of interviews, internal resources, documents, notes from personal observation and various articles. First, the interviews were transcribed. Then, codes were generated with in the NVivo software. The interviews and all the other resources !!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!

(25)

were grouped through these codes around several topics or aspects of 3D printing. This made the data more comprehensible and ready for cross-data analysis. Later, the data was analyzed based on the research question and the sub-questions. The aim was to find they key attributes of 3D printing that make it superior to other technologies as well as the ones that are currently hindering it, in order to discover the specific industries where it could be applied.

(26)

Research Findings

This chapter presents the results from the qualitative data gathered from the combination of four collection methods. The findings are specific to the experience from the internship conducted at 3D Hubs, supplemented by the use of interviews, the analysis of documents and materials, and secondary data from industry analysis. The data is grouped into four identified themes which were deemed most likely to answer the research question, namely the advantages and disadvantages of 3D printing, the current applications of the technology and the analysis of the industry and forecast.

Advantages of 3D printing

Advantages Disadvantages

• Speed and ease of designing and modifying products

• Increased parts complexity

• No need for costly tools and molds • Automated 24-hour manufacturing • Less need for workforce

• Reduced material waste and scrap, more efficient use of raw materials • Ability to recycle waste material

• Minimal inventory risk as there is no unsold finished goods inventory

• Improved working capital management as goods are paid for before being manufactured

• Ability to easily share designs and outsource manufacturing

• Shortened time to market

• Higher costs for large production runs relative to injection molding and other technologies

• Reduced choice for materials, colors, and surface finishes

• Lower precision relative to other technologies

• Limited strength, resistance to heat and moisture, and color stability • Copyright issues

Table 2: Advantages and disadvantages of 3D printing

3D printing has several key advantages over traditional manufacturing processes. Products can be more easily designed on a computer and later modified without any additional costs. Complex shapes and parts can be printed, that cannot be produced by

(27)

any other means, giving free hands to designers. “Complexity doesn’t have a price, a

simple product is just as expensive as an complex one, because the machine doesn’t have an outlining or binding bond and this is especially for advanced technologies like spacecraft” (Interviewee). In metal casting and injection molding, a new product

requires a new mold in which to cast the part. In machining, several tool changes are needed to create the finished product. However, AM is a “single tool” process, regardless of the desired geometry, there is no need to change any aspect of the process, no additional cost or lead time between making an object complex or simple. As such, AM processes are excellent for creating customized, complex geometries. (Campbell et al, 2011)

There is no need for production lines, costly tools and molds as 3D printers create products technically preassembled or with a greatly reduced number of individual parts. This makes small production run applications - such as mass-customized products, prototypes, replacement parts, medical/dental applications, and bridge manufacturing - less costly. It also enables firms to profitably use 3D printing to economically fill custom orders and serve niche markets.

All AM processes create physical parts directly from a standardized digital file, which is a representation of a three-dimensional solid model. The printing itself is a computer-controlled process, which requires a low level of operator expertise and reduce the amount of human interaction needed to create an object. In fact, these processes often operate unmonitored. This allows for overnight builds and dramatically decreases the time to produce products - thus reducing the time between design iterations. Creating the part directly from the computer model ensures that the created part precisely represents the designer’s intent and thus reduces inaccuracies found in traditional manufacturing processes. (Campbell et al, 2011) Furthermore, just as open-source programmers collaborate by sharing software code, engineers are already starting to collaborate on open-source designs for objects and hardware. These characteristics make 3D printing ideal - and already widely used - for prototyping: It enables the production of a single item quickly and cheaply; and then another one after the design has been refined. In contrast to injection molding processes that require costly molds, 3-D printing entails relatively low fixed costs

(28)

Due to the low costs of modifying prototypes, marketers can more easily test different product versions based on customer and design feedback. Product security and privacy can also be increased as prototypes can be produced in-house, instead of being outsourced to outside vendors often in countries with limited copyright protection.

Considering manufacturing, currently most metal and plastic parts are designed to be manufactured, which means they can be clunky and contain material that does not add value to the part's function but is necessary for producing it. 3D printing eliminates this waste making lighter, more elegant and functional designs possible. It’s easy to print custom made: “ For example you have now textile printers. Why do we wear

confection sizes, S, M, L, because of scale production, this will be something of the past. In 10 years we will scan your body and you will get clothes printed for your body. Shoes as well, why would you need 44 or 33 if they have the specific size of your foot.” (Interviewee)

Moreover, there is minimal inventory risk, as products and parts can be printed when there is demand for them. It can improve capital management as goods can be manufactured once they have been paid for. Decentralization of production could be possible as well. Hundreds of thousands of a given product may be done by producing thousands on one hundred printers that are near the source of demand around the world rather than at one factory producing hundreds of thousands of the same item. Production would be also more flexible as the same printers producing thousands of each item could be instantly reprogrammed to produce different products as demanded.

While reducing costs, 3D printing can be beneficial for the environment as well. As material is added layer by layer, there is virtually zero waste and raw materials and waste can be recycled in most cases. Aerospace is one of the most cited relevant examples. In this industrial sector, buy-to-fly ratios are approximately 20:1, i.e. 20 kg are required to produce 1 kg of end product. The remaining 19 kg are waste requiring reprocessing or recycling (Achillas, 2014). Case studies indicate that up to 40% of the raw material-related waste can be avoided through 3D printing, while 95–98% of the unfused raw material can be reused (Petrovic et al., 2011). EADS, an aircraft

(29)

manufacturer has managed to print airplane parts from 10% of the raw materials compared to regular methods.

Lightness is also critical in aviation. A reduction of 1kg in the weight of an airliner will save around $3,000 worth of fuel a year and cut carbon-dioxide emissions at the same time (The Economist, 2011b) adding up to $90,000 savings over an aircraft’s 30-year life-span (Econolyst, 2014). Further indirect manufacturing inputs can be avoided as 3D printing does not require adjuvants as coolants, lubricants or other partly environmentally harmful substances. (Gebler et al., 2014). Carbon footprint of products is reduced, as they can be printed close to their destination, reducing shipping costs and emissions. One of the most common materials for desktop 3D printers is polylactic acid (PLA), which is a bioplastic derived from renewable resources, such as corn.

Perhaps the most exciting aspect of the technology is that it lowers the barriers to entry into business, therefore promoting innovation. Since 3D printing does not require expensive tooling, forms, or punches, it is particularly cost effective for very small production runs and market testing. Instead of setting up a factory or looking for a mass-producer to manufacture a product, 3D printers could offer a less risky route to market. An entrepreneur could build a small batch to test if his or her idea works in practice and easily modify it if necessary, making testing of new products less risky and expensive. If the results are promising, they can scale up with conventional mass production or larger 3D print runs, whichever is deemed more economical. This suggests that success in manufacturing will depend less on scale and more on the quality of ideas (The Economist, 2011b). The process is already observable on crowdfunding platforms, as entrepreneurs create 3D printed proof of concepts to test their ideas and gather funding for large-scale production.

Disadvantages of 3D printing

As we have seen, 3D printing has several critical advantages over traditional manufacturing processes, but there are inherent limitations as well, even though many of them are likely to be reduced as the technology advances.

(30)

known as ABS, is the most common 3D-printing material. A mass manufacturer using plastic injection molding might buy ABS in bulk for about $2 per kilogram, but as a filament for 3D printing it can cost as much as $40-80 a kilo. (The Economist, 2013a) In part the price difference is due to higher standards of purity and composition required for 3D printing, but it can be attributed to the fact that some 3D printer manufacturers require users to buy materials from them and mark up the price, just as conventional inkjet printer manufacturers. This strategy will probably not be sustainable long term as third-party suppliers have already entered the business, pushing down the prices and most 3D printers are not dependent on the supplier of the material.

3D printing is not yet competitive with regular manufacturing when it comes to large production runs. According to different sources, printing is cost effective on production runs of up to 1,000 or 10,000 units, depending on material and the item’s design (Interviewee, The Economist, 2011a). In the future, experts believe that the range of efficient production may be further increased as raw material prices will drop as more firms will use 3D printing to produce finished goods and as final consumers will begin purchasing 3D printers.

Material choices are also limited; currently there are about 50 available, most of them being varieties of plastics. Colors and surface finishes suitable for 3D printing are also more limited than with typical mass-production processes (Stratasys, 2014) However, this field is rapidly catching up, the number of new materials added to the 3D printing palette is growing rapidly. “It’s fair to say that in the future you will be able to print

more or less everything” (Interviewee). Materials are also key from a sustainability

point of view: “One of the advantages of 3D printing is that you don’t have shipping,

but then you have the problem with shipping materials, you just shift the problem to another part of the supply chain. So in the future we probably need biomaterials that you could turn into filament. “(Interviewee)

Precision and strength also need to be improved. Today, 3D printing works with plastics, resins, and metals, with a maximum precision of around 20-100 microns, one-tenth of a millimeter (Interviewee, The Economist, 2011a), which is still not sufficient to compete with industrial engineering processes. Most AM processes use proprietary polymers that are not well characterized, and are weaker than their

(31)

traditionally manufactured counterparts. “The stainless steel we’re using is not

comparable to regular stainless steel. The normal stainless steel is melted at such high temperatures, and 3D printing is at much lower, which makes it less strong.” (Interviewee)

Also, in some AM processes, part strength is not uniform—due to the layer-by-layer fabrication process, parts are often weaker in the direction of the build. Repeatability is in need of improvement as well; parts made on different machines can often have varying properties (Campbell et al, 2011). This is especially true for current desktop 3D printer models. It is fair to say that final precision is hugely dependent on the expertise of the operator and the calibration of the machine.

Additive manufacturing has other limitations. It can be slow – it takes several hours to print a body panel for a car, for example. But speed is relative. What may be too slow for a large production run might be fine for an individual item, which alternatively would take weeks to make in a machine shop. While the speed of additive manufacturing processes continues to increase, expert argue if it will ever be able create parts as fast as for example molding technologies. Some say that the bottleneck lies in the fundamental physics of the processes - it is not possible to scan a laser, cure material and recoat each layer or to depose subsequent layers of material at a speed comparable to that of injection molding (Campbell et al, 2011). Nevertheless, this limitation is only valid for the production of several thousand of a common part. Since tooling must be created for each unique part one wishes to injection mold, AM is the preferred process when custom parts, or low-volume production runs are needed. Meanwhile, some companies are claiming to be working on fast systems with success, creating ultra-fast machines and effectively breaking the “speed barrier” (McCue, 2014; Krassenstein, 2014b)

The energy use of 3D printers is also egregiously underestimated. 3D printing a plastic object uses 5 to 10 times more energy per pound compared to conventional industrial injection molding, while printing metals is even more energy inefficient, due to the slow printing process and the heat needed to melt the materials. (Faludi, 2013) The picture is much better if we look at the whole lifecycle, supply chain and shipping. Additive manufacturing is also likely to get faster; there is plenty of room to

(32)

As 3D printing is a new field, it inevitably brings up legal issues. Responsibility is yet unclear. “If you have mass produced cup, you buy it at the supermarket and it breaks.

Who is responsible? The supermarket? The supplier? The designer? The printer? Yourself? You cannot ask money for something and say you cannot get sued” (Interviewee)

Copyright and privacy issues are also prevalent. Good ideas can be copied even more rapidly with 3D printing, so battles over intellectual property may become even more intense. It will be easier for imitators as well as innovators to get goods to market fast. Competitive advantages may thus be shorter-lived than ever before. Just like when music became digital, pirated versions of tracks and albums started circulating the Internet. Appleyard (2015) defines significant parallels between P2P technologies and 3D printing. These include: (1) IP is held as a computer file requiring hardware/software to realize; (2) piracy is undertaken in the individual’s home; and (3) online communities support participation. Aside from ownership, the barrier to online piracy is the ability to use relevant technology effectively, which is currently lacking for 3D printing.

If anyone – including competitors - with the right equipment is able to print parts or products, designers and producing companies will be at risk of losing business. Eventually the greatest beneficiaries may not be the companies, but their customers. Nevertheless, increased security measures, development of new digital rights management solutions and an update of intellectual property rights will be required to protect producers.

3D printing applications

Initially, additive manufacturing was referred to as “rapid prototyping,” as it was primarily used to quickly fabricate conceptual models of new products and prototypes for form and fit evaluation. An architect can design a new building on a computer and print out a 3D model to get a glimpse of how it would look like in reality and to show a client and to further refine the design. Automotive engineers can design and print a prototype of a vehicle in small size, make adjustments before creating an expensive and time-consuming model from clay.

(33)

As material properties and process repeatability improved, AM technologies’ use has evolved from solely creating prototypes, to creating parts for functional testing, to creating tooling for injection molding and sand casting, and finally, to directly producing end-use parts (Campbell et al, 2011).

From 2003 to 2012, the 3-D printed finished goods market grew from 3.9% of all 3-D printing revenues to 28.3%. (Wohlers, 2014) This trend is expected to continue in the future as the technology is continuously improving. McKinsey (2013) expects the market for complex, low-volume, highly customizable parts, such as medical implants and engine components to be $770 billion annually by 2025, and it is possible that some 30 to 50 percent of these products could be 3D-printed.

A similar trend can be observed on the consumer market as well. Prototyping is still the most popular use case, but gadgets, hobby, household items and art/fashion 3D prints are gaining traction. (3D Hubs Trend Report, 2015) “This is typically what

people make with technologies in their early stages. If a technology is new, the first thing people start to make is prototypes, when it’s a bit better, people start to make gadgets, when it’s maturing more, people start to make fashion (Interviewee)”.

Industrial uses of additive manufacturing include:

Aerospace: SuperDraco, the engine of the Dragon V2 spacecraft by SpaceX is manufactured using AM. The technology enables the engine to be manufactured as one finished part, which requires no assembly, increases the part's structural integrity, improves its reliability, and lowers the overall cost to manufacture. Building a SuperDraco engine would've been next to impossible using conventional manufacturing methods as it is made of a super alloy that is very difficult to machine. (Heller, 2014)

Figure'2:'3D'printed'finished'goods'revenue'share'of'3D'printing' market'(Source:'Wohler’s'Report'2014)'

(34)

Aircraft: The aircraft industry is spearheading the development, especially development of metal 3D printing. Boeing has installed environmental control system ducting made by AM for its commercial and military aircraft for many years. In fact, tens of thousands of AM parts are flying on 16 different production aircraft—both commercial and military. GE has announced that it would be using AM to produce major parts of jet engines and has committed to manufacturing 30,000 fuel nozzles annually for its new engine. The new design consolidates 18 parts into one, and is 25% lighter and five times more durable than the previous fuel nozzle (Caffrey & Wohlers, 2014). Today, a typical F-18 fighter jet is likely to contain some 90 3D-printed parts, even though the F-18 has been in service for two decades, before 3D printing took off. This is because replacement bits, like parts of the cockpit and cooling ducts, are now 3D printed. The F-35, a new strike aircraft entering service in America, has around 900 parts that have been identified as suitable for additive manufacturing (The Economist, 2011a). Meanwhile, Airbus, Pratt & Whitney, Rolls-Royce, Honeywell, NASA, the Aviation Industry Corporation of China, and other aerospace companies are accelerating their involvement and investment in AM. Increasing investment is the technology and involvement of such huge companies is likely to speed up the development of the industry. Just a few years ago, uncertainty surrounded metal AM. Concerns included surface porosity, a lack of full density, and unpredictable microstructure. Today, AM systems are making parts with material properties that exceed the properties of castings, and match the properties of wrought materials. (Caffrey & Wohlers, 2014)

Automobile components: While AM is not yet used for mass production, it is increasingly used to create components for high-end, specialized automobiles. Engine parts for Formula 1 racecars have been fabricated using direct metal laser sintering. As with aircraft components, it is also used for creation of hard-to-find spare parts, during classic car restoration.

Health care: Medical companies have also embraced 3D printing, as it is hard to find an industry where customization is more important. Patient-specific knee replacements can mimic the patient’s own anatomy, artificial limbs provide custom fit and cost a tenth of comparable artificial limbs made using traditional methods. Since an individual’s ear canal is unique in size and shape, a hearing aid needs to fit perfectly for optimal performance. Using CAD software, the contours of a patient’s

(35)

ear can be digitized within minutes. And within hours, a custom-made hearing aid shell can be produced from liquid photopolymer (Berman, 2012)

Dentistry: Within healthcare, the most applications can be found most likely in dentistry. According to EOS of Germany, one of the largest manufacturers of 3D printers, about 19,000 dental copings are manufactured every day using the company’s direct metal laser sintering systems. (The Economist, 2013a) Up to 450 dental crowns, each tailored for an individual patient, can be manufactured in one go in a day by a single machine. The alternative - craft producers of crowns - would do well to manage a dozen a day (The Economist, 2011b) In another example, 3D Systems has worked with Align Technology of San Jose, California. Instead of using metal braces for straightening teeth, Align produces sets of transparent plastic “aligners”. A scan of the patient’s mouth is used to devise a treatment plan, which in turn generates a digital file that is used to 3D-print a set of 20 or so molds. Each mold is slightly different, and from them a series of clear plastic braces is cast. When worn over several months, each brace steadily moves the patient’s teeth into the desired position. Last year Align 3D printed 17 million of them (The Economist, 2013a). The technology has started to trickle down to the final user level as well. New desktop 3D printers using stereolithography are capable of producing castable molds from resin, which can be widely used for dental treatments.

A survey of 100 top manufacturers in the world by PwC revealed that two-thirds are already using 3D printing, some for rapid prototyping and others for production or custom parts. The majority (30%) of the companies, however, are still experimenting with the technology to determine how they can apply it to their production processes. Interestingly, only one-third of the companies surveyed by PricewaterhouseCoopers (PwC) said they are not implementing 3D printing in some way (Mearian, 2014). The popularity of 3D printers has been increasing on the consumer market as well. Thanks to expiring intellectual property and the open-source or crowdsource nature of these projects, the cost of desktop printer ownership is decreasing. The price of a basic 3D printers using fused deposition modeling technology have declined from $30,000 a few years ago to less than $1,000 for some models.

Is 3D printing a case of disruptive innovation?