The influence of industrial designers and developmental technologists on design-driven innovation and related mechanisms

ABSTRACT Design-driven innovation (DDI) is a concept that promotes product innovation by giving new meaning to existing products. However, there are divergent views on who starts the DDI process and what kinds of activities are carried out within it. Therefore, this study analyzed four cases that succeeded in de-maturing the Japanese electric fan market to confirm whether there are differences in DDI processes, outcomes of new products, and post-launch spillover effects on new businesses between DDI originating from industrial designers (IDs) and that originating from developmental technologists (DTs). This study also aimed to elucidate the mechanisms by which these differences may arise. Our results showed that the difference in the way of thinking caused by the different knowledge and skills possessed by IDs and DTs affects the selection of core design concepts and product architecture. This was a fundamental factor that caused differences in DDI processes, the outcomes of new products, and the ripple effect on new businesses.


Introduction
Globalization and rapid advances in science and technology have created a situation in which products and businesses have shorter lifecycles (e.g., Heracleous, 2013).In response, companies are under pressure to promote more efficient research and development (R&D) strategies and launch new products and businesses in shorter periods of time than in the past.However, because it is not easy for companies to achieve these activities, it is becoming increasingly important not only to extend the life of maturity-stage businesses as much as possible but also to pursue a de-maturity strategy through new product developments (NPDs).In the maturity stage, a design-oriented approach that strategically utilizes industrial designers (IDs) in NPD has been advocated (Best, 2006;Borja de Mozota, 2003;Utterback et al., 2006), and the design has become increasingly important (Eisenman, 2013;Rampino, 2011).Although design is defined in variously ways in different domains (e. Pahl et al., 2007), we define design as follows: the interaction between a person -a 'user' -and the man-made environment, considering aesthetic, functional, contextual, cultural, and societal considerations (International Council of Design, n.d.).Verganti (2009) has proposed design-driven innovation (DDI) as a new method of innovation creation that differs from technology push (e.g., Abernathy & Clark, 1985;Gatignon & Xuereb, 1997) and demand pull (e.g., Brown & Eisenhardt, 1995;Rothwell et al., 1974).The definition of design in DDI is 'making sense of things,' rather than improving appearance.The act of 'creating new meaning,' a central concept in DDI, is also related to employee-driven innovation (EDI), and the possibility of extending the concept of creativity from the individual or team level to a companybased level has been noted (Cascini et al., 2022).In addition, in recent years, there has been growing interest in using DDI as an effective approach for de-maturity (e.g., Bellini et al., 2017;Conti & Chiarini, 2021;Goto, 2017;Trabucchi et al., 2017).
However, as empirical case studies on who initiates the DDI process and what kinds of activities take place accumulate, views on the rationale for the creation of new meanings in DDI and their processes are divided.In terms of the rationale for creating new meaning, the following views have emerged: finding new uses for existing products (Sugino, 2013), trend forecasting (Kembaren et al., 2014), individual ideas (Verganti & Shani, 2016), materials-based R&D (Aydin & Erkarslan, 2019), and so on.Concerning the DDI process, some have suggested that small, informal discussion activities (radical circles) and discussions with team members with cultural expertise are important (Altuna et al., 2017;Verganti, 2017).A number of research findings have asserted that IDs are suitable starting points for DDI (e.g., Dell'era & Verganti, 2011;Dell'era et al., 2011;Simoni et al., 2014;Verganti, 2006).Although developmental technologists (DTs) could also create new meanings and start the process of DDI, there are currently limited reports of DDI originating from DTs (Kubo & Sato, 2024;Verganti, 2009).In this study, IDs are defined as experts who focus on product appearance, functionality, and manufacturability and propose related concepts, while DTs are defined as experts who develop new technologies, improve existing technologies, and are responsible for realizing functionality and manufacturability based on their specialty.
Although electric fans (hereafter referred to as 'E-fans') are one of the oldest household appliances in the world, only incremental technological improvements have been made in the last 100 years following the creation of affordable E-fans with alternating current (AC) motors in Japan.Combined with the fact that inexpensive E-fans produced in newly industrializing countries began to be imported into Japan, E-fans have been recognized as a low-price product since 1990 in Japan.As a result, Japanese major electronics companies were unable to devote resources to E-fan NPD, and shipments in the Japanese market continued to decline after 2000 (Onishi, 2010).However, in 2010, bladeless E-fans and E-fans with direct current (DC)motors were launched in the Japanese market at a price that was almost 10 times higher than the volume zone price of conventional E-fans, triggering the creation of a high-end Japanese E-fan market (Tamehiro, 2022).Accordingly, major Japanese electronics companies reentered the E-fan market with high-end models, which accounted for about half of the E-fan market in value terms in FY2016 (Ougawara, 2017), and the de-maturity of the E-fan business in Japan was realized.
Against this background, we studied two cases of DDI at companies that contributed to the dematuration of the Japanese E-fan market.We found a rare DDI case with a DT origin and identified differences in DDI processes, outcomes of new products, and post-launch spillover effects between the case in which the DDI origin was an ID and the case in which it was a DT (Kubo & Sato, 2024).In this study, we considered a DDI process to be series of processes from the conception of new meanings to the social implementation of products and services with new meanings.In the present study, we add to the number of cases from the previous study and clarify the starting point of DDI in each case.Moreover, we re-confirm whether there are differences in DDI processes, outcomes of new products, and post-launch spillover effects between cases in which the DDI origins were IDs and those in which they were DTs.When differences exist, we elucidate the mechanism by which they arose.

Materials and methods
We relied on Yin's (2017) approach.Several NPD cases practiced in different companies were taken and traced the historical dynamic causal processes that led to successful NPD by chronologically process-tracing the events practiced in each company.Comparative analysis was then conducted to identify differences and similarities among the cases.Based on the data and information obtained by the above methods, the newly created meaning, the creators of that meaning, the DDI processes, the outcomes of the new products, and the post-launch spillover effects were derived.This study focuses on NPD cases of four pioneering high-end E-fans that were launched in the Japanese market from 2010 to 2015 and that triggered the de-maturity of the mature-market E-fan business.Data and information on the NPD processes of the four high-end E-fans were obtained from primary data sources as much as possible.The information for Case A was collected through reliable secondary sources.For Case C, primary data were obtained from a research group to which one of the authors belonged in 2014, and secondary data were collected for the missing pieces.In Cases B and D, questionnaires, and interviews, respectively, were used to obtain a detailed understanding of the DDI process.In Case B, answers to the questions were provided by the head of public relations at Company B's headquarters and by the founder of Company B at AXIS Inc. (on 19 June 2013, and12 April 2017, respectively).In Case D, unstructured interviews were also conducted with the DT who created the new meaning and the ID who assisted the DT and was responsible for the aesthetic design, etc. (on 12 April 2016, and23 August 2016, respectively).In all cases, data on product specifications were obtained from manufacturer-issued catalog data and Kakaku.com(https://kakaku.com/kaden/fan/),and point-of-sales data on sales performance were obtained from Growth from Knowledge Japan (GfK Japan).Furthermore, award-winning information on each product was obtained from released documents of the Good Design Award, one of the most historical design awards in Japan (Good Design Award, 2010a, 2010b, 2011, 2015, 2021).
Then, the mechanisms by which the DDI processes, outcomes of the new products, and postlaunch spillover effects of products on new business differed between the cases in which the DDI origins were IDs and those in which they were DTs were discussed.Moreover, to clarify the mechanisms that resulted in significant differences in the DDI process, outcomes, and postlaunch spillover effects, we analyzed the product architecture of the conventional E-fan and the new products in each of the four cases.Product architecture is a method of allocating product functions to components at the design stage and refers to the correspondence between functions and components as well as the interface specifications between components.The impact of product structure on corporate activities is actively discussed from various perspectives (Ulrich, 1995).
Additionally, we adopted an innovation diagram to analyze the factors that led to the significant differences in product architecture as well as post-launch spillover effects between the ID-and DToriginating DDIs, focusing on the differences in knowledge, skills, and thinking methods possessed by the IDs and DTs.The innovation diagram takes knowledge creation (research) on the horizontal axis and knowledge realization (development) on the vertical axis, and a series of innovation processes are depicted in this two-dimensional plane (Yamaguchi, 2019).

Case A
Company A is a major Japanese heavy electric appliance manufacturer that started manufacturing and selling the first E-fan in Japan in 1894 (Sudo, 1980).The E-fan business was stagnant in 2007 owing to the spread of home air conditioners.
Against this background, an ID who belonged to the design center at the head office started to develop new E-fans in 2007.In September 2010, the ID proposed a new concept to the business division: 'a cool wind-blowing object that can be placed next to the television (TV) in the living room like an audio system' (Pd Web interviews, 2011).
The DT who received the ID's concept proposed integrating it with their concept of an E-fan that can be used all year round with an air conditioner, keeping in mind that Company A's development department envisioned a lifestyle change in which people use the E-fans in conjunction with air conditioners would increase (Nagai, 2011, July 8).To realize the concept, the function of 'generating slow air movement' was important.Therefore, the DT decided to replace the AC motor that had been used in E-fans for a long time with a DC motor, which is quiet and can rotate at a low speed, and completed the concept design by making the impellers thinner and changing the number of blades from four to seven.After the product planning stage, the DT added left/right and up/down swing functions to enable efficient air circulation following a request from the business division (Nagai, 2011;Seido, 2012b).
In May 2011, a new product named SIENT (F-DLN100) was launched.To compete with the E-fans of Company B, which had already been selling E-fans with DC motors, and the those of Company D, which had entered the Japanese market the previous year, the market price of the F-DLN100 was set at 25,000 yen, about 10,000 yen less than the E-fans from the other companies.This price was much higher than the volume zone price (3,000 yen to 5,000 yen) in the Japanese E-fan market at the time.After its launch, the F-DLN100 successfully sold approximately 20,000 units in its first year on the market.In October 2011, the F-DLN100 won the Good Design Award (Good Design Award, 2011).Since then, Company A has expanded its product lineup every year, and while the number of E-fans shipped by the Japan Electrical Manufacturers' Association (JEMA) member companies has declined by 13.1%, the SIENT series achieved a compound annual growth rate (CAGR) of + 4.7% (from FY2011 to FY2014), successfully de-maturing the E-fan business (Kubo & Sato, 2024).

Case B
Company B was established by Mr. T in 2003.Since its inception, Company B has placed its commitment to stylish industrial design at the center of its management.The company had enjoyed a successful start-up period but was on the verge of bankruptcy due to the Lehman Brothers collapse in September 2008.Under these circumstances, in 2009, Mr. T selected 'air conditioning' and 'energy conservation' as keywords for his new business in view of worsening global warming issues (Seido, 2012a).At the time, major electrical companies were exiting the E-fan business one after another, but Mr. T was convinced that people did not want to use air conditioning, even in the summer, and that there were areas that conventional E-fans had not yet explored or challenged.
Based on his own past experiences, Mr. T arrived at the concept of 'experiencing the pleasant feeling of a breeze outdoors and not getting tired even after prolonged exposure to the air movement' (Koyasu, 2013, p. 9).To realize this concept, Company B started developing a new E-fan in January 2009 (Terao, 2017).Because the air from conventional E-fans is a swirling current, the fan creates 'hard' air movement, and people become tired after being exposed to it for a long time.Mr. T and his colleagues made full use of a 3D printer, repeatedly made prototypes of about 50 impellers, and conducted experiments based on his self-taught fluid mechanics.They invented an impeller structure with five inner and nine outer blades facing away from each other, resulting in the disappearance of the swirling flow and the realization of their concept (Kishi, 2013;Koyasu, 2013;Terao, 2017).Mr. T himself conducted the detailed functional and appearance design, drawing a design sketch of the new E-fan (Y.Kubo, personal communication, June 19, 2013), and he dared to select a typical E-fan appearance for it (Nikkei Design, 2012;Terao, 2017).In the process of product planning and preparation for mass production, Mr. T found that rotating the new impeller with an AC motor caused too much airflow; this problem was resolved by using a DC motor, leading to the completion of the new product (Terao, 2017).
Thus, in April 2010, the new E-fan, named GreenFan (EGF-1000), was launched.Despite the high price of 35,000 yen 12,000 units -twice as many as the target -were sold in the first year of sales (Terao, 2017).In October 2010, GreenFan also won the Good Design Award (Good Design Award, 2010b).

Case C
Company C was founded in the United Kingdom and sells their products in 52 countries.Dr. D, the founder, studied industrial design and furniture design at the Royal College of Art.With the creation of a new vacuum cleaner without a paper pack and a hand dryer named Air Blade, Company C is positioned as a company that develops electrical products with technical features and a stylish appearance design.In 2005, a DT discovered the induced airflow phenomenon during the development of an air-blade hand dryer.The aim was to reduce paper consumption in consideration of global warming issues and the idea was to apply it to E-fans (Mishina & Mishina seminar, 2013).However, the induced airflow phenomenon had a negative impact on hand dryers due to dust entrapment and spraying, which he reported to Dr. D, then the chief engineer at Company C. Dr. D considered that the use of the induced airflow phenomenon could solve his own previous dissatisfaction with conventional E-fans (Abe, 2010;Mishina & Mishina seminar, 2013;Nusca, 2012).The first problem in conventional E-fans was the sense of fatigue that people experience when the air is divided by the rotating blades, resulting in uneven airflow.The second was the safety issue of rotating blades injuring the human body.The third was the cleaning load for the safety guard protecting people from the rotating blades and the blades themselves.Thus, Company C intended to develop a new bladeless E-fan with impellers inside the main body, which would send high-pressure air into a cavity ring and blow it out through a narrow gap in the ring (Yoshida, 2009).Mr. Alex Knox, a senior designer at Company C, described the background at this time in an interview: 'When we design a product, we start right at the basics.It's all about solving the problem.We'll often use new technology or a new approach to solve that problem' (Nusca, 2012).
A project to develop this new E-fan, named Air Multiplier, was launched.In the first stage of the project, the project members created design sketches of the shape of the product when the induced airflow phenomenon was applied to it, and the conceptual design involved the mutual influence of mechanical and aesthetic design.Subsequently, product planning and detailed product design were conducted simultaneously.Numerous product prototypes were developed, including a new mixedflow impeller for sucking air into the main unit, and after four years of development, the product was commercialized (Abe, 2010).Thus, the DT's discovery of the induced airflow phenomenon and Dr. D's beliefs combined to lead to the new E-fan's development by Company C (Dyson, 2021, pp. 162-163).
The first Air Multiplier model, the tabletop AM01, was launched in the Japanese market in April 2010.Although the Air Multiplier was extremely expensive with a retail price of around 40,000 yen, it attracted a great deal of attention in the Japanese market due to the reduction of fatigue caused by uneven airflow, the stylish bladeless design, improved safety due to the absence of blades, and the reduced cleaning load.Since then, the product lineup has been enhanced with the addition of two tower models, and in October 2010, the Air Multiplier product line won Japan's Good Design Grand Award (best in 2010) (Good Design Award, 2010a).Company C also succeeded in developing a bladeless hair dryer by applying the induction airflow technology developed for the Air Multiplier.Launched in May 2016, this was the first beauty appliance developed by Company C, allowing them to expand into the beauty appliance market (Serizawa, 2016).

Case D
Company D is a major electronics company that has manufactured and sold E-fans in Japan since 1913.Company D had been promoting natural air movement and design features, such as a thinner appearance, since 1988 (Hirano, 2009).However, after 2000, Company D was affected by the lowprice orientation of the E-fan market, and they discontinued NPD of E-fans in FY2005.In 2008, Mr. O, a DT who was in charge of airflow technology in Company D's R&D department, began exploring the application of the induced airflow technology developed by the same department in 1993 to E-fans.In 2009, based on Mr. O's desire to create things that he himself thought interesting, he devised a pot-shaped blower that generates airflow similar to natural air movement with high linearity and excellent functional beauty (Y.Kubo, personal communication, April 12, 2016).Inspired by the launch of Company C's Air Multiplier and the power supply shortages caused by the Great East Japan Earthquake in March 2011, in the fall of 2011, Mr. O began conducting informal discussion activities on product concepts and detailed designs with diverse members of other sections (Y.Kubo, personal communication, April 12, 2016).These included DTs from other fields; product planners from business divisions; and Mr. T.O, an ID from Company D's headquarters (Y.Kubo, personal communication, April 12, 2016).Based on these informal discussion activities, Mr. O formally proposed a product concept for the pot-shaped blower using induced airflow technology to his division manager in December 2011, but the proposal was rejected for various reasons (Y.Kubo, personal communication, April 12, 2016).
When Mr. O and his colleagues were revising the concept, a product planner from a department unrelated to the E-fan business proposed a spherical design, similar to a soccer ball.Initially, Mr. O rejected this proposal, but Mr. T.O convinced him to accept the spherical shape, and the prototype shape was modified (Y.Kubo, personal communication, August 23, 2016).In the spring of 2013, the division manager, who was considering entering the high-end E-fan market, accepted the revised E-fan concept based on Mr. O's insistence on focusing on the new E-fan concept he had devised to differentiate themselves from other companies (Y.Kubo, personal communication, April 12, 2016).Just prior to the press release for the new product, the product planning manager, who was concerned about user acceptance of the novel shape, requested that a user acceptance survey be conducted.The results of this survey confirmed that there was a market need for this novel spherical blower, despite its high price point.However, the launch of the new product was delayed for one year (Y.Kubo, personal communication, May 16, 2019).
In May 2015, the novel E-fan named Bladeless Air Generator 'Q:' Ball Fan was launched, priced at 40,000 yen, which was more expensive than other companies' high-end E-fans.Despite this, Air Generator 'Q' almost achieved its sales target (4,000 units) for the first year, and it won the Good Design Award in October of the same year (Good Design Award, 2015).Air Generator 'Q' was hardly a success from a business perspective because the CAGR of the sales volume trend for the first three years after its launch (FY2015-FY2018) decreased by 3.4%, while the overall shipment volume of E-fans by JEMA member companies decreased by 1.6%.However, Company D used the E-fan's technology to create Air Hospitality, a space solution that incorporated an induced airflow mechanism into ceilings and partitions to provide a static and air-conditioned experience with little airflow sensation (Kubo & Sato, 2024;Panasonic News Release, 2020).

Case comparison
Based on the cases introduced in the previous section, Table 1 summarizes the new created meaning, the origin of the new meaning creation, the DDI processes, the outcomes of new products, and the ripple effects that these products had on new businesses.In Cases A and B, new meaning originated from the ID, and in Cases C and D, it originated from the DT.The new meanings created in all four DDIs were similar: the realization of a stylish appearance and airflow similar to natural air movement.In addition, four DDIs also realized lower power consumption through the combined use of E-fans and air conditioners as well as lower power consumption of the E-fans by utilizing DC motors.
In the DDI processes for Cases A and B, IDs and DTs worked together to promote concept design based on new meanings drafted by the ID.Subsequently, the product planning and mass production preparation stages were conducted, leading to commercialization.Specifically, the DDI process in Case A corresponds to the Type 4 process reported by Kim and Lee (2016) based on a survey of six Korean consumer electronics companies, which is considered more likely to lead to market success.In this process, the IDs lead the entire DDI process, a typical process used by major Japanese electronics companies.In contrast, the DDI processes of Cases C and D were rare cases in which the new meaning was drafted by the DT.Various members, including the DT and ID, participated in the DDI process, and the conceptual designs were completed under the mutual influence of functional and aesthetic design.The subsequent processes also did not proceed according to the typical steps as reported by Kim and Lee (2016) because many product prototypes, including the new development of major parts based on the new phenomenon, and user acceptance surveys for novel shapes were conducted.The DDI processes in Cases C and D were also novel and did not correspond to any of the previously reported DDI processes.
Regarding the outcomes of new products in four DDIs, in terms of sales performance, all new products in the four DDIs achieved their first-year sales plans, despite their higher price compared to conventional E-fans.Furthermore, because the CAGR of the retail sales volume of the new E-fans exceeded that of the E-fan market during the same period in Cases A, B, and C, all four DDIs succeeded in establishing the high-end E-fan market and contributed to the de-maturity of the E-fan business.In addition, all four DDIs received the Good Design Award, which evaluates the novelty of the design concept.Moreover, while the appearance of Cases A and B followed that of conventional E-fans, those of Cases C and D were novel and differed greatly from conventional E-fans.In terms of spillover effects, although each company succeeded in developing the high-end E-fan market, Companies A and B did not enter new businesses based on the technologies developed in the DDIs.In contrast, in Cases C and D, each company utilized the technologies introduced in the DDIs and created new businesses.
Thus, it was confirmed that significant differences in the DDI processes, the appearance of the new products, and the post-launch spillover effect on the new businesses were caused by the drafters of the new meanings.

Success factors of four DDIs
As mentioned above, in all four DDI cases, the outcomes of new products were excellent.The four DDIs created the high-end E-fan market in Japan and contributed to the de-maturity of Japanese E-fan market.As described in the 'Case Introduction' section (Cases A -D), at the time, Japanese society was under pressure to respond to global warming issues and the power shortages caused by the Great East Japan Earthquake in 2011.To solve these issues, Japanese customers and society demanded lower power consumption through the combined use of E-fans and air conditioners as well as lower power consumption of the E-fans themselves.The success of the four DDIs was achieved not only by responding to customer needs for stylish design but also to the social needs of a changing socio-cultural model in Japan.

Factors that caused differences in the DDI process and outcomes
To elucidate the mechanism by which significant differences in the DDI process, the appearance of the new products, and the post-launch spillover effects on new businesses arose, we analyzed the product architecture of the conventional E-fan and the new products in each of the four cases.Because the coordination tasks inherent in product design influence the organizational design required to realize the creation of the product, it is known that there is a relationship between product architecture and organization (Sanchez & Mahoney, 1996).Hence, different product architectures are likely to cause variations in the DDI process due to differences in coordination tasks and organizational forms.In addition, different product architectures are likely to affect the appearance of the product owing to differences in the correspondence between functions and components and in the components used.
The conventional E-fan is defined as equipment that generates air movement to obtain a cool feeling.Therefore, the top-level function of the conventional E-fan is 'to blow air in an arbitrary direction.'To realize this function, there exist two key subordinate functions: one for generating wind and the other for changing the airflow direction.The 'generating air movement' function is realized by the 'forward blowing' and 'air movement speed adjustment' functions for adjusting to the arbitrary air movement speed.The 'forward blowing' function is realized by forming a pressure difference between the front and rear of the space, and the 'air flow distance difference generation' and 'rotation' functions exist as subordinate functions of the 'forward blowing' function.The 'air flow distance difference generation' function is realized by the impeller, and the 'rotation' function is realized by the AC motor.These two functions are combined to form the core design concept (CDC) of the conventional E-fan.The CDC functions at the top of the hierarchy of functions and characterizes a product like the function that generates dynamic power in an automobile.Once the CDC is defined, the related design elements (complementary functions) are expected to follow (Henderson & Clark, 1990).The 'air movement speed adjustment' function of conventional E-fans is achieved by changing the revolution speed of the AC motor.To control the speed of the AC motor, taps that change the strength of the magnetic field according to the number of coil windings are provided, and air movement speed adjustment is achieved via the magnetic field strength change switch.Furthermore, the conventional E-fan has a 'safety measure' function to protect fingers from the rotating blades of the fan and to prevent burns from the heat of the motor, an 'air direction adjustment' function for the rotating blades, and an 'installation stability' function to support the E-fan.The 'safety' function is provided by the enclosure and the front and rear guards, the 'wind direction adjustment' function is provided by worm gears via a link mechanism, and the 'installation stability' function is provided by the support column and the pedestal that supports it.Figure 1 shows the product architecture of the conventional E-fan.
In Cases A and B, the top-level function of the new E-fans is 'to blow air in an arbitrary direction,' which is same as that of conventional E-fans.In addition, the functional structure, the relationship between functions and components, and the CDC of the new E-fans in Cases A and B are almost identical to those of conventional E-fans.The differences between conventional E-fans and the E-fans in Cases A and B are as follows: (1) the conventional E-fan's AC motor has been replaced by a brushless DC motor as a means of realizing the 'rotation' function, (2) a touch-type electronic switch is used as In Cases C and D, the top-level function of the new E-fans is 'to blow air in an arbitrary direction,' which is same as that of conventional E-fans and the new E-fans in Cases A and B. However, the functional structure and the relationship between functions and components of the new E-fans in Cases C and D differ significantly from those of conventional E-fans and those in Cases A and B. In Cases C and D, the 'generating air movement' function is achieved by the 'induced airflow' function, which results from the surrounding air being drawn into the cavity by the viscous effect of the jet stream.To achieve this function, the 'air flow distance difference generation' and 'rotation' functions are required to form a jet stream, and the 'air flow direction control' function is required to control the direction of the air flow.In Cases C and D, these three functions are integrated to form the CDC.Each of these functions is realized by the components of impeller, the DC motor, and the induced airway, respectively.However, the components that realize the 'air movement speed adjustment' function are virtually the same as those in Cases A and B. In Cases C and D, the CDC has been changed so that the rotor blades are now housed in the enclosure, eliminating the need for the parts that were required as a 'safety measure' function in Cases A and B, that is, the front and rear guards.Figure 3 shows the product architecture of the new E-fans in Cases C and D. As seen, the CDC of Cases C and D is much different compared with Q that of conventional E-fans or the new E-fans in Cases A and B. Accordingly, the product architecture of the new E-fans in Cases C and D is also significantly different from that of conventional E-fans and the new E-fans in Cases A and B. Thus, the DDIs in Cases C and D could be considered radical innovations.
Because the CDC and product architecture of the new E-fans in Cases A and B were almost identical to those of conventional E-fans, the coordination tasks and the organizational structure of DDIs in Cases A and B were also similar to those of conventional E-fans.As a result, the DDI processes in Cases A and B were considered to follow the NPD process of conventional E-fans.In addition, because the CDC and product architecture of the new E-fans in Cases A and B were same

Mechanisms by which the DDI origin causes differences
In this subsection, we analyze the factors that led to the significant differences in CDC, product architecture, and post-launch spillover effects on new businesses between Cases A and B and Cases C and D, focusing on the differences in knowledge, skills, and thinking methods possessed by the IDs and DTs.
Figure 4 shows the DDI processes for Cases A and B and Cases C and D on the innovation diagram.The 'soil of knowledge' refers to scientifically meaningful knowledge that does not have economic or social value.When the knowledge in the 'soil of knowledge' is connected to the needs of society or customers and sprouts from the 'soil of knowledge,' the scientific knowledge comes to have economic or social value, and a so-called invention is realized (Yamaguchi, 2019).
The left half of Figure 4 shows the DDI processes for conventional E-fans and the new E-fan in Cases A and B. Point a shows the state in which the scientific knowledge of 'airflow by rotation of the impeller' was created.On one occasion, this scientific knowledge was combined with societal or customer needs for 'generating air movement to feel cool,' leading to the invention of the E-fan at Point b.Subsequently, through incremental development activities, the functions and performance of the E-fan were improved (Point c).As mentioned above, the DDIs in Cases A and B were incremental innovations involving component changes.Therefore, the DDIs in Cases A and B are a change from Point c to Point d.Most IDs lack sufficient knowledge and skills in science and technology.While some are well versed in the technology, few fully understand the science behind the technology, making it difficult for them to choose to return to the 'soil of knowledge.'However, IDs are adept at gathering user and market information and have extensive external networks, good foresight, and strong trend forecasting capabilities (e.g., Perks et al., 2005;Walsh, 1996).Hence, IDs seek novelty in functional and aesthetic concepts, and DTs support engineering aspects such as mechanical and electrical design to realize IDs' innovative concepts (e.g., Marsili & Salter, 2006).Because the DDIs in Cases A and B were processes in which DTs proceeded with product design based on the concepts that IDs devised from an aesthetic perspective, the CDC did not change from the conventional E-fan, and incremental improvements were made by changing some components.As a result, the DDI processes in Cases A and B generally followed those of conventional E-fans, and the appearance of the new E-fans was also similar to that of conventional E-fans.In contrast, because DTs have knowledge and skills in science and technology, DTs are knowledgeable regarding science and technology; however, their user needs are less concerned with market trends.Hence, they tend to prefer innovation through scientific and technological change.In Cases C and D, they did not opt for incremental improvement but, instead, chose to return to the 'soil of knowledge.'Subsequent emergent activities allowed them to obtain new knowledge.As a result of this process, the CDC and product architecture in Cases C and D differed from those of conventional E-fans and the new E-fans in Cases A and B. Accordingly, the DDI processes and the appearance of the products in Cases C and D also differed significantly from those of conventional E-fans and the new E-fans in Cases A and B.
In Cases C and D, the DTs entered the beauty appliance hair dryer market (Point i) and created a new business of space solutions that allowed users to experience air conditioning with static and low airflow (Point j), respectively.These were based on the new scientific knowledge that DTs had achieved by returning to the 'soil of knowledge' and engaging in emergence activities.It can be considered easier to create new scientific knowledge or discover overlooked scientific knowledge by DTs returning to the 'soil of knowledge,' increasing the possibility of acquiring the means to meet the needs of customers and society.The reason for the large difference in the post-launch ripple effect on new businesses between Cases A and B and Cases C and D can also be attributed to the difference in the knowledge and skills possessed by IDs and DTs.
Thus, the difference in the way of thinking caused by the different knowledge and skills possessed by IDs and DTs is the fundamental factor that resulted in differences in the DDI process, the appearance of the new product, and the post-launch spillover effect on the new businesses.application on the vertical axis.Among these, those with a change in meaning corresponding to the right half of the figure are DDIs, and those in which both a change in meaning and the application of new technology are present are called technology epiphany.It is considered appropriate to judge the presence or absence of new technology application by the presence or absence of CDC changes.Based on the above criteria, the conventional E-fans are positioned in the lower left of the figure .Although the E-fans in Cases A and B are DDI, these are general DDIs positioned in the lower right of the figure because they are regarded as no new technology application owing to no change in CDC; however, there were changes in the meaning.In contrast, the E-fans in Cases C and D are positioned in the upper right of the figure and correspond to technology epiphanies because there are both change in meaning and new technology application by changes in CDC.In the four DDI cases, the new meanings created were similar, but the DDIs from IDs and DTs were classified as different DDIs.The core concept of DDI is the 'creation of new meaning' and that technology push and demand pull should be construed as a way to create new meanings.
The cases in this study also illustrate the advantages and disadvantages of ID-type and DT-type DDIs: While the ID-type DDIs can expect steady sales of new products due to concepts that are consistent with the company's strategy and excellent aesthetics, they have the problem of being easily imitated by other companies because existing technologies are followed.However, DTs can develop new technologies based on the creation of new scientific knowledge or the discovery of overlooked scientific knowledge, and DT-type DDIs can develop novel products and even new businesses utilizing new technologies.In contrast, there are concerns that the probability of development success is lower, the development period is longer, and development costs are higher than in the case of the ID-type DDIs.

Conclusions
In this study, we examined four DDI cases, of which two featured IDs as the creators of new meaning and two featured DTs.In all four DDIs, new meanings were created, and these were similar, but the DDIs from IDs and DTs were classified as different DDIs.The new market creation for high-end E-fans and the de-maturity of the E-fan market were successfully realized through a synergy between the new meanings created and the changing socio-cultural model in Japan.The IDs, DTs who were the originators of DDI, and their colleagues sensed the needs of customers and changing socio-cultural models, which led to DDI success through each other's strengths.
In contrast, this study also confirmed that there were significant differences in the DDI processes, the appearance of the new E-fans, and the post-launch ripple effect on new business between the IDand the DT-originated DDIs.In the two DDIs that originated from IDs, the DDI process follows the general ID-driven NPD process, albeit with some differences, and the appearance of the new E-fans was similar to that of conventional E-fans because the CDC followed the previous one, and only the component units were changed.In addition, based on the knowledge obtained from the DDI, no new business was created.In contrast, the DDI processes of the two DT-originated DDIs were series of trial and error and differed from the conventional process.Furthermore, the appearance of the DToriginated E-fans was much different from that of conventional E-fans and the E-fans developed by IDs because a CDC completely different from the previous one was adopted, and the functional structure and components were changed accordingly.Moreover, new businesses were created based on the new scientific knowledge acquired in the DDI process in two DT-originated DDIs.
In addition, we clarified that the different knowledge and skills possessed by IDs and DTs resulted in differences in their ways of thinking, which resulted in changes to CDC and product architecture.These changes impacted coordination tasks, organization, and the appearance of the products.Thus, the difference in the way of thinking caused by the different knowledge and skills possessed by IDs and DTs is the fundamental factor that causes differences in the DDI process, the appearance of the new product, the post-launch spillover effect to the new businesses, and DDI classification.
In the conventional DDI theoretical framework, those who were skilled in design capability were assumed to be creators of new meanings (Verganti, 2017), and almost all DDI cases shown thus far have been cases of creating new meanings by following or combining existing technologies.Therefore, this study contributed to the elaboration and extension of DDI theory in five ways.(1) DDI cases originated by DT were found and revealed that there are two types of DDI: ID-and DT-originated.(2) In DDI cases originating from DT, the cases in which new meanings were created by adopting new technologies were found.
(3) It was clarified that there were many differences in the DDI process, outcomes, postlaunch ripple effects, and DDI classification between DDI originating from ID and that originating from DT. (4) It was clarified that the fundamental factor that affects the DDI processes, outcomes, and post-launch ripple effects was the difference in the way of thinking caused by IDs' and DTs' different knowledge and skills, contributing to the clarification of part of the mechanism of new meaning creation.(5) It became possible to consider aesthetic design, functional design, and innovation in an integrated manner through product architecture.In addition, this study's practical contributions to the practice of DDI are threefold.(1) The advantages and disadvantages of ID-type and DT-type DDIs were identified.(2) The informal discussion activities were demonstrated to also be effective in DDI in large companies.(3) The importance and effectiveness of utilizing DTs in DDI aimed at de-maturation were clarified.Based on our results, companies can use different DDI originators depending on the resources they have and the results they aim to achieve.
In general, a design-oriented approach that strategically utilizes ID in NPD is advocated in mature industries.In contrast, our results indicate the importance of utilizing not only IDs but also DTs hitherto neglected to achieve de-maturation.To put this into practice, it is considered important to secure the resources of the R&D department and to ensure DT's activities and informal discussion activities even during the maturity stage.Through the above activities, extending the life of mature businesses as long as possible and dematuration are important ways for companies to counter the shortening lifecycles of their products and businesses.
However, this study was conducted in the limited business field of high-end E-fans, and it is necessary to conduct studies on a wider range of business fields to determine the universality of the results.We intend to conduct further study and contribute to the refinement of DDI theory and the brushing up of practical methods.

Figure 1 .
Figure 1.Product architecture of conventional E-fans.The area above the dotted line indicates the functional area, and the area below indicates the entity (component) area.The functional area enclosed by the bold blue line is the CDC.(Source: prepared by the authors)

Figure 2 .
Figure 2. Product architecture of the new E-fans in cases a and B. The area above the dotted line indicates the functional area, and the area below indicates the entity (component) area.The functional area enclosed by the bold blue line is the CDC.(Source: prepared by the authors)

Figure 3 .
Figure 3. Product architecture of the new E-fans in cases C and D. The area above the dotted line indicates the functional area, and the area below indicates the entity (component) area.The functional area enclosed by the bold blue line is the CDC.(Source: prepared by the authors)

Figure 4 .
Figure 4. Innovation diagram of the four DDI cases.The blue-colored area represents the 'soil of knowledge.' the photographs at points b, c, d, g, h, i, and j show an early mass-produced E-fan (1928), the conventional E-fans (1979: left, 2008: right) before the introduction of high-end E-fans, the E-fan developed in case a (left), the E-fan developed in case B (right), the E-fan developed in case C (left), the E-fan developed in case D (right), a hair dryer newly commercialized as the spillover effect of case C, and an integrated air conditioning system for living space newly commercialized as the spillover effect of case D, respectively.(Source: Prepared by the authors with reference to Yamaguchi (2019), p. 52, Figure 3.2)

Table 1 .
Summary of cases A -D.
(Source: prepared by the authors)