2.1. Innovation as a phenomenon

Writing on innovation is becoming more frequent and is growing to include many perspectives and levels of analysis (Fagerberg & Verspagen Reference Fagerberg and Verspagen2009). A main strand of the innovation literature analyzes innovation as a phenomenon. Numerous studies provide empirical evidence of the patterns of innovation across nations, industry sectors, firms, and technology classes (Furman, Porter & Stern Reference Furman, Porter and Stern2002; Castellacci Reference Castellacci2007; Griliches Reference Griliches1990; Scherer Reference Scherer1965; Pavitt Reference Pavitt1985; Koh & Magee Reference Koh and Magee2006). Other studies reveal and define types of innovation, such as architectural and modular innovation (Henderson & Clark Reference Henderson and Clark1990; Baldwin & Clark Reference Baldwin and Clark2000), radical and incremental innovation (Daft & Becker Reference Daft and Becker1978; Hage Reference Hage1980), product versus process innovation (Abernathy & Utterback Reference Abernathy and Utterback1978), and recently, disruptive innovation (Christensen Reference Christensen1997), open innovation (Chesbrough Reference Chesbrough2003), and distributed innovation (Anderson & Joglekar Reference Anderson and Joglekar2012; Baldwin Reference Baldwin2012). Despite being useful for analyzing the subsequent competitive implications of innovation (Abernathy & Clark Reference Abernathy and Clark1985), such studies of innovation as phenomena are naturally limited to explaining how innovation arises.

Meanwhile, since Schumpeter (Reference Schumpeter1934), scholars (primarily economists) have sought to understand the contexts and factors that influence the success and failure of innovation (Schmookler Reference Schmookler and Nelson1962; Freeman Reference Freeman1974). Studied factors include R&D spending and manpower (Arrow Reference Arrow and Nelson1962; Nelson Reference Nelson1962; Mowery & Rosenberg Reference Mowery and Rosenberg1998), industrial and competitive dynamics (Rosenberg Reference Rosenberg1963; Porter Reference Porter1990), institutional environments (Lundvall Reference Lundvall1992; Nelson Reference Nelson1993; Freeman Reference Freeman1995), regional ecosystems (Saxenian Reference Saxenian1996; Kenny Reference Kenny2000; Blank Reference Blank2008; Delgado, Porter & Stern Reference Delgado, Porter and Stern2010), culture (Florida Reference Florida2004; Senor & Singer Reference Senor and Singer2009), venture capital (Samila & Sorenson Reference Samila and Sorenson2010), intellectual property protection (Merges & Nelson Reference Merges and Nelson1990), and employment law (Marx, Strumsky & Fleming Reference Marx, Strumsky and Fleming2009), among others. Alternately, psychologists have examined cognitive, individual, team, organizational, and cultural factors that may restrict or empower creativity (Weisberg Reference Weisberg2006; Hennessey & Amabile Reference Hennessey and Amabile2010).

Management scholars have particularly focussed on firm-level organizational capabilities for learning and experiments that enable firms to continually innovate (Cohen & Levinthal Reference Cohen and Levinthal1990; Teece & Pisano Reference Teece and Pisano1994). The structural dimensions of organizations, such as the decentralization of decision making, loose links among divisions, and vertical integration versus disintegration, are associated with different types of firm innovation capabilities (Sanchez & Mahoney Reference Sanchez and Mahoney1996; Fang, Lee & Schilling Reference Fang, Lee and Schilling2010; Kapoor Reference Kapoor2013). Studies of inventor networks (Sorenson, Rivkin & Fleming Reference Sorenson, Rivkin and Fleming2006; Singh & Fleming Reference Singh and Fleming2010), inter-firm alliances and collaboration networks (Gulati Reference Gulati1998; Schilling & Phelps Reference Schilling and Phelps2007), and transaction networks (Luo, Olechowski & Magee Reference Luo, Olechowski and Magee2012) have shed light on how firms access and leverage complementary resources and knowledge in a larger ecosystem to pursue innovation opportunities. Taking the ecosystem and network perspective to the national level are studies on ‘national innovation systems’ (Lundvall Reference Lundvall1992; Nelson Reference Nelson1993; Freeman Reference Freeman1995), which emphasize the interactions of different institutions, such as companies, universities, government agencies, and intermediary organizations, to collectively nurture innovation.

2.2. Innovation as a process

Innovation scholars who are interested in operations have focussed on the process of new product development (NPD) (Ulrich & Eppinger Reference Ulrich and Eppinger2001; Krishnan & Ulrich Reference Krishnan and Ulrich2001; Cagan & Vogel Reference Cagan and Vogel2002). The NPD process is often referred to as the ‘innovation process’. This process has been conceptualized, analyzed, and organized as the sequential stages of opportunity identification, concept generation, product design, development, and commercial production (Clark & Wheelwright Reference Clark and Wheelwright1993; Otto & Wood Reference Otto and Wood2000; Schilling Reference Schilling2010). Managers often make go/kill decisions on whether to move a project from one completed stage to the next (Cooper Reference Cooper2000). This is often referred to as a ‘stage-gate process’ and is well accepted by researchers and practitioners. A survey shows that nearly 60% of firms use some type of stage-gate process to control project quality and reduce development costs (Schilling Reference Schilling2010, pp. 249).

In contrast, partially parallel development processes, in which a later stage is initiated before an earlier stage is finalized, have received much attention because this approach may enable close coordination, prevent costly iterations between stages, and shorten overall development time (de Meyer & van Hooland Reference de Meyer and van Hooland1990). With advanced technology and high market uncertainty, however, the tight coupling of stages may result in frequent and costly reworking of both the product and the process (Eisenhardt & Tabrizi Reference Eisenhardt and Tabrizi1995; Terwiesch & Loch Reference Terwiesch and Loch1999). Research has also suggested that involving customers and suppliers in the NPD process can often help match NPD efforts with potential demands and needs (Herstatt & von Hippel Reference Herstatt and von Hippel1992; Bonaccorsi & Lipparini Reference Bonaccorsi and Lipparini1994). Concrete frameworks and methods have been proposed to aid in understanding customers with regard to social–economic–technological trends and identifying new product design opportunities based on this understanding (Cagan & Vogel Reference Cagan and Vogel2002).

Recent studies have called for the utilization of design and creative thinking techniques to ensure the novelty of the outcome of the product development process (Seidel & Fixson Reference Seidel and Fixson2013). Such techniques may include the theory of inventive problem solving (TRIZ; Altshuller and Shapiro Reference Altshuller and Shapiro1956; Cascini & Russo Reference Cascini and Russo2007), design by analogy (Chan et al. Reference Chan, Fu, Schunn, Cagan, Wood and Kotovsky2011; Fu et al. Reference Fu, Chan, Cagan, Kotovsky, Schunn and Wood2013), and the crowdsourcing of design features (Burnap et al. Reference Burnap, Ren, Papalambros, Gonzalez and Gerth2015). Design theoreticians have created several formal models of innovative design processes, such as function–behavior–structure (Gero Reference Gero1990), infused design (Shai & Reich Reference Shai and Reich2004), concept–knowledge (C–K) theory (Hatchuel & Weil Reference Hatchuel and Weil2009), idea matrixes, and creativity operators (Tang & Luo Reference Tang and Luo2013). Such formal models may guide the development of new design tools and creativity methods. These models emphasize knowledge management and creativity in design processes.

The design literature has developed a fruitful understanding of a broad set of human activities that begin with abstraction and end with useful novelty in the design process (Antonsson & Cagan Reference Antonsson and Cagan2001; Magee et al. Reference Magee, Wood, Frey and Moreno2013). Such activities include conception, problem definition, idea generation and evaluation, prototyping, experimentation, and refining. Empirical and theoretical research has shown that flexibly structured processes, rather than rigidly structured ones, yield greater design success (Frey et al. Reference Frey, Herder, Wijnia, Subrahmanian, Katsikopolous and Clausing2009; Brooks Reference Brooks2010). Creative processes often require iterations between divergent and convergent thinking (Dym et al. Reference Dym, Agogino, Frey, Eris and Leifer2005; Hennessey & Amabile Reference Hennessey and Amabile2010). Design expertise, which comprises domain-specific knowledge plus the ability to transform it to derive novelty, is a fundamental enabler of creative design outcomes (Reyna Reference Reyna, Mio and Katz1996; Weisberg Reference Weisberg2006; Luo et al. Reference Luo, Baldwin, Whitney and Magee2014).

As indicated in these diverse but related studies, innovation results from a broad range of relevant activities and factors. The brief review above was not intended to cover all innovation studies, but to reveal the limited scope in individual studies and coupling of the existing understanding of concepts related to innovation. Such limitations may lead to oversight regarding valuable collective actions for enhancing innovation. This paper aims to reveal and connect the broadly relevant sub-processes that contribute to innovation, and then develop a more united view of the innovation process.

3.1. Three sub-processes of innovation

An innovation is recognized or acknowledged when its associated invention is successfully implemented (Hennessey & Amabile Reference Hennessey and Amabile2010); that is, when the invention has successfully diffused into broad use, received massive public recognition, and achieved real economic or social impacts (Utterback Reference Utterback1974; Rogers Reference Rogers2003; Luo et al. Reference Luo, Baldwin, Whitney and Magee2014). Inventions, however, do not necessarily lead to innovations. Despite their useful novelty, many noticeable inventions fail to move from R&D labs to markets to become recognized as innovations (Times Magazine 2010); the Segway and Google Glass are just two examples. Additionally, many inventors fail to capture the profits from their own inventions; an example is the computer graphic user interface (GUI), initially invented at Xerox Palo Alto Research Center (PARC). Although the GUI later became a standard, Xerox made little revenue from its invention. The first successful commercial GUI product was the Apple Macintosh, which was inspired by the inventive work at PARC. Microsoft also gained phenomenal profit from its Windows operating system, also GUI-based. The varied degrees and pathways of the implementation and diffusion of inventions indicate that the process through which inventions become innovations is highly relevant to the eventual realization of innovation.

The process by which new inventions reach the market and ultimately the users is the ‘entrepreneurship process’. It is the effort of entrepreneurship that matches inventions with market opportunities and needs. Entrepreneurship has been considered a pivotal element of innovation over time (Wennekers & Thurik Reference Wennekers and Thurik1999). Schumpeter (Reference Schumpeter1934) was the first to add the concept of entrepreneurship to innovation and to consider entrepreneurs as the agents who push creative destruction. Stevenson & Gumpert (Reference Stevenson and Gumpert1985) explicitly distinguished the entrepreneurial process from the administrative process and defined entrepreneurship as ‘the pursuit of opportunity beyond the resources you currently control’. Note that while general entrepreneurship does not necessarily involve technology or invention (Stevenson Reference Stevenson2006; Shane Reference Shane2008), an entrepreneurial process is required to turn an invention into innovation.

The novel nature of inventions implies the resulting uncertainty about markets, demand, and supporting ecosystems, and thus determines that the process of nurturing, commercializing, or implementing inventions is always an entrepreneurial one, but to varied degrees. Higher novelty in an invention demands greater entrepreneurial effort to nurture it and further drive it to enter and penetrate the market, where it becomes an innovation. For example, Steve Jobs created and managed a highly entrepreneurial process that was separate from Apple’s main operations to design and market the Macintosh when Apple I and Apple II were the successful main product lines of Apple in the early 1980s (Wasserman Reference Wasserman2011). Entrepreneurship is needed for the innovation-oriented practices of not only new venture startup firms but also established companies, where it is called ‘corporate entrepreneurship’ or ‘intrapreneurship’ (Pinchot Reference Pinchot and Katz1997; Wolcott & Lippitz Reference Wolcott and Lippitz2007).

The sub-process that results in invention is the ‘creative design process’. Design is the process of applying knowledge to create artifacts that are novel and useful to others (Simon Reference Simon1996; Luo et al. Reference Luo, Baldwin, Whitney and Magee2014). Design processes often include a series of activities, such as conception, problem definition, idea evaluation, prototyping, and refining. Design does not guarantee the achievement of an invention, although it creates the possibility. Some design processes may result in only limited novelty that is insufficient to be considered an invention, or limited usefulness that is insufficient to drive the entrepreneurial process. Such design processes are less relevant to innovation. This paper focuses on the creative design process that delivers sufficient novelty for invention and further innovation.

The design process requires knowledge as inputs. Some knowledge concerns the natural phenomena related to design problems: this is scientific knowledge. Other knowledge is related to the practices, solutions, methods, tools, social contexts, and any other existing artifacts in a specific design domain: this is design knowledge. For example, the design of piston engines requires a large volume of domain-specific design knowledge about the existing artifacts (including existing engine designs, tools, methods, processes, regulations, etc.), in addition to the basic scientific knowledge about the physical phenomenon of combustion.

Design knowledge is generated and accumulated through continual design and learning activities in the design process, as suggested by the C–K theory (Hatchuel & Weil Reference Hatchuel and Weil2009). In contrast, scientific knowledge is normally created in the ‘scientific process’, preceding the design process that uses knowledge to create new artifacts. Continual scientific knowledge creation enables new design opportunities and the creation of novel artifacts. For example, the creation of scientific knowledge of photoelectric effect and an improved understanding of silicon materials enabled the invention of solar PV cells. Despite providing knowledge ingredients for the design process, scientific processes are not typically aimed at designing artifacts, but at the discovery of unknown truths in the natural world. Scientists are mainly driven by curiosity. For example, Galileo Galilei discovered gravitation and Isaac Newton formulated the law of universal gravitation without thought to applications. These types of discoveries and theories from scientists allowed many useful artifacts based on them to be designed later.

The creative design processes that aim for highly novel and path-breaking inventions may need to rely more on scientific knowledge than on design knowledge because by definition, little relevant design knowledge (e.g., existing solutions, relevant tools/equipment, regulations, and policies) exists prior to the path-breaking invention. After an invention has opened a new design path, related design knowledge will be generated and accumulated. In fact, related design knowledge will increasingly be required in further design iterations along the same trajectory. Later activities in the same trajectory would be more oriented for incremental improvements and optimization rather than invention. Therefore, the goal of achieving innovation suggests the value of scientific knowledge for creative design, and the relevance of the scientific process that continually creates new scientific knowledge to enable creative design opportunities.

Earlier studies have shown that scientific research increases the rate and quality of related technological inventions (Jaffe Reference Jaffe1989; Jaffe & Trajtenberg Reference Jaffe and Trajtenberg1996), and engineers can benefit from reading the relevant scientific literature to solve difficult technical problems (Gibbons & Johnston Reference Gibbons and Johnston1974; Allen Reference Allen1977). Dugan & Gabriel (Reference Dugan and Gabriel2013) attribute the success of DARPA in producing many path-breaking inventions to its uniting of scientists and engineers. At DARPA, scientists and engineers collaborate in the same project teams, which seek to solve challenging engineering design problems that require pushing the frontiers of basic science. Fleming & Sorenson (Reference Fleming and Sorenson2004) theoretically argued that scientific knowledge acts as a ‘map’ that can help inventors perceive and explore a wider space of design possibilities and identify the directions and shortest paths to better solutions. His empirical analysis of patent data showed that the contribution of science to invention is greater when the technologies or design problems are more complex and that the use of scientific knowledge in design can mitigate the negative effect of the difficulty of interdisciplinary knowledge integration on the value of invention. These statistical and case studies have suggested the value of uniting scientific research and engineering design efforts.

3.2. Characteristics of the sub-processes

Thus far, we have identified three sub-processes that contribute to innovation—science, design, and entrepreneurship. Despite their differences, these three sub-processes also share certain characteristics in common (Table 1). First, they are all creative processes. Science is about creating previously unknown understanding and knowledge of the natural world. Design creates new artifacts that did not exist previously. Entrepreneurship establishes new businesses, uses, or installments that did not exist previously. Creativity distinguishes these three processes from production or service delivery processes, for which precise repetition is essential.

Second, science, design, and entrepreneurship processes are all uncertain. The output from such processes is unpredictable and idiosyncratic by nature. To cope with uncertainty, scientists, engineering designers, and entrepreneurs must continually experiment, collect feedback, and learn through trial and error. As a consequence, the success of experimental processes often comes after many failures and experiential learning. To fail cheaply and adjust quickly, flexibility is important in the design and management of all three sub-processes (Rajan et al. Reference Rajan, van Wie, Wood, Campbell and Otto2005; Frey et al. Reference Frey, Herder, Wijnia, Subrahmanian, Katsikopolous and Clausing2009; de Neufville & Scholtes Reference de Neufville and Scholtes2011). In brief, experimentation and learning are the core activities of all three sub-processes, and a flexible process and mindset facilitates experimentation and learning. In contrast, rule-based mindsets and organizations are inappropriate for such creative and uncertain processes.

Furthermore, the required experiments and anticipated failures determine that all three processes are costly. To sustain experimentation for learning and afford the necessary failures requires economic resources. Therefore, to succeed in their respective processes, scientists, engineering designers, and entrepreneurs need to search for resources to sustain their activities. An entrepreneurial process is often funded by external venture capital funds, or the internal venture funds of large companies in the case of corporate entrepreneurship. The economic returns of new ventures or corporate ventures can be used to compensate and sustain their upstream creative design activities, which in turn, are expected to consider market and customer needs. The scientific process is mainly funded and sustained by public resources, such as the National Science Foundation, partly because science does not directly generate economic returns. When relevant scientific, design, and entrepreneurial activities are closely united to result in innovation, however, the contribution of science to invention and innovation will be straightforward and visible. Thus, the related scientific research may be better rewarded, appropriating a portion of the economic return from the resulting innovation.

Because of the shared creative, uncertain, and costly nature of the three sub-processes, scientists, engineering designers, and entrepreneurs can learn from one another to improve their respective processes. In particular, design scientists have developed rich theories and methods that can potentially be leveraged to improve the understanding, design, and management of scientific and entrepreneurship processes. For instance, design scholars have shown that C–K theory and infused design can indeed support scientific discoveries (Shai, Reich & Hatchuel Reference Shai, Reich and Hatchuel2013). Recently, scholars and practitioners advocated the adoption of design thinking and ideation methods to improve creativity in broader business practices (Beckman & Barry Reference Beckman and Barry2007; Verganti Reference Verganti2008; Brown Reference Brown2009; Seidel & Fixson Reference Seidel and Fixson2013). Brainstorming processes and techniques are a focus for ideation methods in different contexts (Osborn Reference Osborn1953; Taylor, Berry & Block Reference Taylor, Berry and Block1958; Sutton & Hargadon Reference Sutton and Hargadon1996; Girotra, Terwiesch & Ulrich Reference Girotra, Terwiesch and Ulrich2010). Additional theories and methods developed for design process (e.g., the flexibility principle, design by analogy, and creativity operators) may also inspire or guide scientists and entrepreneurs in undertaking their respective creative activities. The growing design science research may continually benefit scientific and entrepreneurial processes.

6.1. Case exampleFootnote ³

The research group of MIT mechanical engineering professor Douglas Hart works on fluid diagnostics, optics, and image processing. Professor Hart and Janos Rohaly, a senior lecturer, invented a single lens 3D imaging technology in 2002 on the basis of many years of scientific research. Their technology uses a single lens with a rotating, off-axis aperture to generate 3D surface images in real time, supplanting dual-lens cameras that imaged an object from different angles. This applied development effort was supported by a commercialization-oriented grant from the MIT Deshpande Center for Technology Innovation. The center funds applied development efforts to move MIT inventions to the market.

Later, at a networking event of the MIT $50K business plan competition, Professor Hart met two Harvard Business School MBA students, Eric Paley and Micah Rosenbloom, who eventually became the CEO and COO of the company they later founded. At the time, the startup team also included MIT research scientists, postdoctoral scholars, and doctoral students. Since then, they have explored the market potential of various specific applications of the 3D imaging technology, from facial recognition for security and 3D animation for entertainment to medical applications. They presented alternative application and commercialization ideas at the Cambridge Enterprise Forum Concept Clinic. That experience helped them to formulate their business plan for the MIT $50K business plan competition, where they were selected as a runner-up.

While developing prototypes, they discovered a significant need in dental imaging. With a clear target market, Brontes Technologies was incorporated in June 2003 with the mission of providing technology for dentists to scan the inside of the mouth in real time; the technology was licensed through the MIT technology licensing office. They returned to present at the Enterprise Forum Startup Clinic and received two rounds of seed capital from the audience, followed by venture capital funds from David Frankel, Flybridge Capital, Charles River Ventures, and Bain Capital Ventures in 2004. In October 2006, 3M acquired Brontes Technologies for $95 million and the company delivered its first products in 2007.

This story demonstrates how the innovation ecosystem around MIT integrates and unites the relevant scientific, design, and entrepreneurial sub-processes for innovation, as well as their respective supporting activities, to nurture innovation. As illustrated in Figure 3, the researchers at the MIT lab were responsible for the scientific sub-process. Some postdocs and students later left the lab and joined the startup company to focus on the engineering design of applications and commercial products. The two MBA students from Harvard University drove the entrepreneurial process. Scientists, engineers, and entrepreneurs worked closely to pursue innovation.

Figure 3. United innovation sub-processes in the Boston/MIT innovation ecosystem.

Meanwhile, supporting organizations also helped unite the sub-processes. For instance, the MIT Deshpande Center supported commercialization-oriented engineering design efforts. The MIT Technology Licensing Office was responsible for generating the financial returns from the invention, which would, in turn, be reallocated to support basic research activities at MIT. And, the MIT $50K Business Plan Competition and the Cambridge Enterprise Forum connected researchers, entrepreneurs, investors, and potential clients, and synchronized their complementary activities.

Clearly, more than just hosting the diverse actors in its local community, the MIT ecosystem was also able to motivate and drive the actors to think and act beyond their specialties to connect, interact, and collaborate to incorporate deep and advanced sciences in technology applications, and, in turn, take inventions from the lab to the market. The MIT innovation ecosystem is not only comprehensive, but the co-located actors are also closely connected. As a result, relevant science, design, and entrepreneurship sub-processes are united for innovation.

6.2. Building an innovation ecosystem: challenges and strategies

Inspired by the success of the MIT and Stanford innovation ecosystems (Roberts & Eesley Reference Roberts and Eesley2009; Eesley & Miller Reference Eesley and Miller2012), many cities and governments around the world have attempted to build their own ‘science parks’, ‘innovation clusters’, or ‘ecosystems’. According to the COSEPUP report of NAS, NAE, and IoM (COSEPUP 2013), there are now over 460 science parks worldwide, although very few have succeeded. In practice, although it is straightforward to use financial and policy incentives to attract industrial firms, startups, and research and education institutions to reside in a specific science park, it is also naturally difficult to foster fluid and intensive communication, interaction, and cooperation among individuals and organizations specializing in the distinct science, design, and entrepreneurship sub-processes.

The specialized individuals and organizations required for science, design, and entrepreneurship have distinctive goals and interests. Scientists are driven by curiosity, creating gadgets fascinates engineers, and entrepreneurs are keen on economic and social impacts. Such differences may limit the incentives to collaborate and unite activities Starbuck & Farjoun (Reference Starbuck and Farjoun2005). In addition, scientists, engineers/designers, and entrepreneurs receive different types of education and develop different types of personalities, knowledge bases, skill sets, capabilities, work habits, and communication styles along their respective career paths. Scientific institutions and commercial firms also require and often develop different cultures, codes of conduct, and organizational routines over time. For these reasons, the effectiveness of communication and cooperation between different types of groups for science, design, and entrepreneurship may be limited. Without necessary interactions and effective cooperation, the mechanistic aggregation of scientists, designers, and entrepreneurs in a geographically proximate region may not result in an innovation ecosystem.

What can incentivize scientists, designers (or inventors), and entrepreneurs to reach out to each other and interact, thereby effectively collaborating with and learning from each other? At the 2013 workshops organized by COSEPUP of NAS, NAE, and IoM, speakers repeatedly highlighted culture as key to the success of innovation ecosystems, as opposed to operational methods (COSEPUP 2013). Generally, culture may imply the shared expectations among a group of people about how they and others will interact. As for the innovation culture, it should center on the shared understanding that innovation is a result of collective efforts ranging from science through design to entrepreneurship and the shared belief that the value from the co-created innovation can be shared. Such a culture may motivate scientists, designers, and entrepreneurs to accommodate others’ differences and specialties and to interact, collaborate, and unite their activities. The motivated interactions and influences will further facilitate confluence and the emergence of a coherent innovative culture, which in turn, facilitates organic interactions and seamless integration.

The importance of culture for uniting the sub-processes of innovation in a local ecosystem has been implied in previous studies on Silicon Valley and Israel (Saxenian Reference Saxenian1996; Florida Reference Florida2004; Senor & Singer Reference Senor and Singer2009). In the case example of Brontes Technologies, the self-driven interactions among individuals specializing in science, design, and entrepreneurship were largely facilitated by the entrepreneurial culture of the MIT community at large. In brief, to foster the possible emergence of a culture that nurtures innovation, ecosystem builders may evangelize the united view of the innovation process and its value added for enhancing innovation.

7.1. Case exampleFootnote ⁴

In the late 1990s, a few Chinese entrepreneurs were keen to drive the wide adoption of solar-PV-generated electricity by reducing the cost of PV cells via large-scale manufacturing; however, the scientific knowledge base, design capability, supply chains, and market needs for solar PV cells were lacking in China at the time. Facing local constraints, Shi Zhengrong started Suntech Power in 2001, which later became one of the global leaders in the solar industry, and developed it rapidly through successful global engagement and integration of required resources. For instance, Suntech’s research and development mainly relied on the University of New South Wales (UNSW) in Australia (where the founder studied and worked for many years). Suntech sought and imported used equipment from Italy, Japan, and the United States, procured key raw materials from Germany and the United States, supplied its products to customers in Europe, and also received investments from international venture capitalists and private equity firms (Zhang, Luo & White Reference Zhang, Luo and White2014).

This global integration model fueled the rapid growth of Suntech without comprehensive R&D capacities in-house and without being part of a comprehensive local ecosystem for solar PV technologies. Suntech became the world’s first solar company to list on New York Stock Exchange (NYSE) in 2005 and is the global industry leader with shipments of 1.6 GW PV products in 2010. Its radical growth and successful NYSE initial public offering, which raised US$2.3 billion, attracted several other China-based solar ventures to imitate its growth model characterized by building global linkages, which also allowed them to rapidly grow into world-class technology-based businesses (see Table 3).

Interestingly, Suntech’s entrepreneurship and engineering design activities were closely linked to the fundamental research conducted at the Solar Photovoltaics Research Group of the UNSW in Australia, led by Professor Martin Green, and the applied research at Pacific Solar Pty, a spinoff and affiliate of Martin Green’s group. Shi Zhengrong, the founder and CEO (2001–2011) of Suntech, was a Ph.D. student of Professor Green from 1992 to 1995 and a research scientist, and later deputy director of research at Pacific Solar Pty Ltd from 1995 to 2000. As a scientist, Shi had published extensively about multicrystalline silicon thin film solar technologies and accumulated a number of patents. Having spent 15 years in scientific research, he has the cutting-edge scientific knowledge of solar PV technologies and the ability to comprehend the latest scientific advances in the field.

Since its founding in 2001, Suntech has maintained close ties with the R&D activities at UNSW. In 2002, Suntech signed a cooperation agreement with the Centre of Excellence for Advanced Silicon Photovoltaics and Photonics (CEASPP) at UNSW. The collaboration covered a full spectrum of activities from basic research to technology development and commercialization. One example was the jointly developed ‘semiconductor finger’ technology (Wenham et al. Reference Wenham, Mai, Tjahjono, Jia and Shi2005; Wenham, Mai & Tjahjono Reference Wenham, Mai and Tjahjono2006), which achieved 18% efficiency, compatibility with Suntech’s manufacturing infrastructure, and immediate commercialization by Suntech in 2007. The basic research grounding the semiconductor finger technology was conducted at UNSW and received some funding from Suntech. The resultant patent was jointly held by UNSW and Suntech.

In addition to directly funding scientific research at UNSW in Australia, Suntech also sponsored the visits of UNSW researchers and students to Suntech and sent its own engineers to visit UNSW laboratories. Such interactions facilitated intensive informal learning and synchronized scientific, design, and entrepreneurship efforts across regional borders. A few R&D personnel from UNSW also joined Suntech later (see Table 4). Among them, Professor Stuart Wenham split his time as the director of the CEASPP at UNSW and the CTO of Suntech. Over time, the scientific research at UNSW continually served as a main source of scientific and technological knowledge required for the product design and entrepreneurship activities of Suntech. To access the broader cutting-edge scientific knowledge relevant to solar technologies, Suntech also collaborated with other universities in China, Europe, and the United States.

This case demonstrates how Suntech united the relevant scientific, design, and entrepreneurship sub-processes for solar innovation on a global scale, which is illustrated in Figure 4. Scientists at the UNSW research lab contributed to the scientific sub-process. UNSW and Suntech engineers collaborated and contributed to the engineering design of solar cells and modules. Suntech, together with its international suppliers and investors, drove the entrepreneurship sub-process. Clearly, Suntech played a central role in coupling and uniting sub-processes across regional borders.

Figure 4. United innovation sub-processes across regions for solar photovoltaic innovation.

As Suntech cultivated local suppliers and generated spinoffs, much of its technical knowledge spilled over to other local firms over time, resulting in a cluster of globally competitive solar companies in China. Some of these Chinese companies were also founded by UNSW researchers and students and former Suntech executives, and have grown by uniting science, design, and entrepreneurship activities across regional borders. Photovoltaic shipments from Chinese firms grew rapidly from 3% of the global total in 1997 to 54% in 2010 (Mints Reference Mints2011; National Research Council 2012). As shown in Table 2, of the top 10 global PV firms by shipment in 2010, 5 were based in China. The top 10 Chinese solar PV firms are all publically listed on either the NYSE or the NASDAQ. In particular, none of these firms are state-owned companies or receive state subsidies. Policy support from the Chinese government was absent until 2007. Without the linkages that successfully united globally distributed science, design, and entrepreneurship activities, the Chinese solar firms would not have been able to grow to their current scales so quickly and, as a result, drive the diffusion and implementation of solar PV technologies across the globe.

Today, the capacity for basic research on the PV effect; the design and development of solar PV cells and application modules; and the manufacturing, sales, and installation of solar power systems are globally distributed. The leading firms in the solar industry operate and compete globally, maintaining strong ties with universities and research institutionsFootnote ⁵ , which continually advances the scientific understanding of the PV effect. The global integration of relevant scientific research, application design, and entrepreneurial activities enabled solar PV innovations via the immediate use of the latest scientific knowledge for solar PV technology design and implementation.