Get 20M+ Full-Text Papers For Less Than $1.50/day. Start a 14-Day Trial for You or Your Team.

Learn More →

Empirical study on innovation motivators and inhibitors of Internet of Things applications for industrial manufacturing enterprises

Empirical study on innovation motivators and inhibitors of Internet of Things applications for... ETH Zurich, Product Development Group Zurich, Leonhardstrasse 21, Industrial manufacturing enterprises in export-oriented economies rely on product or 8092 Zurich, Switzerland service innovation to maintain their competitive advantage. Decreasing costs of Full list of author information is available at the end of the article computing power, connectivity and electronic components have facilitated a wide range of innovations based on Internet of Things (IoT) applications. However, only few successful IoT applications specific to industrial manufacturing enterprises are known. Although academics have been investigating challenges related to realising IoT, existing literature does not explain this situation integrally. Therefore, interest and engagement in and motivators and inhibitors of IoT application development and deployment are investigated based on a literature review and empirically based on a survey with N = 109 participants from enterprises in the Swiss metal, electrical and machine industries. Most enterprises are interested and are often engaged in IoT application development. Improving service and aftersales activities through IoT applications is a common motivator. Inhibitors from four domains hinder the development of IoT applications: business, organisational, technological and industrial. Business and organisational inhibitors are perceived to be more challenging than the technological and industrial ones. The business and organisational issues presented herein have essential impacts on the success of innovation in IoT applications. The results indicate future research directions for the innovation and development of IoT applications, and they can be used by organisations interested in IoT-based innovations to refine policy and decision-making. Keywords: Internet of things, Digitalisation, Industry survey, Innovation, Motivators, Inhibitors Background Context The decreasing costs of computing power, connectivity and electronic components have facilitated the realisation of the vision of Internet of Things (IoT) and have created poten- tial for all sorts of applications. Various media channels, the World Economic Forum (Schwab 2015) and academic researchers (Brynjolfsson and McAfee 2014)haveclaimed that a technology-driven revolution is underway that would change the way mankind lives and runs the economy. At the core of this revolution is the merger of physical and © The Author(s). 2018 Open Access This article is distributed under the terms of the Creative Commons Attribution 4.0 International License (http://creativecommons.org/licenses/by/4.0/), which permits unrestricted use, distribution, and reproduction in any medium, provided you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons license, and indicate if changes were made. (2018) 7:10 Heinis et al. Journal of Innovation and Entrepreneurship Page 2 of 22 digital spaces as cyber-physical systems. “Cyber-physical systems (CPS) will transform how humans interact with and control the physical world” (Rajkumar et al. 2010). The term is defined as follows: “cyber-physical systems (CPS) are physical and engineered systems whose operations are monitored, coordinated, controlled and integrated by a computing and communication core” (Rajkumar et al. 2010). The communication capa- bilities of such systems drive the technological revolution because they allow multiple systems to be connected, thereby creating the IoT. The IoT paradigm is the result of the convergence of three perspectives: (1) thing-oriented, (2) Internet-oriented and (3) semantic-oriented (Atzori et al. 2010). In other words, physical things can be sensed or can sense data automatically (1). The data are then communicated automatically to other things or humans (2). The data are interpreted and evaluated automatically to derive meaning (3). These three perspectives help realise the vision of an Internet containing information about the physical world without depending on human input (Ashton 2009). Technological progress in the information and communication technology (ICT) domain has reduced the costs of computing power, connectivity and electronic components. This decrease has helped the IoT to become increasingly real and has created potential for a wide range of promising innovations. Unsurprisingly, there is a lot of hype about IoT (Burton and Walker 2015). Clearly, several IoT applications are appearing in domains such as smart homes, wearables and smart cities. Well-known IoT applications are often desirable consumer gadgets (e.g. colour-changing light balls or fitness trackers). Known applications with a positive economic or ecological impact remain scarce (e.g. park- ing or bin fill level monitoring) (SAS Institute Inc 2016). In industrial manufacturing enterprises of specialised and export-oriented economies, even fewer IoT-based innova- tions are known, despite the opportunities offered by various technological enablers. The question then is, what inhibits the development and deployment of IoT applications in industrial manufacturing enterprises and thus prevents potential innovations—missing motivation or existing inhibitors? Need Although the potential of IoT and the challenges associated with realising it have been reviewed and discussed conceptually (Saarikko et al. 2017;Russo et al. 2015; Lee and Lee 2015;Lietal. 2014; Perera et al. 2014;Khanetal. 2012), the literature does not cover well the motivators and inhibitors of IoT application development and deployment among industrial manufacturing enterprises. The scope of the key challenges identified is generic and valid for the entire IoT ecosystem—for example Naming and Identity Management (Khan et al. 2012)—but it might not be equally relevant for individual enterprises aim- ing to develop specific IoT applications. The existing theoretical work is based mainly on conceptual models and is not backed up by empirical data. Existing empirical works on digitalisation, the fourth industrial revolution and IoT do not or only partially cover inter- est, engagement, motivators and inhibitors of the development and deployment of IoT applications among industrial manufacturing enterprises (Table 1). Studies that do cover motivators and inhibitors do not focus specifically on IoT application development among industrial manufacturing enterprises (Geissbauer et al. 2016; Weiss et al. 2016); instead, they target all sorts of industries (Twentyman and Swabey 2015;SAS InstituteInc 2016)— for example healthcare, retail and consumer goods—or they survey large enterprises (LEs) only. (2018) 7:10 Heinis et al. Journal of Innovation and Entrepreneurship Page 3 of 22 Table 1 Comparison of existing empirical studies on IoT development and deployment, IoT in general, digitalisation and fourth industrial revolution Publication Main focus Covers Covers Covers Participant’s Enterprise Industry Location (2 digit textitN Type of interest and motivators inhibitors role type ISO) (approx.) research engagement and challenges (Dijkman et al. 2015) IoT business models, No No No IoT n/a n/a Global (mainly 103 Academic building block professionals NL, US) identification (Geissbauer et al. 2016) 4th industrial No (Yes) Yes Chief digital LE Industrial Global 2000 Industrial revolution, officers, senior product expectations in executives suppliers 5 years (Geppetal. 2015) Engineering trends, (Yes) No No Engineering LE Engineering- DE 30 Academic 4th industrial professionals to-order revolution (ETO) (Greif et al. 2016) Digitalisation, degree No No (Yes) Industry SME All CH 300 Industrial of digitalisation professionals (Hsu and Lin 2016) IoT services, usage No No No Consumers n/a n/a TW 508 Academic intentions of consumers (Kinkel et al. 2016) Digitalisation, (Yes) No No Industry LE and MEM DE 150 Industrial competencies for professionals SME digitalisation (SAS Institute Inc 2016) IoT deployment No Yes Yes IoT LE All Global 75 Industrial process professionals (Skinner 2016) IoT as service, service No No No Industry LE and ICT and IoT Global 900 Industrial provider perspective professionals SME service providers (Twentyman and Swabey 2015) IoT products, smart Yes Yes Yes R&D, n/a Retail, Global (mainly 200 Industrial product development innovation, healthcare, US, GB) product manufac- development turing executives (Weiss et al. 2016) Digitalisation, degree No (Yes) (Yes) Senior SME All IT 53 Academic of digitalisation executives (2018) 7:10 Heinis et al. Journal of Innovation and Entrepreneurship Page 4 of 22 Small- and medium-sized enterprises (SMEs) with fewer than 250 employees have not surveyed in this regard, although the exports of high-wage economies are domi- nated by industrial manufacturing SMEs—for example, the Swiss metal, electrical and machine industries (MEM) are responsible for more than 30% of the total goods exports, and the majority of the enterprises in these industries are SMEs (Swissmem 2016). Export-oriented industrial manufacturing enterprises rely on innovations to maintain their competitive advantage in current globalised markets (Kaleka 2002). Raymond et al. (2018) argued that information technology (IT) capabilities can be used for innovation purposes in industrial SMEs. Thus, IoT-based innovations can help achieve competitive advantages in globalised markets. Studies in the literature do not exhaustively cover the motivators or inhibitors of IoT- based innovations but focus on a limited range of topics such as privacy issues or data analytics capabilities. By focusing on IoT products (smart products) only and excluding enterprise-internal IoT applications, Twentyman and Swabey (2015) analysed one per- spective on IoT application development in depth but missed out on providing a holistic view. A holistic and consolidated study on the motivators and inhibitors of IoT application development from the perspective of manufacturing enterprises is needed to understand the entire system of technology-based innovations, such as access to technology, business and financial issues and research and development knowledge and skills. Only a holis- tic view of the system allows us to compare the relevance of individual motivators and inhibitors effectively, as well as to define measures that can foster IoT-based innovations in industrial manufacturing enterprises. Task The present study investigates the interests and engagement of industrial manufacturing enterprises in IoT-based innovations and aims to provide a holistic understanding of the motivators and inhibitors of innovation in IoT application development and deployment in these enterprises. This knowledge is necessary to refine policy and decision-making in governments or industry associations interested in fostering IoT-based innovations or in enterprises operating and innovating in the era of technology-driven digital trans- formation. The authors address three specific research questions (RQ). They investigate (1) interest and engagement in, (2) motivators and (3) inhibitors of development and deployment of IoT applications in industrial manufacturing enterprises by presenting two conceptual models based on a literature review—one on motivators and one on inhibitors—and empirical data from a survey with 109 participants from Swiss MEM industries. 1 RQ1. Are manufacturing enterprises interested and engaged in the development and deployment of IoT applications? 2 RQ2. What benefit (added value) do enterprises expect from the development and deployment of IoT applications? 3 RQ3. Which inhibitors hinder the development and deployment of IoT applications? Interest and engagement (RQ1) The first RQ targets the interest and engagement in the development and deployment of IoT applications. In relation to the first RQ, three hypotheses (H) can be formulated. These hypotheses claim differences in interest and engagement based on the type of (2018) 7:10 Heinis et al. Journal of Innovation and Entrepreneurship Page 5 of 22 enterprise (SME or LE) and based on the importance rating of manufactured products integrated into an IoT application (digitalised products). Developing and deploying IoT applications is a form of innovation. Heck (2017) discussed the characteristics relevant to the innovation capabilities of SMEs, which differ from those of LEs. These differences allowed us to hypothesise different interests and levels of engagement in IoT application development and deployment for SMEs and LEs. H1a. Importance of digitalised products is rated differently depending on the type of enterprise. H1b. Level of engagement in IoT application development and deployment differs depending on the type of enterprise. H1c. Enterprises that assign more importance to digitalised products are more likely to engage in the development of IoT applications. Motivators (RQ2) The second RQ targets the reasons and motivators for the development and deployment of IoT applications. A conceptual model is needed to holistically collect and map the moti- vators driving the development and deployment of IoT applications. Thus, the first step to answering RQ2 would be the development of a conceptual model capturing motivators (RQ2a). The differences between SMEs and LEs allow us to hypothesise different moti- vators for different enterprise types (H2a). Furthermore, engaged enterprises may have different motivators than non-engaged enterprises (H2b). RQ2a. Which conceptual model allows us to holistically collect and map motivators? H2a. SMEs and LEs rate the importance of motivators differently. H2b. Engaged enterprises and non-engaged enterprises rate the importance of motivators differently. Inhibitors (RQ3) The third RQ targets the inhibitors and challenges associated with the development and deployment of IoT applications. Inhibitors can be identified and collected from the lit- erature on the topic. A conceptual model is needed to map the inhibitors exhaustively (RQ3a). Furthermore, inhibitors can be identified based on the statements of industry members (RQ3b). Not all inhibitors are expected to have the same significance (H3a). RQ3a. Which conceptual model allows us to holistically collect and map inhibitors? RQ3b. Which inhibitors are perceived by industrial manufacturing enterprises? H3a. Some inhibitors are perceived to be more challenging than the others. Results Conceptual frameworks The term IoT application as used in this study is defined in this section. In addition, a literature review on motivators and inhibitors of the development and deployment of IoT applications and two resulting conceptual models are presented. An extended value- chain model for motivators and four domains of inhibitors emerged from the literature review. (2018) 7:10 Heinis et al. Journal of Innovation and Entrepreneurship Page 6 of 22 Definition of IoT application The term IoT application used in this study is defined based on three main elements: (1) physical object, (2) data processing functionality and (3) added value (Heinis et al. 2017). The second element contains three functional sub-elements: (2.1) data sensing, (2.2) data transmission and (2.3) data evaluation. This framework allows us to capture and describe IoT applications integrally, covering the technological, business, engineering design and innovation aspects. Motivators along value-chain A list of eight motivators (m1–m8) related to the development and deployment of IoT applications, as well as the potential added value of IoT applications, emerged from the literature presented in the “Background”section (Table 2). The extended value-chain model is introduced as a conceptual model that allows us to capture exhaustively the motivators of IoT application development (Fig. 1). In the- ory, enterprises conduct activities or investments only if they expect added economic value. This theory is true for engagement in IoT application development and deploy- ment. Thus, by implication, expected added value underlies each motivator to develop or deploy IoT applications. The value-chain framework describes the activities conducted by an enterprise to generate value (Porter 1985). Added value is created within an enter- prise by increasing the efficiency or effectiveness of value-chain activities or by defining new value-generating activities. Therefore, the value-chain model allows us to allocate the relevant enterprise-internal motivators of IoT application development and deploy- ment. The traditional value-chain model proposed by Porter (1985) does not explicitly cover the creation of added value for enterprise customers. External motivators based on added value for the customer cannot be allocated. In the MEM industries, customers are typically other enterprises that use a product or service to create value for their down- stream customers. The relevant activities for allocating potential external motivators are thus related to product or service usage. By extending the value-chain model with product or service usage activities according to the ideas of McPhee and Wheeler (2006), potential internal or external IoT applications can be identified. Inhibitors in four domains A conceptual model consisting of four domains of inhibitors allowed us to cluster into four domains the wide range of inhibitors of IoT application development and deployment Table 2 Possible motivators of development and deployment of IoT applications presented to survey participants for selection Motivator ID Motivators m1 Offer shared products and services as an alternative to individual ownership m2 Use collected data to improve decision-making m3 Gain revenues through new or different business models m4 Improve manufacturing and production process m5 Enhance market research for better customer segmentation or pricing strategies m6 Monitor product state and usage for predictive maintenance and repair m7 Assess product usage and performance to improve product design and development m8 Track product location to improve logistics mOT Other (2018) 7:10 Heinis et al. Journal of Innovation and Entrepreneurship Page 7 of 22 Fig. 1 Extended value-chain model to facilitate allocation of enterprise motivators for development and deployment of IoT applications identified in literature: organisational, business, technological and industrial (Table 3). The model helps cover a wide range aspects of technology-based innovations. Organisational domain Seven unique inhibitors (o1–o7) belonging to the organisa- tional domain were identified in the literature (Table 3). Geissbauer et al. (2016)reported a lack of clear vision and strategy for digital operations and a lack of leadership from top management as important inhibitors. An unsuitable organisational structure or missing key functional areas were identified as inhibitors by Porter and Heppelmann (2015)and by Twentyman and Swabey (2015). A non-existent digital culture and lack of training were mentioned as inhibitors in multiple studies (Geissbauer et al. 2016;SAS InstituteInc 2016; Twentyman and Swabey 2015). Another inhibitor mentioned in multiple sources is the lack of in-house expertise or skills (Geissbauer et al. 2016; Porter and Heppelmann 2015; Curran et al. 2015; Twentyman and Swabey 2015). Pech (2016) stated that the adoption of disruptive technologies decelerates when special training is involved. Data analytics capabilities is a more specific but indispensable skill (Geissbauer et al. 2016). Integration between physical product development and software development processes is essential for the development of IoT applications. This can be difficult when product design and engineering occur over a lengthy, linear development cycle, as opposed to digital and software design, which proceed in short, modular development loops (Hui 2014). Business domain Seven unique inhibitors (b1–b7) belonging to the business domain were identified in the literature (Table 3). Expectations related to the impact of IoT appli- cations on demand and revenues are contradictory (Twentyman and Swabey 2015). The same applies to the expected impact on cost: decrease (Geissbauer et al. 2016)versus increase (Twentyman and Swabey 2015). There is considerable uncertainty around rev- enues and costs. This could inhibit enterprises from deploying IoT applications because of the higher perceived risk and volatility of IoT investments. Enterprises face challenges (2018) 7:10 Heinis et al. Journal of Innovation and Entrepreneurship Page 8 of 22 Table 3 Conceptual model covering inhibitors of development and deployment of IoT applications Inhibitor ID Organisational inhibitor o1 Lack of clear digital operations vision/strategy o2 Unsuitable organisational structure or missing key functional areas o3 Lack of leadership from top management o4 Lack of digital culture and training o5 Lack of in-house expertise or skills o6 Lack of capabilities in data analytics o7 Lack of integration between physical product development and software development oNA Not applicable Inhibitor ID Business inhibitor b1 Insufficient information to predict demand and revenues, resulting in high uncertainty b2 Weak value proposition of IoT applications and resulting low customer demand b3 Difficulty in identifying market opportunities b4 Insufficient information to predict costs or required investment b5 Issues related to monetisation under current business model b6 Issues related to collaboration with suppliers or partners on digital solutions b7 Issues related to choosing level of vertical integration for IoT applications bNA Not applicable Inhibitor ID Technological inhibitor t1 Availability of basic infrastructure technologies t2 Difficulties related to selecting enabling technologies to realise IoT applications t3 Difficulties related to interoperability with internal or external systems t4 Need for standardised identification and addressing protocols t5 Internet scalability to handle increase in traffic and requests t6 Issues related to physical product design measures to prevent unauthorised data access t7 Issues related to software measures to prevent unauthorised data access t8 Insufficient tools to manage user authentication process t9 Difficulties related to integration of digital components into physical product t10 Access to tools and database to handle big data tNA Not applicable Inhibitor ID Industrial inhibitor i1 Undefined regulations and laws around customer privacy and data collection i2 Undefined regulations and laws around the use and sharing of data i3 Lack of comprehensive and widely accepted service intermediaries i4 Lack of certification to improve trust among customers and industry participants i5 Potential loss of intellectual property iNA Not applicable in terms of identifying market opportunities, establishing appropriate channels, defin- ing the value proposition of IoT applications and handling the new demands associated with a closer customer relationship (Porter and Heppelmann 2015;Fleisch et al. 2015; SAS Institute Inc 2016; Twentyman and Swabey 2015). Fleisch et al. (2015)reportedthat IoT changes existing business models. The business model of an industrial manufactur- ing enterprise is usually based on product sales, and service-oriented business models are rare (Adrodegari et al. 2014). Enterprises thus face issues with monetisation, for example, of data under current business models (Tobler et al. 2013). Enterprises face issues in terms of collaborating with suppliers or partners on digital solutions. Missing digital expertise (2018) 7:10 Heinis et al. Journal of Innovation and Entrepreneurship Page 9 of 22 forces industrial manufacturing enterprises to collaborate with new suppliers and part- ners. The bargaining power of these suppliers can be high, thus allowing them to capture a large share of the IoT applications’ value (Porter and Heppelmann 2014). Developing IoT applications can expand the range of activities of an enterprise, which is necessary to create value. Given the broad range and complexity of activities and limited resources, SMEs, especially, may face challenges in terms of choosing the level of vertical integration to develop IoT applications. Technological domain Ten inhibitors (t1–t10) can be assigned to the technological domain (Table 3). As mechanical parts are replaced with software, the physical complexity of a product usually diminishes (Porter and Heppelmann 2015). IoT applications may have fewer physical components, but the number of sensors required and the pervasiveness of software use rise. These requirements create new challenges for industrial manufacturing enterprises because digital components must be integrated in physical products. Several studies have explored open issues related to middleware or architecture for IoT (Atzori et al. 2010; Bandyopadhyay and Sen 2011;Khanetal. 2012) that impede IoT deployment. Given the complexity of IoT middleware, industrial manufacturing enterprises face the challenge of identifying the appropriate architecture and selecting enabling technologies. The sheer volume of data available from IoT applications and the business impacts of its use go hand-in-hand with the challenges pertaining to handling and analysing data (Gubbi et al. 2013; Lee and Lee 2015). In addition, the integration of IoT applications with existing internal or external software systems necessary to create a lasting business impact is challenging. The complexity and availability of infrastructure technologies can pose a significant challenge for industrial manufacturing enterprises. Essential connectiv- ity requirements include internet scalability and the need for standardisation to connect and integrate technologies (Bandyopadhyay and Sen 2011,Atzorietal.2010,Porterand Heppelmann 2015). Additional technological inhibitors are based on the implementation of hardware or software measures to prevent any unauthorised data access. Atzori et al. (2010) outlined the various reasons why IoT applications are especially vulnerable to attacks. First, the physical components of an IoT application are mostly exposed and unat- tended, and it is difficult to protect them with physical measures. Second, communication is often wireless, which arguably makes it easy for unauthorised persons to intercept them. Last, IoT components cannot implement complex schemes to support security because they typically have limited energy and computing resources. Industrial domain Five inhibitors (i1–i5) belonging to the industrial domain were iden- tified in literature (Table 3). These inhibitors affect the industry overall and are largely external to the enterprise. The lack of clear regulations on the collection, sharing and use of data can pose significant legal challenges for enterprises developing IoT applications. Issues pertaining to data access and collection are tied to the basic right to privacy, which includes concealing personal information and the ability to control what happens with this information (Weber 2010). IoT applications will remain limited to a few niche mar- kets if there is continued public concern about privacy (Sundmaeker et al. 2010). The lack of comprehensive, trustworthy and widely accepted service intermediaries prevents the (2018) 7:10 Heinis et al. Journal of Innovation and Entrepreneurship Page 10 of 22 deployment of IoT applications (Haller et al. 2009). Thelackofamethod forcertifica- tion of IoT applications can pose a challenge to enterprises trying to establish credibility and improve customer trust in their IoT applications. A few emerging certification pro- grams for IoT-specific applications are available (Underwriters Laboratories 2016;ICSA Labs 2017; IoT Security Foundation 2017). However, these certification programs are in the early stages of development, and they may not have a significant impact on the mar- ket acceptance of IoT applications. Data ownership and intellectual property issues within the IoT domain are important and widely discussed topics which could inhibit enterprises from developing IoT applications (Porter and Heppelmann 2014; Geissbauer et al. 2016). Empirical evidence This section presents the results of our survey and statistical evaluation of the gathered data in three subsections, one each on (1) interest and engagement, (2) motivators and (3) inhibitors. Interest and engagement Survey respondents from SMEs are not more likely to rate the importance of digitalised products differently than LEs, even though the values for each category differ (Table 4). The null hypothesis corresponding to H1a is not rejected. There is no significant differ- ence between SMEs and LEs (Mann-Whitney U = 1176.5, p = 0.075, α = 0.05, two tailed). Survey respondents from SMEs are more likely to not engage in IoT application development. The results do not contradict hypothesis H1b. Two chi-square tests of inde- pendence were performed to examine the relationship between enterprise type and the level of interest and engagement. The level of interest and engagement in IoT applica- tion development is one categorical dependent variable, and it is measured based on the selection of five possible answers (no plans and no interest, no plans but interest, plans, in progress, experienced)(Table 5). The relationship between enterprise type and level of 2 ∗ interest and engagement is significant, χ (4, N = 109) = 11.45 , α = .05, p = 0.0219. A second dependent variable can be obtained by dividing the participants into two groups (non-engaged, engaged) based on their levels of interest and engagement (Table 6). Partic- ipants who selected in progress or experienced were assigned to the engaged group, and the rest were assigned to the non-engaged group. The relationship between enterprise type and level of interest and engagement based on the two groups is very significant, χ ∗∗ (1, N = 109) = 7.85 , α = .01, p = 0.00508. Survey respondents who assigned higher importance to digitalised products are more likely to engage in the development and deployment of IoT applications. These results do not contradict hypothesis H1c. Again, two chi-square tests of independence were per- formed to examine the relationship between the importance of digitalised products and the level of interest and engagement. The two dependent variables describing the level of Table 4 Importance of digitalised products separated by type of enterprise Enterprise type N Not important (%) Somewhat important (%) Important (%) Very important (%) SME 60 8 25 32 35 LE 49 0 12 45 43 All 109 5 19 38 39 (2018) 7:10 Heinis et al. Journal of Innovation and Entrepreneurship Page 11 of 22 Table 5 Interest and engagement in developing IoT applications separated by enterprise type Enterprise type N No plans and No plans but Plans (%) In progress (%) Experienced (%) no interest (%) interest (%) SME 60 28 12 22 15 23 LE 49 6 10 18 29 37 All 109 18 11 20 21 29 interest and engagement used to test H1b were used to test H1c as well. The relationship between the importance of digitalised products and the level of interest and engagement 2 ∗ is significant, χ (12, N = 109) = 24.73 , α = .05, p = 0.0162 (Table 7). The relationship between the importance of digitalised products and the level of interest and engage- 2 ∗∗ ment based on the two groups is very significant, χ (3, N = 109) = 11.96 , α = .01, p = 0.00751 (Table 8). Motivators In total, 91 statements extracted from the open-ended question on motivators and expected added value of IoT application development and deployment were coded and assigned to a value-chain activity (Table 9). Of these statements, 56% (51) were assigned to the domain of primary activities. The highest number of statements was assigned to the marketing and sales activity segment (26%, 24). Examples include the expectation of new business models, competitive advantage, strengthening of market position and bet- ter access to customer data. Other important activity segments are service and aftersales (16%, 15) and operations and production (11%, 10). Examples include predictive mainte- nance and improved manufacturing. In total, 31% (28) of the statements exhibited strong customer or product user focus and were allocated to user activities. Not all statements could be assigned to one specific activity (e.g. use/consume, standby/store). Statements related to the entire domain of user activities were thus assigned to all three activities, but with a weight of one third only (e.g. increased customer benefit). Only 13% (12) of the statements were assigned to support activities. Noteworthy is the motivation to use IoT applications for research and technology development (8%, 7). Participants revealed that they expect to use the data generated by IoT applications to improve products or services. Based on participants’ selections, the top three motivators for developing IoT appli- cations are (1) monitoring product state and usage for predictive maintenance and repair (m6, 69%), (2) using collected data to improve decision-making (m2, 61%) and (3) improving manufacturing and production processes (m4, 60%) (Table 10). On average, participants selected M=3.74 motivators (SD=1.92). The number of motivators selected by participants does not depend either on enterprise type (SME, LE) or on engagement of an enterprise (not engaged, engaged). There are significant relationships between enterprise type (SME, LE) and selection of the motivators m2, m3 and m7. The relationship between enterprise type and selection of Table 6 Engagement in IoT application development separated by enterprise type Enterprise type N Non-engaged (%) Engaged (%) SME 60 62 38 LE 49 35 65 All 109 50 50 (2018) 7:10 Heinis et al. Journal of Innovation and Entrepreneurship Page 12 of 22 Table 7 Relationship between importance of digitalised products and level of interest and engagement Importance N No plans and No plans but Plans (%) In progress (%) Experienced (%) of digitalised no interest (%) interest (%) products Not important 5 60 20 0 0 20 Somewhat 21 38 10 24 24 5 important Important 41 12 15 27 22 24 Very important 42 10 7 14 21 48 All 109 18 11 20 21 29 m2 use collected data to improve decision-making is very significant, χ (1, N = 109) = ∗∗ 9.7217 , p < .01. LEs are more likely than SMEs to select m2. The relationship between enterprise type and selection of m3 gain revenues through new or different business models 2 ∗ is significant, χ (1, N = 109) = 4.4793 , p < .05. LEs are more likely to select m3. The relationship between enterprise type and selection of m7 assess product usage and performance to improve product design and development is very significant, χ (1, N = ∗∗ 109) = 7.1376 , p < .01. Again, LEs are more likely to select m7. For all other motivators, no significant difference was found between SMEs and LEs (p > 0.05) (Table 11). There are significant relationships between the level of engagement of an enterprise (non-engaged, engaged) and selection of m3 and m7. The relationship between the level of engagement of an enterprise and selection of m3 gain revenues through new or differ- 2 ∗ ent business models is significant, χ (1, N = 109) = 5.736 , p < .05. Engaged enterprises are more likely to select m3 than non-engaged enterprises. The relationship between enterprise type and selection of m7 assess product usage and performance to improve 2 ∗∗ product design and development is very significant, χ (1, N = 109) = 7.940 , p < .01. Again, engaged enterprises are more likely to select m7 than non-engaged enterprises. For all other motivators, no significant differences were found between engaged and non-engaged enterprises (Table 12). Inhibitors An open-ended question before the closed-ended questions on inhibitors was included in the survey to determine the key inhibitors before presenting our response options. In total, 120 statements were collected and coded into the four inhibitor domains (Table 13). A total of 37.6% of respondents indicated an inhibitor that fit into the business domain. Common responses included the weak value proposition of IoT applications and issues with the existing business model. After the business domain, 26.4% of the collected state- ments were related to technological inhibitors. Statements covering organisational or industrial inhibitors were the least represented at 21.6% and 14.4% of the responses, Table 8 Relationship between importance of digitalised products and level of engagement Importance digitalised products N Non-engaged (%) Engaged (%) Not important 5 80 20 Somewhat important 21 71 29 Important 41 54 46 Very important 42 31 69 All 109 50 50 (2018) 7:10 Heinis et al. Journal of Innovation and Entrepreneurship Page 13 of 22 Table 9 Assignment of statements from open-ended question on motivators and expected added value of IoT application development to value-chain activities Activity domain # statements % statements Value-adding activities # statements % statements Primary activities 51 56% Inbound logistics 1 1% Operations and production 10 11% Service and aftersales 15 16% Outbound logistics 1 1% Marketing and sales 24 26% Support activities 12 13% Company infrastructure 1 1% R&D development 7 8% Procurement (purchase) 2 2% Human resource management 2 2% User activities 28 31% Use/consume 10 11% Standby/store 7 8% Maintain/repair 10 11% Total 91 100% 91 100% respectively. Inhibitors identified in response to this question have largely been covered in the response options. Other commonly cited inhibitors include a lack of knowledge, data security and regulatory issues. With a weighted score of 267, the business domain is the most important domain of challenges. The organisational domain is the second most challenging with a score of 235, followed by the technological domain with a score of 198. The industrial domain ranked the last with a score of 190. The differences in the ranking of the domains are highly 2 ∗∗∗ significant (Friedman test: χ (3, N = 89) = 25.57 , p < 0.001) (Table 14). Even though the ranking and the relative ranking scores of SMEs and LEs deviate from the ranking of all enterprises, a chi-square test did not show any significant differences in the selection of rankings depending on the enterprise type (SME or LE). In the domain of organisational inhibitors, the top three inhibitors are based on the weighted score: (1) lack of clear digital operations vision/strategy with a score of 377, (2) lack of in-house expertise or skills with a score of 309 and (3) lack of integration between physical product development and software development processes with a score of 231 (Table 15). The differences in the ranking of the inhibitors are highly significant (Friedman 2 ∗∗∗ test: χ (7, N = 89) = 52.16 , p < 0.001). Table 10 Ranking of motivators for IoT application development and deployment based on selection frequency Rank Motivator ID # selections % of N = 109 1st m6 75 69% 2nd m2 67 61% 3rd m4 65 60% 4th m3 59 54% 5th m7 47 43% 6th m5 29 27% 7th m1 28 26% 8th m8 27 25% 9th mOT 11 10% (2018) 7:10 Heinis et al. Journal of Innovation and Entrepreneurship Page 14 of 22 Table 11 Comparison of ranking of motivators for IoT application development and deployment for SMEs and LEs based on selection frequency SMEs N = 60 55% LEs N = 49 45% Compared Rank Motivator # %of N Rank Motivator # %of Delta % Chi-square Significance ID selections ID selections N (1, N = 109) 6th m1 17 28% 8th m1 11 22% − 6% 0.489 3rd m2 29 48% 1st m2 38 78% 29% 9.722 ** 4th m3 27 45% 3rd m3 32 65% 20% 4.479 * 1st m4 40 67% 5th m4 25 51% − 16% 2.743 7th m5 16 27% 6th m5 13 27% 0% 0.000 2nd m6 39 65% 2nd m6 36 73% 8% 0.901 5th m7 19 32% 4th m7 28 57% 25% 7.138 ** 8th m8 15 25% 7th m8 12 24% − 1% 0.004 9th mOT 8 13% 9th mOT 3 6% − 7% 1.546 In the domain of business inhibitors, the top three inhibitors are based on the weighted score: (1) insufficient information to predict demand and revenues, resulting in high uncertainty with a score of 265; (2) issues related to monetisation under current business model with a score of 261; and (3) difficulty in identifying market opportunities with a score of 224 (Table 15). The differences in the ranking of the inhibitors are very significant 2 ∗∗ (Friedman test: χ (7, N = 89) = 24.28 , p < 0.01). In the domain of technological inhibitors, the top three inhibitors are based on the weighted score: (1) difficulties related to interoperability with internal or external sys- tems with a score of 385, (2) difficulties in selecting enabling technologies to realise IoT applications with a score of 351 and (3) availability of basic infrastructure technologies with a score of 320 (Table 15). The differences in the ranking of the inhibitors are highly 2 ∗∗∗ significant (Friedman test: χ (10, N = 89) = 41.55 , p < 0.001). In the domain of industrial inhibitors, the top three inhibitors are based on the weighted score: (1) undefined regulations and laws around customer privacy and the collection of data with a score of 230, (2) undefined regulations and laws around the use and sharing of data with a score of 175 and (3) potential loss of intellectual property with a score of 159 (Table 15). The differences in the ranking of the inhibitors are very significant (Friedman 2 ∗∗ test: χ (5, N = 89) = 19.95 , p < 0.01). Table 12 Comparison of ranking of motivators for IoT application development and deployment for non-engaged and engaged enterprises based on selection frequency Non- N = 54 50% Engaged N = 55 50% Compared engaged Rank Motivator # %of N Rank Motivator # %of N Delta Chi-square Significance ID selections ID selections % (1, N = 109) 7th m1 12 22% 7th m1 16 29% 7% 0.673 3rd m2 31 57% 2nd/3rd m2 36 65% 8% 0.745 4th m3 23 43% 2nd/3rd m3 36 65% 23% 5.736 * 1st m4 36 67% 5th m4 29 53% − 14% 2.199 5th m5 17 31% 8th m5 12 22% − 10% 1.303 2nd m6 35 65% 1st m6 40 73% 8% 0.795 6th m7 16 30% 4th m7 31 56% 27% 7.940 ** 8th m8 10 19% 6th m8 17 31% 12% 2.245 9th mOT 4 7% 9th mOT 7 13% 5% 0.850 (2018) 7:10 Heinis et al. Journal of Innovation and Entrepreneurship Page 15 of 22 Table 13 Assignment of statements from open-ended question on inhibitors to four inhibitor domains Inhibitor domain # statements % statements Business 47 38% Organisational 27 22% Technological 33 26% Industrial 18 14% All 125 100% Discussion Large enterprises (LEs) show higher levels of interest and engagement in IoT application development than SMEs. The limited financial and human resources of SMEs, which hinder R&D activities, might explain this result (Hausman 2005; Massa and Testa 2008; Laperche and Liu 2013). SMEs focus strongly on customers (Scozzi et al. 2005). In combination with the difficulty of predicting demand for and revenues of potential customer-oriented IoT applications, the strong focus of SMEs on customers does not have a positive impact on their levels of interest and engagement. However, SMEs are more flexible and can adapt quickly to changes in technologies or markets (Scozzi et al. 2005). Owing to this innovation advantage, SMEs could be expected to be more experienced in developing or deploying IoT applications. The relatively high levels of interest and engagement in IoT application development and deployment reported in the survey may be ascribed to a selection bias, in that enter- prises interested in deploying IoT applications are more likely to participate in the survey. The results in “Interest and engagement” section show that LEs have a higher interest and are more engaged in IoT application development than SMEs. The sample contains larger share of participants working in LEs than can be expected from the target population. A total of 33% of the employees in the Swiss MEM industries work in LEs (Swissmem 2016). By contrast, 45% of the survey participants work in LEs. The larger share of participants from LEs could be an indicator of selection bias. Apart from that, selection bias is hardly measurable. The open-ended and the close-ended questions on motivators delivered consistent results. The top motivator from the responses to the close-ended question is monitoring product state and usage for predictive maintenance and repair (m6). The highest num- ber of statements from the open-ended question were indeed assigned to the marketing and sales activity segment. However, most statements assigned to this activity segment are vague and not very specific. A wide range of statements could thus be assigned to this segment. The activity segment that follows is service and aftersales. The statements belonging to this segment are more specific and often mention predictive maintenance. A few of the answers to the open-ended question cannot be assigned to an activity segment. Table 14 Ranking of inhibitor domains (N = 89) Rank by score Inhibitor domain Ranking score Ranking score % # 1st # 2nd # 3rd # 4th # total ***(p < 0.001) 1st Business 267 100% 34 29 18 8 89 2nd Organisational 235 88% 28 22 18 21 89 3rd Technological 198 74% 13 21 28 27 89 4th Industrial 190 71% 14 17 25 33 89 (2018) 7:10 Heinis et al. Journal of Innovation and Entrepreneurship Page 16 of 22 Table 15 Ranking of inhibitors based on ranking scores of all four inhibitor domains (N = 89) Rank by score Inhibitor ID Ranking score # selections avg score per selection avg score per participants ***(p < 0.001) 1st o1 377 51 7.39 4.24 2nd o5 309 46 6.72 3.47 3rd o7 231 37 6.24 2.60 4th o4 191 30 6.37 2.15 5th o6 184 31 5.94 2.07 6th o2 149 22 6.77 1.67 7th o3 140 23 6.09 1.57 8th oNA 56 7 8.00 0.63 Rank by score Inhibitor ID Ranking score # selections avg score per selection avg score per participants ***(p < 0.001) 1st b1 265 38 6.97 2.98 2nd b5 261 39 6.69 2.93 3rd b3 224 32 7.00 2.52 4th b4 219 32 6.84 2.46 5th b2 213 31 6.87 2.39 6th b7 146 24 6.08 1.64 7th b6 138 22 6.27 1.55 8th bNA 64 8 8.00 0.72 Rank by score Inhibitor ID Ranking score # selections avg score per selection avg score per participants ***(p < 0.001) 1st t3 385 40 9.63 4.33 2nd t2 351 35 10.03 3.94 3rd t1 320 32 10.00 3.60 4th t9 261 30 8.70 2.93 5th t4 248 26 9.54 2.79 6th t7 212 24 8.83 2.38 7th t6 200 22 9.09 2.25 8th tNA 99 9 11.00 1.11 9th t8 98 12 8.17 1.10 10th t10 96 12 8.00 1.08 11th t5 83 11 7.55 0.93 Rank by score Inhibitor ID Ranking score # selections avg score per selection avg score per participants ***(p < 0.001) 1st i1 230 41 5.61 2.58 2nd i2 175 34 5.15 1.97 3rd i5 159 31 5.13 1.79 4th i4 129 25 5.16 1.45 5th iNA 90 15 6.00 1.01 6th i3 74 16 4.63 0.83 This could indicate that the value-chain model is not conceptually suitable for capturing motivators and expected added value, which is not the case. Most of these answers do not cover motivators or expected added value at all. A few are extrinsic motivators such as “this is the future”, “we cannot ignore this trend”, or “market pressure”, which probably do not lead to a lasting engagement in IoT application development. The results of the open-ended question on inhibitors are aligned with the finding from the close-ended questions that business inhibitors are the most challenging. The result related to the second-ranked inhibitor domain from the open-ended question does not (2018) 7:10 Heinis et al. Journal of Innovation and Entrepreneurship Page 17 of 22 align with the result obtained from the close-ended question. Participants answered the open-ended question before being presented the inhibitor domains and the full set of inhibitors. An insight from informal interviews with industry representatives is that the perception of IoT is technology-dominated. This could explain why inhibitors stated in response to the open-ended question are more technology-oriented. After being con- fronted with all inhibitor domains and the entire collection of inhibitors, the participants may reassess their opinion. The results obtained in study show that business and organisational inhibitors hinder the realisation of IoT applications decisively and, therefore, hinder innovations based on IoT applications. This insight is not well represented in extant academic literature. The literature identifies challenges mainly in the technological or industrial domain (Atzori et al. 2010; Bandyopadhyay and Sen 2011;Khanetal. 2012; Sundmaeker et al. 2010;Mio- randi et al. 2012) and considers the realisation of the IoT as the application of a certain technology (Lee and Lee 2015). Of course, the technological and industrial challenges outlined in existing literature must be solved to facilitate the development of IoT appli- cations. However, the landscape of existing and economic IoT technology available in the market is already well developed. Consequently, researchers should focus increasingly on the business and organisational aspects of IoT application development and deployment. There are a few limitations of the present study. The sample population was cre- ated through non-random convenience sampling. In addition, the results of the survey may not be replicable. While we recognise the downside of non-random sampling, this sampling method was selected from the viewpoint of practicality considering the study duration, resources at hand, and availability of the subjects. Although we cannot effectively comment on the parameters of the entire Swiss population or the indus- trial manufacturing enterprises of other nations, the results of the survey do provide meaningful insights about enterprises already interested or engaged in IoT. Moreover, it can be argued that the selection bias in non-random sampling is unlikely to have any effect on the sections pertaining to motivators and inhibitors. For example, enter- prises that encounter technological challenges are not more likely to participate in the survey than those who encounter business challenges. Thus, apart from the results related to interest and activity, the results of all other sections of the survey should represent the trends among the industrial manufacturing enterprises who are already interested in IoT. Conclusions The results of this study show that among LEs in the Swiss MEM industries, the level of interest and engagement in developing IoT applications is generally higher than that among SMEs. The main motivation to develop IoT applications is implementing or improving service and aftersales activities in the value-chain of the enterprises by offering predictive product maintenance, for example. Four domains that covered exhaus- tively the inhibitors that hinder the development and deployment of IoT applications were identified from the literature: business, organisational, technological and industrial. Business and organisational inhibitors proved to be more relevant than technologi- cal and industrial ones. The authors identified business inhibitors, such as insufficient information to predict demand and revenues, resulting in high uncertainty and issues with monetisation under current business model, to be the most challenging ones. The (2018) 7:10 Heinis et al. Journal of Innovation and Entrepreneurship Page 18 of 22 domain of organisational inhibitors tied in second with relevant inhibitors such as lack of clear digital operations vision/strategy and lack of in-house expertise or skills.The most relevant technological inhibitors were difficulties related to interoperability with inter- nal or external systems and difficulties in selecting enabling technologies to realise IoT applications. The industrial domain of inhibitors was found to be the least challenging with inhibitors such as undefined regulations and laws around customer privacy and data collection. The approach of addressing exhaustively the motivators and inhibitors related to the development and deployment of IoT applications and comparing their relevance led to the insight that innovation for the IoT is not only about developing technology and over- coming privacy regulations, as is often discussed in academic literature, but also about developing and deploying successful IoT applications. The challenges relevant to this end at the enterprise level are not mainly about technology or regulations but about busi- ness and enterprise organisation. Business as well as enterprise organisation are driven by human behaviour and, therefore, deserve the increased attention of non-technical research fields, as is happening already in the field of innovation management (e.g. IoT business models). The potential of IoT applications in industrial manufacturing enter- prises is not yet fully exploited. The extended value-chain model used in this study could help to identify novel IoT applications other than the well-known ones, such as predictive maintenance. The results of this study imply that the identified inhibitors can be used by governments or industry associations interested in fostering IoT-based innovations or by enterprises operating in and innovating during the era of technology-driven digital transformation to refine policy and decision-making. Especially, governments and industry associations can define their supportive role for a future digital economy—as proposed by Hanna (2018)— based on the learnings gained from this study. Two possible directions for future work can be derived from this study. The first is research on the tools and methods to overcome the inhibitors identified herein. The unpredictability of demand and revenues and the corresponding high degree of uncertainty could be addressed by using agile development methods, which facilitate rapid incorporation of user feedback. The challenge associated with that approach is managing the different paces of iteration cycles for hardware and software development. To help enterprises to overcome the lack of in-house expertise and skills, methods that allow organisations to acquire new knowledge quickly must be inves- tigated. Second, the extended value-chain model can be investigated as a tool not only for allocating motivators but also for systematically searching for novel IoT applications along the entire value-chain. Methods General approach The design of our empirical study is based on the sequential explorative research design described in Teddlie and Tashakkori (2006). This approach is favoured because academic literature covering the motivators and inhibitors of the development and deployment of IoT applications integrally is non-existent, and the possible range of results must be defined first. Two main working steps were taken to answer the RQs. First, the liter- ature was reviewed to collect a broad spectrum of possible motivators and inhibitors of IoT application development and deployment and to develop a conceptual model (2018) 7:10 Heinis et al. Journal of Innovation and Entrepreneurship Page 19 of 22 which allows us to cover the inhibitors and motivators exhaustively. Second, based on the first step, a survey was designed and used to collect quantitative data for validating and measuring the relevance of the identified motivators and inhibitors and to test the formulated hypotheses. Data measurement A structured survey questionnaire containing four sections relevant to this study was designed and deployed: demographics, interest and engagement, motivators and inhibitors. The questions were designed to cover a wide range of answers because this work is an explorative empirical study on the topic, and the aim is to provide a general overview. Most questions were close-ended, except for the first question in the section pertaining to motivators and inhibitors. There, an open-ended question was used to allow participants to mention motivators and inhibitors without being biased by the answer options of the close-ended questions. The statements from the open-ended questions were coded man- ually into categories. Participants with no intentions to develop IoT applications (Table 5, no interest and no plans) were not asked to provide any answers on inhibitors because their insights were not expected to be valuable. In the section pertaining to interest and engagement, participants were asked to select the statement best describing their situ- ation (single option selection). The close-ended question on motivators presented a set Table 16 Participants’ demographic profile Measure Items # selections % of N = 109 Title Executive 36 33% Department head 29 27% Staff 23 21% Unit head 12 11% Other 9 8% Functional area Management 31 28% Research and development 34 31% Information technology 6 6% Production 5 5% Quality engineering 2 2% Marketing and communication 4 4% Sales 10 9% Other 17 16% MEM industry domain Mechanical engineering 34 41% Electrical engineering/electronics 24 29% Precision instruments, apparatus and devices 16 20% Metals 6 7% Vehicles 2 2% Other 27 33% Product/service category Power engineering transmission 21 19% Assembly and factory automation 17 16% Machine tools and manufacturing technology 16 15% Process engineering equipment 15 14% Precision tools 11 10% Remaining/other 139 128% Multiple selections allowed (2018) 7:10 Heinis et al. Journal of Innovation and Entrepreneurship Page 20 of 22 of motivators to the participants from which they could select multiple options. The rel- evance of a motivator was determined based on its overall selection frequency. In the section on inhibitors, participants were asked to select the relevant inhibitors and rank the selected ones. For the inhibitor domains, participants were asked to state the rank of each domain. The overall rank of an inhibitor or domain was calculated based on the ranking score (Hillmer 2017). Sample The target population of the survey comprised small-, medium- and large-sized enter- prises from the MEM industries in Switzerland. The estimated size of the target popu- lation was 4000 enterprises (Swissmem 2016). The survey was accessible online between 21 February 2017 and 13 April 2017, and it was available in the three languages, namely, English, German, and French. To limit survey access to the target population, the survey was distributed through organisations related closely to the MEM industries. Links to the survey were shared through newsletters or mailing lists of the organisations, for exam- ple, INNOVATION NETWORK,SWISS ENGINEERING and INDUSTRIE2025. In addition, the online survey was distributed directly to members of the Zurich IoT Meetup Group and members of SWISSMEM. Non-probability, convenience sampling was used to generate the sample. The number of complete survey responses and the resulting sample size of the study amounted to 109 enterprises. The enterprises ranged in size up to 350,000 full-time employees (FTEs) (M = 6913, Mdn = 220, SD = 37867). Of all enterprises, 55% were SMEs with up to 250 FTEs (M = 81, Mdn = 215, SD = 81). Forty-five percent of the enterprises with over 250 employees were LEs (M = 15278, Mdn = 1500, SD = 55646). More than 20% of SMEs did not have a research and development (R&D) department, com- pared to only 2% of the LEs. On average, there were 15 FTEs in the R&D departments of the SMEs. LEs had larger R&D departments with an average of 600 FTEs. More than 70% of the survey respondents reported that their positions were at the executive or managerial level (Table 16). The majority of the respondents worked in R&D (31%) or management (28%). Abbreviations FTE: Full-time employees; H: Hypotheses; ICT: Information and communication technology; IoT: Internet of things; LEs: Large enterprises; MEM: Metal, electrical, and machine; RQ: Research question; R&D: Research and development; SMEs: Small- and medium-sized enterprises Funding This research project is funded by ETH Zurich (federally funded) and ETH Zurich Foundation. The ETH Zurich Foundation is an independent, non-profit organisation under private law with the aim of promoting teaching and research at ETH Zurich. The ETH Zurich Foundation awards funds to selected projects within the key strategic areas set by the ETH Zurich Executive Board. Availability of data and materials The survey questionnaire supporting the conclusions of this article is included in the additional files supplied with the article. The data supporting the conclusions of this article will not be shared owing to confidentiality agreements with the participating enterprises. Authors’ contributions TBH is the main author of the article and is responsible for the conceptual architecture, data evaluation, conclusions and writing. JH helped with the questionnaire design and operative execution of the survey. MM co-designed the conceptual architecture. All authors reviewed and approved the final manuscript. Competing interests The authors declare that they have no competing interests. (2018) 7:10 Heinis et al. Journal of Innovation and Entrepreneurship Page 21 of 22 Publisher’s Note Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations. Author details 1 2 ETH Zurich, Product Development Group Zurich, Leonhardstrasse 21, 8092 Zurich, Switzerland. University of Zurich, Rämistrasse 71, 8006 Zurich, Switzerland. Received: 19 April 2018 Accepted: 23 August 2018 References Adrodegari, F., Alghisi, A., Saccani, N. (2014). Towards usage-oriented business models: an assessment of European capital goods manufacturers, In Proceeding of 21st EurOMA conference, Palermo (ITA). Ashton, K. (2009). That “internet of things” thing. RFiD Journal, 22(7), 97–114. http://www.itrco.jp/libraries/RFIDjournal- ThatInternetofThingsThing.pdf. Accessed 14 June 2016. Atzori, L., Iera, A., Morabito, G. (2010). The Internet of Things: a survey. Computer Networks, 54(15), 2787–2805. https://doi. org/10.1016/j.comnet.2010.05.010. Accessed 8 June 2015. Bandyopadhyay, D., & Sen, J. (2011). Internet of Things: applications and challenges in technology and standardization. Wireless Personal Communications, 58(1), 49–69. https://doi.org/10.1007/s11277-011-0288-5. Accessed 27 Sep 2016. Brynjolfsson, E., & McAfee, A. (2014). The second machine age: work, progress, and prosperity in a time of brilliant technologies. New York: Norton. Burton, B., & Walker, M.J. (2015). Hype cycle for emerging technologies. Retrieved October 13, 2016, from https://www. gartner.com/doc/3100227. Accessed 13 Oct 2016. Curran, C., Puthiyamadam, T., Sviokla, J., Verweij, G. (2015). 2015 Global Digital IQ Survey: Lessons from digital leaders. Technical Report.PwC. https://www.pwc.ch/de/publications/2016/digital_iq_survey_2015.pdf. Accessed 7 Aug 2017. Dijkman, R., Sprenkels, B., Peeters, T., Janssen, A. (2015). Business models for the Internet of Things. International Journal of Information Management, 35(6), 672–678. http://linkinghub.elsevier.com/retrieve/pii/S0268401215000766. Fleisch, E., Weinberger, M., Wortmann, F. (2015). Business models and the Internet of Things (extended abstract), (pp. 6–10): Springer International Publishing, Cham. https://doi.org/10.1007/978-3-319-16546-2_2. Geissbauer, R., Vedso, J., Schrauf, S. (2016). Industry 4.0: building the digital enterprise. Technical Report. https://www.pwc. com/gx/en/industries/industries-4.0/landing-page/industry-4.0-building-your-digital-enterprise-april-2016.pdf. Gepp, M., Gölzer, P., Grobholz, B. (2015). Engineer-to-order companies are reserved on adoption of current engineering trends - an empirical study, In 2015 IEEE International Conference on Industrial Engineering and Engineering Management (IEEM). https://doi.org/10.1109/IEEM.2015.7385902 (pp. 1525–1530). Greif, H., Kühnis, N., Warnking, P. (2016). Digital transformation: How mature are Swiss SMEs? http://www.pwc.ch/en/ publications/2016/pwc_digital_transform_how_mature_are_swiss_smes_survey_16_en.pdf. Accessed 27 July 2017. Gubbi, J., Buyya, R., Marusic, S., Palaniswami, M. (2013). Internet of Things (IoT): a vision, architectural elements, and future directions. Future Generation Computer Systems, 29(7), 1645–1660. https://doi.org/10.1016/j.future.2013.01.010. Haller, S., Karnouskos, S., Schroth, C. (2009). The Internet of Things in an enterprise context, (pp. 14–28). Berlin. https://doi. org/10.1007/978-3-642-00985-3_2. Hanna, N. (2018). A role for the state in the digital age. Journal of Innovation and Entrepreneurship, 7(1), 5. https://doi.org/ 10.1186/s13731-018-0086-3. Hausman, A. (2005). Industrial Marketing Management. Innovativeness among small businesses: theory and propositions for future research, 34(8), 773–782. http://linkinghub.elsevier.com/retrieve/pii/S0019850105000039. Heck, J. (2017). Creating mmentum and a positive long-term impact on the Innovation Capability of Swiss SMEs. PhD thesis, ETH, Zürich. https://doi.org/10.3929/ETHZ-A-010898455. Heinis, T., Gomes Martinho, C., Meboldt, M. (2017). Fundamental challenges in developing Internet of Things applications for engineers and product designers, In DS 87-5 Proceedings of the 21st International Conference on Engineering Design (ICED 17) Vol 5: Design for X, Design to X, Vancouver, Canada, 21-25.08 (pp. 279–88). Hillmer, B. (2017). Calculation of weighted ranking score. https://help.surveygizmo.com/help/rank-score. Hsu, C.L., & Lin, J.C.C. (2016). An empirical examination of consumer adoption of Internet of Things services: network externalities and concern for information privacy perspectives. Computers in Human Behavior, 62, 516–527. http:// linkinghub.elsevier.com/retrieve/pii/S0747563216302990. Hui, G. (2014). So You Want to Build an Internet of Things Business: Harvard Business Review. https://hbr.org/2014/01/so- you-want-to-build-an-internet-of-things-business. Accessed 3 June 2016. ICSA Labs (2017). IoT Security & Privacy. Retrieved August 8, 2017, from https://www.icsalabs.com/technology-program/ iot-testing. Accessed: 8 Aug 2017. IoT Security Foundation (2017). IoT is vast and has many security related issues − how do we go about addressing them? Retrieved August 8, 2017, from https://iotsecurityfoundation.org/working-groups/. Accessed 8 Aug 2017. Kaleka, A. (2002). Resources and capabilities driving competitive advantage in export markets: guidelines for industrial exporters. Industrial Marketing Management, 31(3), 273–283. https://www.sciencedirect.com/science/article/pii/ S0019850100001486. Khan, R., Khan, S.U., Zaheer, R., Khan, S. (2012). Future internet: the internet of things architecture, possible applications and key challenges, In 2012 10th International Conference on Frontiers of Information Technology. (pp. 257–260). IEEE. https://doi.org/10.1109/FIT.2012.53. Kinkel, S., Rahn, J., Rieder, B., Lerch, C., Jäger, A. (2016). Digital-Vernetztes Denken In Der Produktion. Technical Report. https://doi.org/10.13140/RG.2.2.23720.37129. Laperche, B., & Liu, Z. (2013). SMEs and knowledge-capital formation in innovation networks: a review of literature. Journal of Innovation and Entrepreneurship, 2(1), 21. https://doi.org/10.1186/2192-5372-2-21. Lee, I., & Lee, K. (2015). The Internet of Things (IoT): applications, investments, and challenges for enterprises. Business Horizons, 58(4), 431–440. https://doi.org/10.1016/j.bushor.2015.03.008. (2018) 7:10 Heinis et al. Journal of Innovation and Entrepreneurship Page 22 of 22 Li, S., Da Xu, L., Zhao, S. (2014). The internet of things: a survey. Information Systems Frontiers, 17(2), 243–259. Massa, S., & Testa, S. (2008). Innovation and SMEs: misaligned perspectives and goals among entrepreneurs, academics, and policy makers. Technovation, 28(7), 393–407. https://doi.org/10.1016/j.technovation.2008.01.002. http:// linkinghub.elsevier.com/retrieve/pii/S0166497208000126. McPhee, W., & Wheeler, D. (2006). Making the case for the added-value chain. Strategy & Leadership, 34(4), 39–46. https:// doi.org/10.1108/10878570610676873. Miorandi, D., Sicari, S., De Pellegrini F., Chlamtac I. (2012). Internet of things: vision, applications and research challenges. Ad Hoc Networks, 10(7), 1497–1516. http://www.sciencedirect.com/science/article/pii/S1570870512000674. Pech, RM. (2016). Achieving the innovative edge in technology, engineering design, and entrepreneurship. Journal of Innovation and Entrepreneurship, 5(1), 6. https://doi.org/10.1186/s13731-016-0035-y. Perera, C., Liu, C.H., Jayawardena, S., Chen, M. (2014). A survey on Internet of Things from industrial market perspective. IEEE Access, 2, 1660–1679. https://doi.org/10.1109/ACCESS.2015.2389854. Porter, M.E. (1985). Competitive advantage: Creating and sustaining, superior performance. New York: The Free Press. Porter, M.E., & Heppelmann, J.E. (2014). How smart, connected products are transforming competition. Harvard Business Review, 92(11), 64–88. Porter, M.E., & Heppelmann, J.E. (2015). How smart, connected products are transforming companies. Harvard Business Review, 93(10), 96–114. Rajkumar, R.R., Lee, I., Sha, L., Stankovic, J. (2010). Cyber-physical systems: the next computing revolution, In Proceedings of the 47th Design Automation Conference. http://doi.acm.org/10.1145/1837274.1837461 (pp. 731–736). New York: ACM. Raymond, L., Uwizeyemungu, S., Fabi, B., St-Pierre, J. (2018). IT capabilities for product innovation in SMEs: a configurational approach. Information Technology and Management, 19(1), 75–87. https://doi.org/10.1007/s10799-017-0276-x. Russo, G., Marsigalia, B., Evangelista, F., Palmaccio, M., Maggioni, M. (2015). Exploring regulations and scope of the Internet of Things in contemporary companies: a first literature analysis. Journal of Innovation and Entrepreneurship, 4(1), 11. https://doi.org/10.1186/s13731-015-0025-5. Saarikko, T., Westergren, U.H., Blomquist, T. (2017). The Internet of Things: are you ready for what’s coming?. Business Horizons, 60(5), 667–676. http://www.sciencedirect.com/science/article/pii/S000768131730068X?via%3Dihub. SAS Institute Inc (2016). Internet of Things visualise the impact. http://www.sas.com/de_ch/insights/big-data/internet- of-things.html. Schwab, K. (2015). The Fourth Industrial Revolution: what it means and how to respond. Foreign Affairs, https://www. foreignaffairs.com/articles/2015-12-12/fourth-industrial-revolution. Scozzi, B., Garavelli, C., Crowston, K. (2005). Methods for modeling and supporting innovation processes in SMEs. European Journal of Innovation Management, 8(1), 120–137. http://www.emeraldinsight.com/doi.org/10.1108/ Skinner, T. (2016). IoT Outlook 2016. Retrieved August 9, 2017, from http://telecoms.com/intelligence/telecoms-com-iot- outlook-2016/. Sundmaeker, H., Guillemin, P., Friess, P., Woelfflé, S. (2010). Vision and challenges for realising the Internet of Things. Cluster of European Research Projects on the Internet of Things. European Commision, 3(3), 34–36. https://doi.org/10. 2759/26127. Swissmem (2016). Panorama 2016 Zahlen und Fakten. Retrieved August 10, 2016, from https://www.swissmem.ch/ fileadmin/user_upload/Swissmem/Publikationen/Panorama/2016_Panorama_DE.pdf. Teddlie, C., & Tashakkori, A. (2006). A general typology of research designs featuring mixed methods. Research in the Schools, 13(1), 12–28. Tobler, A., Grieder, P., Heck, R. (2013). Die Klaviatur für eine wettbewerbsfähige Schweizer MEM-Industrie. Zurich: McKinsey & Company Schweiz. Retrieved July 11, 2016, from https://www.ub.unibas.ch/digi/a125/sachdok/2014/ BAU_1_6218937.pdf. Twentyman, J., & Swabey, P. (2015). Developing smart products. Tech. rep., The Economist Intelligence Unit. Retrieved July 25, 2016, from https://www.eiuperspectives.economist.com/sites/default/files/EIU-Cognizant- Developingsmartproducts.pdf. Underwriters Laboratories (2016). UL launches cybersecurity assurance program. http://www.ul.com/newsroom/ pressreleases/ul-launches-cybersecurity-assurance-program/. Weber, R.H. (2010). Internet of Things - new security and privacy challenges. Computer Law & Security Review, 26(1), 23–30. https://doi.org/10.1016/j.clsr.2009.11.008. Weiss, A., Schade, C., Riedl, M., Matt, D. (2016). Present and future of digitalization in South Tyrolean SMEs. Paper presented at the 1-14. Retrieved from https://search.proquest.com/docview/1860981961?accountid=27229. http://www.deepdyve.com/assets/images/DeepDyve-Logo-lg.png Journal of Innovation and Entrepreneurship Springer Journals

Empirical study on innovation motivators and inhibitors of Internet of Things applications for industrial manufacturing enterprises

Loading next page...
 
/lp/springer-journals/empirical-study-on-innovation-motivators-and-inhibitors-of-internet-of-oWuRWXWlIh

References (32)

Publisher
Springer Journals
Copyright
Copyright © 2018 by The Author(s)
Subject
Business and Management; Entrepreneurship; Economic Geography; Political Economy/Economic Policy
eISSN
2192-5372
DOI
10.1186/s13731-018-0090-7
Publisher site
See Article on Publisher Site

Abstract

ETH Zurich, Product Development Group Zurich, Leonhardstrasse 21, Industrial manufacturing enterprises in export-oriented economies rely on product or 8092 Zurich, Switzerland service innovation to maintain their competitive advantage. Decreasing costs of Full list of author information is available at the end of the article computing power, connectivity and electronic components have facilitated a wide range of innovations based on Internet of Things (IoT) applications. However, only few successful IoT applications specific to industrial manufacturing enterprises are known. Although academics have been investigating challenges related to realising IoT, existing literature does not explain this situation integrally. Therefore, interest and engagement in and motivators and inhibitors of IoT application development and deployment are investigated based on a literature review and empirically based on a survey with N = 109 participants from enterprises in the Swiss metal, electrical and machine industries. Most enterprises are interested and are often engaged in IoT application development. Improving service and aftersales activities through IoT applications is a common motivator. Inhibitors from four domains hinder the development of IoT applications: business, organisational, technological and industrial. Business and organisational inhibitors are perceived to be more challenging than the technological and industrial ones. The business and organisational issues presented herein have essential impacts on the success of innovation in IoT applications. The results indicate future research directions for the innovation and development of IoT applications, and they can be used by organisations interested in IoT-based innovations to refine policy and decision-making. Keywords: Internet of things, Digitalisation, Industry survey, Innovation, Motivators, Inhibitors Background Context The decreasing costs of computing power, connectivity and electronic components have facilitated the realisation of the vision of Internet of Things (IoT) and have created poten- tial for all sorts of applications. Various media channels, the World Economic Forum (Schwab 2015) and academic researchers (Brynjolfsson and McAfee 2014)haveclaimed that a technology-driven revolution is underway that would change the way mankind lives and runs the economy. At the core of this revolution is the merger of physical and © The Author(s). 2018 Open Access This article is distributed under the terms of the Creative Commons Attribution 4.0 International License (http://creativecommons.org/licenses/by/4.0/), which permits unrestricted use, distribution, and reproduction in any medium, provided you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons license, and indicate if changes were made. (2018) 7:10 Heinis et al. Journal of Innovation and Entrepreneurship Page 2 of 22 digital spaces as cyber-physical systems. “Cyber-physical systems (CPS) will transform how humans interact with and control the physical world” (Rajkumar et al. 2010). The term is defined as follows: “cyber-physical systems (CPS) are physical and engineered systems whose operations are monitored, coordinated, controlled and integrated by a computing and communication core” (Rajkumar et al. 2010). The communication capa- bilities of such systems drive the technological revolution because they allow multiple systems to be connected, thereby creating the IoT. The IoT paradigm is the result of the convergence of three perspectives: (1) thing-oriented, (2) Internet-oriented and (3) semantic-oriented (Atzori et al. 2010). In other words, physical things can be sensed or can sense data automatically (1). The data are then communicated automatically to other things or humans (2). The data are interpreted and evaluated automatically to derive meaning (3). These three perspectives help realise the vision of an Internet containing information about the physical world without depending on human input (Ashton 2009). Technological progress in the information and communication technology (ICT) domain has reduced the costs of computing power, connectivity and electronic components. This decrease has helped the IoT to become increasingly real and has created potential for a wide range of promising innovations. Unsurprisingly, there is a lot of hype about IoT (Burton and Walker 2015). Clearly, several IoT applications are appearing in domains such as smart homes, wearables and smart cities. Well-known IoT applications are often desirable consumer gadgets (e.g. colour-changing light balls or fitness trackers). Known applications with a positive economic or ecological impact remain scarce (e.g. park- ing or bin fill level monitoring) (SAS Institute Inc 2016). In industrial manufacturing enterprises of specialised and export-oriented economies, even fewer IoT-based innova- tions are known, despite the opportunities offered by various technological enablers. The question then is, what inhibits the development and deployment of IoT applications in industrial manufacturing enterprises and thus prevents potential innovations—missing motivation or existing inhibitors? Need Although the potential of IoT and the challenges associated with realising it have been reviewed and discussed conceptually (Saarikko et al. 2017;Russo et al. 2015; Lee and Lee 2015;Lietal. 2014; Perera et al. 2014;Khanetal. 2012), the literature does not cover well the motivators and inhibitors of IoT application development and deployment among industrial manufacturing enterprises. The scope of the key challenges identified is generic and valid for the entire IoT ecosystem—for example Naming and Identity Management (Khan et al. 2012)—but it might not be equally relevant for individual enterprises aim- ing to develop specific IoT applications. The existing theoretical work is based mainly on conceptual models and is not backed up by empirical data. Existing empirical works on digitalisation, the fourth industrial revolution and IoT do not or only partially cover inter- est, engagement, motivators and inhibitors of the development and deployment of IoT applications among industrial manufacturing enterprises (Table 1). Studies that do cover motivators and inhibitors do not focus specifically on IoT application development among industrial manufacturing enterprises (Geissbauer et al. 2016; Weiss et al. 2016); instead, they target all sorts of industries (Twentyman and Swabey 2015;SAS InstituteInc 2016)— for example healthcare, retail and consumer goods—or they survey large enterprises (LEs) only. (2018) 7:10 Heinis et al. Journal of Innovation and Entrepreneurship Page 3 of 22 Table 1 Comparison of existing empirical studies on IoT development and deployment, IoT in general, digitalisation and fourth industrial revolution Publication Main focus Covers Covers Covers Participant’s Enterprise Industry Location (2 digit textitN Type of interest and motivators inhibitors role type ISO) (approx.) research engagement and challenges (Dijkman et al. 2015) IoT business models, No No No IoT n/a n/a Global (mainly 103 Academic building block professionals NL, US) identification (Geissbauer et al. 2016) 4th industrial No (Yes) Yes Chief digital LE Industrial Global 2000 Industrial revolution, officers, senior product expectations in executives suppliers 5 years (Geppetal. 2015) Engineering trends, (Yes) No No Engineering LE Engineering- DE 30 Academic 4th industrial professionals to-order revolution (ETO) (Greif et al. 2016) Digitalisation, degree No No (Yes) Industry SME All CH 300 Industrial of digitalisation professionals (Hsu and Lin 2016) IoT services, usage No No No Consumers n/a n/a TW 508 Academic intentions of consumers (Kinkel et al. 2016) Digitalisation, (Yes) No No Industry LE and MEM DE 150 Industrial competencies for professionals SME digitalisation (SAS Institute Inc 2016) IoT deployment No Yes Yes IoT LE All Global 75 Industrial process professionals (Skinner 2016) IoT as service, service No No No Industry LE and ICT and IoT Global 900 Industrial provider perspective professionals SME service providers (Twentyman and Swabey 2015) IoT products, smart Yes Yes Yes R&D, n/a Retail, Global (mainly 200 Industrial product development innovation, healthcare, US, GB) product manufac- development turing executives (Weiss et al. 2016) Digitalisation, degree No (Yes) (Yes) Senior SME All IT 53 Academic of digitalisation executives (2018) 7:10 Heinis et al. Journal of Innovation and Entrepreneurship Page 4 of 22 Small- and medium-sized enterprises (SMEs) with fewer than 250 employees have not surveyed in this regard, although the exports of high-wage economies are domi- nated by industrial manufacturing SMEs—for example, the Swiss metal, electrical and machine industries (MEM) are responsible for more than 30% of the total goods exports, and the majority of the enterprises in these industries are SMEs (Swissmem 2016). Export-oriented industrial manufacturing enterprises rely on innovations to maintain their competitive advantage in current globalised markets (Kaleka 2002). Raymond et al. (2018) argued that information technology (IT) capabilities can be used for innovation purposes in industrial SMEs. Thus, IoT-based innovations can help achieve competitive advantages in globalised markets. Studies in the literature do not exhaustively cover the motivators or inhibitors of IoT- based innovations but focus on a limited range of topics such as privacy issues or data analytics capabilities. By focusing on IoT products (smart products) only and excluding enterprise-internal IoT applications, Twentyman and Swabey (2015) analysed one per- spective on IoT application development in depth but missed out on providing a holistic view. A holistic and consolidated study on the motivators and inhibitors of IoT application development from the perspective of manufacturing enterprises is needed to understand the entire system of technology-based innovations, such as access to technology, business and financial issues and research and development knowledge and skills. Only a holis- tic view of the system allows us to compare the relevance of individual motivators and inhibitors effectively, as well as to define measures that can foster IoT-based innovations in industrial manufacturing enterprises. Task The present study investigates the interests and engagement of industrial manufacturing enterprises in IoT-based innovations and aims to provide a holistic understanding of the motivators and inhibitors of innovation in IoT application development and deployment in these enterprises. This knowledge is necessary to refine policy and decision-making in governments or industry associations interested in fostering IoT-based innovations or in enterprises operating and innovating in the era of technology-driven digital trans- formation. The authors address three specific research questions (RQ). They investigate (1) interest and engagement in, (2) motivators and (3) inhibitors of development and deployment of IoT applications in industrial manufacturing enterprises by presenting two conceptual models based on a literature review—one on motivators and one on inhibitors—and empirical data from a survey with 109 participants from Swiss MEM industries. 1 RQ1. Are manufacturing enterprises interested and engaged in the development and deployment of IoT applications? 2 RQ2. What benefit (added value) do enterprises expect from the development and deployment of IoT applications? 3 RQ3. Which inhibitors hinder the development and deployment of IoT applications? Interest and engagement (RQ1) The first RQ targets the interest and engagement in the development and deployment of IoT applications. In relation to the first RQ, three hypotheses (H) can be formulated. These hypotheses claim differences in interest and engagement based on the type of (2018) 7:10 Heinis et al. Journal of Innovation and Entrepreneurship Page 5 of 22 enterprise (SME or LE) and based on the importance rating of manufactured products integrated into an IoT application (digitalised products). Developing and deploying IoT applications is a form of innovation. Heck (2017) discussed the characteristics relevant to the innovation capabilities of SMEs, which differ from those of LEs. These differences allowed us to hypothesise different interests and levels of engagement in IoT application development and deployment for SMEs and LEs. H1a. Importance of digitalised products is rated differently depending on the type of enterprise. H1b. Level of engagement in IoT application development and deployment differs depending on the type of enterprise. H1c. Enterprises that assign more importance to digitalised products are more likely to engage in the development of IoT applications. Motivators (RQ2) The second RQ targets the reasons and motivators for the development and deployment of IoT applications. A conceptual model is needed to holistically collect and map the moti- vators driving the development and deployment of IoT applications. Thus, the first step to answering RQ2 would be the development of a conceptual model capturing motivators (RQ2a). The differences between SMEs and LEs allow us to hypothesise different moti- vators for different enterprise types (H2a). Furthermore, engaged enterprises may have different motivators than non-engaged enterprises (H2b). RQ2a. Which conceptual model allows us to holistically collect and map motivators? H2a. SMEs and LEs rate the importance of motivators differently. H2b. Engaged enterprises and non-engaged enterprises rate the importance of motivators differently. Inhibitors (RQ3) The third RQ targets the inhibitors and challenges associated with the development and deployment of IoT applications. Inhibitors can be identified and collected from the lit- erature on the topic. A conceptual model is needed to map the inhibitors exhaustively (RQ3a). Furthermore, inhibitors can be identified based on the statements of industry members (RQ3b). Not all inhibitors are expected to have the same significance (H3a). RQ3a. Which conceptual model allows us to holistically collect and map inhibitors? RQ3b. Which inhibitors are perceived by industrial manufacturing enterprises? H3a. Some inhibitors are perceived to be more challenging than the others. Results Conceptual frameworks The term IoT application as used in this study is defined in this section. In addition, a literature review on motivators and inhibitors of the development and deployment of IoT applications and two resulting conceptual models are presented. An extended value- chain model for motivators and four domains of inhibitors emerged from the literature review. (2018) 7:10 Heinis et al. Journal of Innovation and Entrepreneurship Page 6 of 22 Definition of IoT application The term IoT application used in this study is defined based on three main elements: (1) physical object, (2) data processing functionality and (3) added value (Heinis et al. 2017). The second element contains three functional sub-elements: (2.1) data sensing, (2.2) data transmission and (2.3) data evaluation. This framework allows us to capture and describe IoT applications integrally, covering the technological, business, engineering design and innovation aspects. Motivators along value-chain A list of eight motivators (m1–m8) related to the development and deployment of IoT applications, as well as the potential added value of IoT applications, emerged from the literature presented in the “Background”section (Table 2). The extended value-chain model is introduced as a conceptual model that allows us to capture exhaustively the motivators of IoT application development (Fig. 1). In the- ory, enterprises conduct activities or investments only if they expect added economic value. This theory is true for engagement in IoT application development and deploy- ment. Thus, by implication, expected added value underlies each motivator to develop or deploy IoT applications. The value-chain framework describes the activities conducted by an enterprise to generate value (Porter 1985). Added value is created within an enter- prise by increasing the efficiency or effectiveness of value-chain activities or by defining new value-generating activities. Therefore, the value-chain model allows us to allocate the relevant enterprise-internal motivators of IoT application development and deploy- ment. The traditional value-chain model proposed by Porter (1985) does not explicitly cover the creation of added value for enterprise customers. External motivators based on added value for the customer cannot be allocated. In the MEM industries, customers are typically other enterprises that use a product or service to create value for their down- stream customers. The relevant activities for allocating potential external motivators are thus related to product or service usage. By extending the value-chain model with product or service usage activities according to the ideas of McPhee and Wheeler (2006), potential internal or external IoT applications can be identified. Inhibitors in four domains A conceptual model consisting of four domains of inhibitors allowed us to cluster into four domains the wide range of inhibitors of IoT application development and deployment Table 2 Possible motivators of development and deployment of IoT applications presented to survey participants for selection Motivator ID Motivators m1 Offer shared products and services as an alternative to individual ownership m2 Use collected data to improve decision-making m3 Gain revenues through new or different business models m4 Improve manufacturing and production process m5 Enhance market research for better customer segmentation or pricing strategies m6 Monitor product state and usage for predictive maintenance and repair m7 Assess product usage and performance to improve product design and development m8 Track product location to improve logistics mOT Other (2018) 7:10 Heinis et al. Journal of Innovation and Entrepreneurship Page 7 of 22 Fig. 1 Extended value-chain model to facilitate allocation of enterprise motivators for development and deployment of IoT applications identified in literature: organisational, business, technological and industrial (Table 3). The model helps cover a wide range aspects of technology-based innovations. Organisational domain Seven unique inhibitors (o1–o7) belonging to the organisa- tional domain were identified in the literature (Table 3). Geissbauer et al. (2016)reported a lack of clear vision and strategy for digital operations and a lack of leadership from top management as important inhibitors. An unsuitable organisational structure or missing key functional areas were identified as inhibitors by Porter and Heppelmann (2015)and by Twentyman and Swabey (2015). A non-existent digital culture and lack of training were mentioned as inhibitors in multiple studies (Geissbauer et al. 2016;SAS InstituteInc 2016; Twentyman and Swabey 2015). Another inhibitor mentioned in multiple sources is the lack of in-house expertise or skills (Geissbauer et al. 2016; Porter and Heppelmann 2015; Curran et al. 2015; Twentyman and Swabey 2015). Pech (2016) stated that the adoption of disruptive technologies decelerates when special training is involved. Data analytics capabilities is a more specific but indispensable skill (Geissbauer et al. 2016). Integration between physical product development and software development processes is essential for the development of IoT applications. This can be difficult when product design and engineering occur over a lengthy, linear development cycle, as opposed to digital and software design, which proceed in short, modular development loops (Hui 2014). Business domain Seven unique inhibitors (b1–b7) belonging to the business domain were identified in the literature (Table 3). Expectations related to the impact of IoT appli- cations on demand and revenues are contradictory (Twentyman and Swabey 2015). The same applies to the expected impact on cost: decrease (Geissbauer et al. 2016)versus increase (Twentyman and Swabey 2015). There is considerable uncertainty around rev- enues and costs. This could inhibit enterprises from deploying IoT applications because of the higher perceived risk and volatility of IoT investments. Enterprises face challenges (2018) 7:10 Heinis et al. Journal of Innovation and Entrepreneurship Page 8 of 22 Table 3 Conceptual model covering inhibitors of development and deployment of IoT applications Inhibitor ID Organisational inhibitor o1 Lack of clear digital operations vision/strategy o2 Unsuitable organisational structure or missing key functional areas o3 Lack of leadership from top management o4 Lack of digital culture and training o5 Lack of in-house expertise or skills o6 Lack of capabilities in data analytics o7 Lack of integration between physical product development and software development oNA Not applicable Inhibitor ID Business inhibitor b1 Insufficient information to predict demand and revenues, resulting in high uncertainty b2 Weak value proposition of IoT applications and resulting low customer demand b3 Difficulty in identifying market opportunities b4 Insufficient information to predict costs or required investment b5 Issues related to monetisation under current business model b6 Issues related to collaboration with suppliers or partners on digital solutions b7 Issues related to choosing level of vertical integration for IoT applications bNA Not applicable Inhibitor ID Technological inhibitor t1 Availability of basic infrastructure technologies t2 Difficulties related to selecting enabling technologies to realise IoT applications t3 Difficulties related to interoperability with internal or external systems t4 Need for standardised identification and addressing protocols t5 Internet scalability to handle increase in traffic and requests t6 Issues related to physical product design measures to prevent unauthorised data access t7 Issues related to software measures to prevent unauthorised data access t8 Insufficient tools to manage user authentication process t9 Difficulties related to integration of digital components into physical product t10 Access to tools and database to handle big data tNA Not applicable Inhibitor ID Industrial inhibitor i1 Undefined regulations and laws around customer privacy and data collection i2 Undefined regulations and laws around the use and sharing of data i3 Lack of comprehensive and widely accepted service intermediaries i4 Lack of certification to improve trust among customers and industry participants i5 Potential loss of intellectual property iNA Not applicable in terms of identifying market opportunities, establishing appropriate channels, defin- ing the value proposition of IoT applications and handling the new demands associated with a closer customer relationship (Porter and Heppelmann 2015;Fleisch et al. 2015; SAS Institute Inc 2016; Twentyman and Swabey 2015). Fleisch et al. (2015)reportedthat IoT changes existing business models. The business model of an industrial manufactur- ing enterprise is usually based on product sales, and service-oriented business models are rare (Adrodegari et al. 2014). Enterprises thus face issues with monetisation, for example, of data under current business models (Tobler et al. 2013). Enterprises face issues in terms of collaborating with suppliers or partners on digital solutions. Missing digital expertise (2018) 7:10 Heinis et al. Journal of Innovation and Entrepreneurship Page 9 of 22 forces industrial manufacturing enterprises to collaborate with new suppliers and part- ners. The bargaining power of these suppliers can be high, thus allowing them to capture a large share of the IoT applications’ value (Porter and Heppelmann 2014). Developing IoT applications can expand the range of activities of an enterprise, which is necessary to create value. Given the broad range and complexity of activities and limited resources, SMEs, especially, may face challenges in terms of choosing the level of vertical integration to develop IoT applications. Technological domain Ten inhibitors (t1–t10) can be assigned to the technological domain (Table 3). As mechanical parts are replaced with software, the physical complexity of a product usually diminishes (Porter and Heppelmann 2015). IoT applications may have fewer physical components, but the number of sensors required and the pervasiveness of software use rise. These requirements create new challenges for industrial manufacturing enterprises because digital components must be integrated in physical products. Several studies have explored open issues related to middleware or architecture for IoT (Atzori et al. 2010; Bandyopadhyay and Sen 2011;Khanetal. 2012) that impede IoT deployment. Given the complexity of IoT middleware, industrial manufacturing enterprises face the challenge of identifying the appropriate architecture and selecting enabling technologies. The sheer volume of data available from IoT applications and the business impacts of its use go hand-in-hand with the challenges pertaining to handling and analysing data (Gubbi et al. 2013; Lee and Lee 2015). In addition, the integration of IoT applications with existing internal or external software systems necessary to create a lasting business impact is challenging. The complexity and availability of infrastructure technologies can pose a significant challenge for industrial manufacturing enterprises. Essential connectiv- ity requirements include internet scalability and the need for standardisation to connect and integrate technologies (Bandyopadhyay and Sen 2011,Atzorietal.2010,Porterand Heppelmann 2015). Additional technological inhibitors are based on the implementation of hardware or software measures to prevent any unauthorised data access. Atzori et al. (2010) outlined the various reasons why IoT applications are especially vulnerable to attacks. First, the physical components of an IoT application are mostly exposed and unat- tended, and it is difficult to protect them with physical measures. Second, communication is often wireless, which arguably makes it easy for unauthorised persons to intercept them. Last, IoT components cannot implement complex schemes to support security because they typically have limited energy and computing resources. Industrial domain Five inhibitors (i1–i5) belonging to the industrial domain were iden- tified in literature (Table 3). These inhibitors affect the industry overall and are largely external to the enterprise. The lack of clear regulations on the collection, sharing and use of data can pose significant legal challenges for enterprises developing IoT applications. Issues pertaining to data access and collection are tied to the basic right to privacy, which includes concealing personal information and the ability to control what happens with this information (Weber 2010). IoT applications will remain limited to a few niche mar- kets if there is continued public concern about privacy (Sundmaeker et al. 2010). The lack of comprehensive, trustworthy and widely accepted service intermediaries prevents the (2018) 7:10 Heinis et al. Journal of Innovation and Entrepreneurship Page 10 of 22 deployment of IoT applications (Haller et al. 2009). Thelackofamethod forcertifica- tion of IoT applications can pose a challenge to enterprises trying to establish credibility and improve customer trust in their IoT applications. A few emerging certification pro- grams for IoT-specific applications are available (Underwriters Laboratories 2016;ICSA Labs 2017; IoT Security Foundation 2017). However, these certification programs are in the early stages of development, and they may not have a significant impact on the mar- ket acceptance of IoT applications. Data ownership and intellectual property issues within the IoT domain are important and widely discussed topics which could inhibit enterprises from developing IoT applications (Porter and Heppelmann 2014; Geissbauer et al. 2016). Empirical evidence This section presents the results of our survey and statistical evaluation of the gathered data in three subsections, one each on (1) interest and engagement, (2) motivators and (3) inhibitors. Interest and engagement Survey respondents from SMEs are not more likely to rate the importance of digitalised products differently than LEs, even though the values for each category differ (Table 4). The null hypothesis corresponding to H1a is not rejected. There is no significant differ- ence between SMEs and LEs (Mann-Whitney U = 1176.5, p = 0.075, α = 0.05, two tailed). Survey respondents from SMEs are more likely to not engage in IoT application development. The results do not contradict hypothesis H1b. Two chi-square tests of inde- pendence were performed to examine the relationship between enterprise type and the level of interest and engagement. The level of interest and engagement in IoT applica- tion development is one categorical dependent variable, and it is measured based on the selection of five possible answers (no plans and no interest, no plans but interest, plans, in progress, experienced)(Table 5). The relationship between enterprise type and level of 2 ∗ interest and engagement is significant, χ (4, N = 109) = 11.45 , α = .05, p = 0.0219. A second dependent variable can be obtained by dividing the participants into two groups (non-engaged, engaged) based on their levels of interest and engagement (Table 6). Partic- ipants who selected in progress or experienced were assigned to the engaged group, and the rest were assigned to the non-engaged group. The relationship between enterprise type and level of interest and engagement based on the two groups is very significant, χ ∗∗ (1, N = 109) = 7.85 , α = .01, p = 0.00508. Survey respondents who assigned higher importance to digitalised products are more likely to engage in the development and deployment of IoT applications. These results do not contradict hypothesis H1c. Again, two chi-square tests of independence were per- formed to examine the relationship between the importance of digitalised products and the level of interest and engagement. The two dependent variables describing the level of Table 4 Importance of digitalised products separated by type of enterprise Enterprise type N Not important (%) Somewhat important (%) Important (%) Very important (%) SME 60 8 25 32 35 LE 49 0 12 45 43 All 109 5 19 38 39 (2018) 7:10 Heinis et al. Journal of Innovation and Entrepreneurship Page 11 of 22 Table 5 Interest and engagement in developing IoT applications separated by enterprise type Enterprise type N No plans and No plans but Plans (%) In progress (%) Experienced (%) no interest (%) interest (%) SME 60 28 12 22 15 23 LE 49 6 10 18 29 37 All 109 18 11 20 21 29 interest and engagement used to test H1b were used to test H1c as well. The relationship between the importance of digitalised products and the level of interest and engagement 2 ∗ is significant, χ (12, N = 109) = 24.73 , α = .05, p = 0.0162 (Table 7). The relationship between the importance of digitalised products and the level of interest and engage- 2 ∗∗ ment based on the two groups is very significant, χ (3, N = 109) = 11.96 , α = .01, p = 0.00751 (Table 8). Motivators In total, 91 statements extracted from the open-ended question on motivators and expected added value of IoT application development and deployment were coded and assigned to a value-chain activity (Table 9). Of these statements, 56% (51) were assigned to the domain of primary activities. The highest number of statements was assigned to the marketing and sales activity segment (26%, 24). Examples include the expectation of new business models, competitive advantage, strengthening of market position and bet- ter access to customer data. Other important activity segments are service and aftersales (16%, 15) and operations and production (11%, 10). Examples include predictive mainte- nance and improved manufacturing. In total, 31% (28) of the statements exhibited strong customer or product user focus and were allocated to user activities. Not all statements could be assigned to one specific activity (e.g. use/consume, standby/store). Statements related to the entire domain of user activities were thus assigned to all three activities, but with a weight of one third only (e.g. increased customer benefit). Only 13% (12) of the statements were assigned to support activities. Noteworthy is the motivation to use IoT applications for research and technology development (8%, 7). Participants revealed that they expect to use the data generated by IoT applications to improve products or services. Based on participants’ selections, the top three motivators for developing IoT appli- cations are (1) monitoring product state and usage for predictive maintenance and repair (m6, 69%), (2) using collected data to improve decision-making (m2, 61%) and (3) improving manufacturing and production processes (m4, 60%) (Table 10). On average, participants selected M=3.74 motivators (SD=1.92). The number of motivators selected by participants does not depend either on enterprise type (SME, LE) or on engagement of an enterprise (not engaged, engaged). There are significant relationships between enterprise type (SME, LE) and selection of the motivators m2, m3 and m7. The relationship between enterprise type and selection of Table 6 Engagement in IoT application development separated by enterprise type Enterprise type N Non-engaged (%) Engaged (%) SME 60 62 38 LE 49 35 65 All 109 50 50 (2018) 7:10 Heinis et al. Journal of Innovation and Entrepreneurship Page 12 of 22 Table 7 Relationship between importance of digitalised products and level of interest and engagement Importance N No plans and No plans but Plans (%) In progress (%) Experienced (%) of digitalised no interest (%) interest (%) products Not important 5 60 20 0 0 20 Somewhat 21 38 10 24 24 5 important Important 41 12 15 27 22 24 Very important 42 10 7 14 21 48 All 109 18 11 20 21 29 m2 use collected data to improve decision-making is very significant, χ (1, N = 109) = ∗∗ 9.7217 , p < .01. LEs are more likely than SMEs to select m2. The relationship between enterprise type and selection of m3 gain revenues through new or different business models 2 ∗ is significant, χ (1, N = 109) = 4.4793 , p < .05. LEs are more likely to select m3. The relationship between enterprise type and selection of m7 assess product usage and performance to improve product design and development is very significant, χ (1, N = ∗∗ 109) = 7.1376 , p < .01. Again, LEs are more likely to select m7. For all other motivators, no significant difference was found between SMEs and LEs (p > 0.05) (Table 11). There are significant relationships between the level of engagement of an enterprise (non-engaged, engaged) and selection of m3 and m7. The relationship between the level of engagement of an enterprise and selection of m3 gain revenues through new or differ- 2 ∗ ent business models is significant, χ (1, N = 109) = 5.736 , p < .05. Engaged enterprises are more likely to select m3 than non-engaged enterprises. The relationship between enterprise type and selection of m7 assess product usage and performance to improve 2 ∗∗ product design and development is very significant, χ (1, N = 109) = 7.940 , p < .01. Again, engaged enterprises are more likely to select m7 than non-engaged enterprises. For all other motivators, no significant differences were found between engaged and non-engaged enterprises (Table 12). Inhibitors An open-ended question before the closed-ended questions on inhibitors was included in the survey to determine the key inhibitors before presenting our response options. In total, 120 statements were collected and coded into the four inhibitor domains (Table 13). A total of 37.6% of respondents indicated an inhibitor that fit into the business domain. Common responses included the weak value proposition of IoT applications and issues with the existing business model. After the business domain, 26.4% of the collected state- ments were related to technological inhibitors. Statements covering organisational or industrial inhibitors were the least represented at 21.6% and 14.4% of the responses, Table 8 Relationship between importance of digitalised products and level of engagement Importance digitalised products N Non-engaged (%) Engaged (%) Not important 5 80 20 Somewhat important 21 71 29 Important 41 54 46 Very important 42 31 69 All 109 50 50 (2018) 7:10 Heinis et al. Journal of Innovation and Entrepreneurship Page 13 of 22 Table 9 Assignment of statements from open-ended question on motivators and expected added value of IoT application development to value-chain activities Activity domain # statements % statements Value-adding activities # statements % statements Primary activities 51 56% Inbound logistics 1 1% Operations and production 10 11% Service and aftersales 15 16% Outbound logistics 1 1% Marketing and sales 24 26% Support activities 12 13% Company infrastructure 1 1% R&D development 7 8% Procurement (purchase) 2 2% Human resource management 2 2% User activities 28 31% Use/consume 10 11% Standby/store 7 8% Maintain/repair 10 11% Total 91 100% 91 100% respectively. Inhibitors identified in response to this question have largely been covered in the response options. Other commonly cited inhibitors include a lack of knowledge, data security and regulatory issues. With a weighted score of 267, the business domain is the most important domain of challenges. The organisational domain is the second most challenging with a score of 235, followed by the technological domain with a score of 198. The industrial domain ranked the last with a score of 190. The differences in the ranking of the domains are highly 2 ∗∗∗ significant (Friedman test: χ (3, N = 89) = 25.57 , p < 0.001) (Table 14). Even though the ranking and the relative ranking scores of SMEs and LEs deviate from the ranking of all enterprises, a chi-square test did not show any significant differences in the selection of rankings depending on the enterprise type (SME or LE). In the domain of organisational inhibitors, the top three inhibitors are based on the weighted score: (1) lack of clear digital operations vision/strategy with a score of 377, (2) lack of in-house expertise or skills with a score of 309 and (3) lack of integration between physical product development and software development processes with a score of 231 (Table 15). The differences in the ranking of the inhibitors are highly significant (Friedman 2 ∗∗∗ test: χ (7, N = 89) = 52.16 , p < 0.001). Table 10 Ranking of motivators for IoT application development and deployment based on selection frequency Rank Motivator ID # selections % of N = 109 1st m6 75 69% 2nd m2 67 61% 3rd m4 65 60% 4th m3 59 54% 5th m7 47 43% 6th m5 29 27% 7th m1 28 26% 8th m8 27 25% 9th mOT 11 10% (2018) 7:10 Heinis et al. Journal of Innovation and Entrepreneurship Page 14 of 22 Table 11 Comparison of ranking of motivators for IoT application development and deployment for SMEs and LEs based on selection frequency SMEs N = 60 55% LEs N = 49 45% Compared Rank Motivator # %of N Rank Motivator # %of Delta % Chi-square Significance ID selections ID selections N (1, N = 109) 6th m1 17 28% 8th m1 11 22% − 6% 0.489 3rd m2 29 48% 1st m2 38 78% 29% 9.722 ** 4th m3 27 45% 3rd m3 32 65% 20% 4.479 * 1st m4 40 67% 5th m4 25 51% − 16% 2.743 7th m5 16 27% 6th m5 13 27% 0% 0.000 2nd m6 39 65% 2nd m6 36 73% 8% 0.901 5th m7 19 32% 4th m7 28 57% 25% 7.138 ** 8th m8 15 25% 7th m8 12 24% − 1% 0.004 9th mOT 8 13% 9th mOT 3 6% − 7% 1.546 In the domain of business inhibitors, the top three inhibitors are based on the weighted score: (1) insufficient information to predict demand and revenues, resulting in high uncertainty with a score of 265; (2) issues related to monetisation under current business model with a score of 261; and (3) difficulty in identifying market opportunities with a score of 224 (Table 15). The differences in the ranking of the inhibitors are very significant 2 ∗∗ (Friedman test: χ (7, N = 89) = 24.28 , p < 0.01). In the domain of technological inhibitors, the top three inhibitors are based on the weighted score: (1) difficulties related to interoperability with internal or external sys- tems with a score of 385, (2) difficulties in selecting enabling technologies to realise IoT applications with a score of 351 and (3) availability of basic infrastructure technologies with a score of 320 (Table 15). The differences in the ranking of the inhibitors are highly 2 ∗∗∗ significant (Friedman test: χ (10, N = 89) = 41.55 , p < 0.001). In the domain of industrial inhibitors, the top three inhibitors are based on the weighted score: (1) undefined regulations and laws around customer privacy and the collection of data with a score of 230, (2) undefined regulations and laws around the use and sharing of data with a score of 175 and (3) potential loss of intellectual property with a score of 159 (Table 15). The differences in the ranking of the inhibitors are very significant (Friedman 2 ∗∗ test: χ (5, N = 89) = 19.95 , p < 0.01). Table 12 Comparison of ranking of motivators for IoT application development and deployment for non-engaged and engaged enterprises based on selection frequency Non- N = 54 50% Engaged N = 55 50% Compared engaged Rank Motivator # %of N Rank Motivator # %of N Delta Chi-square Significance ID selections ID selections % (1, N = 109) 7th m1 12 22% 7th m1 16 29% 7% 0.673 3rd m2 31 57% 2nd/3rd m2 36 65% 8% 0.745 4th m3 23 43% 2nd/3rd m3 36 65% 23% 5.736 * 1st m4 36 67% 5th m4 29 53% − 14% 2.199 5th m5 17 31% 8th m5 12 22% − 10% 1.303 2nd m6 35 65% 1st m6 40 73% 8% 0.795 6th m7 16 30% 4th m7 31 56% 27% 7.940 ** 8th m8 10 19% 6th m8 17 31% 12% 2.245 9th mOT 4 7% 9th mOT 7 13% 5% 0.850 (2018) 7:10 Heinis et al. Journal of Innovation and Entrepreneurship Page 15 of 22 Table 13 Assignment of statements from open-ended question on inhibitors to four inhibitor domains Inhibitor domain # statements % statements Business 47 38% Organisational 27 22% Technological 33 26% Industrial 18 14% All 125 100% Discussion Large enterprises (LEs) show higher levels of interest and engagement in IoT application development than SMEs. The limited financial and human resources of SMEs, which hinder R&D activities, might explain this result (Hausman 2005; Massa and Testa 2008; Laperche and Liu 2013). SMEs focus strongly on customers (Scozzi et al. 2005). In combination with the difficulty of predicting demand for and revenues of potential customer-oriented IoT applications, the strong focus of SMEs on customers does not have a positive impact on their levels of interest and engagement. However, SMEs are more flexible and can adapt quickly to changes in technologies or markets (Scozzi et al. 2005). Owing to this innovation advantage, SMEs could be expected to be more experienced in developing or deploying IoT applications. The relatively high levels of interest and engagement in IoT application development and deployment reported in the survey may be ascribed to a selection bias, in that enter- prises interested in deploying IoT applications are more likely to participate in the survey. The results in “Interest and engagement” section show that LEs have a higher interest and are more engaged in IoT application development than SMEs. The sample contains larger share of participants working in LEs than can be expected from the target population. A total of 33% of the employees in the Swiss MEM industries work in LEs (Swissmem 2016). By contrast, 45% of the survey participants work in LEs. The larger share of participants from LEs could be an indicator of selection bias. Apart from that, selection bias is hardly measurable. The open-ended and the close-ended questions on motivators delivered consistent results. The top motivator from the responses to the close-ended question is monitoring product state and usage for predictive maintenance and repair (m6). The highest num- ber of statements from the open-ended question were indeed assigned to the marketing and sales activity segment. However, most statements assigned to this activity segment are vague and not very specific. A wide range of statements could thus be assigned to this segment. The activity segment that follows is service and aftersales. The statements belonging to this segment are more specific and often mention predictive maintenance. A few of the answers to the open-ended question cannot be assigned to an activity segment. Table 14 Ranking of inhibitor domains (N = 89) Rank by score Inhibitor domain Ranking score Ranking score % # 1st # 2nd # 3rd # 4th # total ***(p < 0.001) 1st Business 267 100% 34 29 18 8 89 2nd Organisational 235 88% 28 22 18 21 89 3rd Technological 198 74% 13 21 28 27 89 4th Industrial 190 71% 14 17 25 33 89 (2018) 7:10 Heinis et al. Journal of Innovation and Entrepreneurship Page 16 of 22 Table 15 Ranking of inhibitors based on ranking scores of all four inhibitor domains (N = 89) Rank by score Inhibitor ID Ranking score # selections avg score per selection avg score per participants ***(p < 0.001) 1st o1 377 51 7.39 4.24 2nd o5 309 46 6.72 3.47 3rd o7 231 37 6.24 2.60 4th o4 191 30 6.37 2.15 5th o6 184 31 5.94 2.07 6th o2 149 22 6.77 1.67 7th o3 140 23 6.09 1.57 8th oNA 56 7 8.00 0.63 Rank by score Inhibitor ID Ranking score # selections avg score per selection avg score per participants ***(p < 0.001) 1st b1 265 38 6.97 2.98 2nd b5 261 39 6.69 2.93 3rd b3 224 32 7.00 2.52 4th b4 219 32 6.84 2.46 5th b2 213 31 6.87 2.39 6th b7 146 24 6.08 1.64 7th b6 138 22 6.27 1.55 8th bNA 64 8 8.00 0.72 Rank by score Inhibitor ID Ranking score # selections avg score per selection avg score per participants ***(p < 0.001) 1st t3 385 40 9.63 4.33 2nd t2 351 35 10.03 3.94 3rd t1 320 32 10.00 3.60 4th t9 261 30 8.70 2.93 5th t4 248 26 9.54 2.79 6th t7 212 24 8.83 2.38 7th t6 200 22 9.09 2.25 8th tNA 99 9 11.00 1.11 9th t8 98 12 8.17 1.10 10th t10 96 12 8.00 1.08 11th t5 83 11 7.55 0.93 Rank by score Inhibitor ID Ranking score # selections avg score per selection avg score per participants ***(p < 0.001) 1st i1 230 41 5.61 2.58 2nd i2 175 34 5.15 1.97 3rd i5 159 31 5.13 1.79 4th i4 129 25 5.16 1.45 5th iNA 90 15 6.00 1.01 6th i3 74 16 4.63 0.83 This could indicate that the value-chain model is not conceptually suitable for capturing motivators and expected added value, which is not the case. Most of these answers do not cover motivators or expected added value at all. A few are extrinsic motivators such as “this is the future”, “we cannot ignore this trend”, or “market pressure”, which probably do not lead to a lasting engagement in IoT application development. The results of the open-ended question on inhibitors are aligned with the finding from the close-ended questions that business inhibitors are the most challenging. The result related to the second-ranked inhibitor domain from the open-ended question does not (2018) 7:10 Heinis et al. Journal of Innovation and Entrepreneurship Page 17 of 22 align with the result obtained from the close-ended question. Participants answered the open-ended question before being presented the inhibitor domains and the full set of inhibitors. An insight from informal interviews with industry representatives is that the perception of IoT is technology-dominated. This could explain why inhibitors stated in response to the open-ended question are more technology-oriented. After being con- fronted with all inhibitor domains and the entire collection of inhibitors, the participants may reassess their opinion. The results obtained in study show that business and organisational inhibitors hinder the realisation of IoT applications decisively and, therefore, hinder innovations based on IoT applications. This insight is not well represented in extant academic literature. The literature identifies challenges mainly in the technological or industrial domain (Atzori et al. 2010; Bandyopadhyay and Sen 2011;Khanetal. 2012; Sundmaeker et al. 2010;Mio- randi et al. 2012) and considers the realisation of the IoT as the application of a certain technology (Lee and Lee 2015). Of course, the technological and industrial challenges outlined in existing literature must be solved to facilitate the development of IoT appli- cations. However, the landscape of existing and economic IoT technology available in the market is already well developed. Consequently, researchers should focus increasingly on the business and organisational aspects of IoT application development and deployment. There are a few limitations of the present study. The sample population was cre- ated through non-random convenience sampling. In addition, the results of the survey may not be replicable. While we recognise the downside of non-random sampling, this sampling method was selected from the viewpoint of practicality considering the study duration, resources at hand, and availability of the subjects. Although we cannot effectively comment on the parameters of the entire Swiss population or the indus- trial manufacturing enterprises of other nations, the results of the survey do provide meaningful insights about enterprises already interested or engaged in IoT. Moreover, it can be argued that the selection bias in non-random sampling is unlikely to have any effect on the sections pertaining to motivators and inhibitors. For example, enter- prises that encounter technological challenges are not more likely to participate in the survey than those who encounter business challenges. Thus, apart from the results related to interest and activity, the results of all other sections of the survey should represent the trends among the industrial manufacturing enterprises who are already interested in IoT. Conclusions The results of this study show that among LEs in the Swiss MEM industries, the level of interest and engagement in developing IoT applications is generally higher than that among SMEs. The main motivation to develop IoT applications is implementing or improving service and aftersales activities in the value-chain of the enterprises by offering predictive product maintenance, for example. Four domains that covered exhaus- tively the inhibitors that hinder the development and deployment of IoT applications were identified from the literature: business, organisational, technological and industrial. Business and organisational inhibitors proved to be more relevant than technologi- cal and industrial ones. The authors identified business inhibitors, such as insufficient information to predict demand and revenues, resulting in high uncertainty and issues with monetisation under current business model, to be the most challenging ones. The (2018) 7:10 Heinis et al. Journal of Innovation and Entrepreneurship Page 18 of 22 domain of organisational inhibitors tied in second with relevant inhibitors such as lack of clear digital operations vision/strategy and lack of in-house expertise or skills.The most relevant technological inhibitors were difficulties related to interoperability with inter- nal or external systems and difficulties in selecting enabling technologies to realise IoT applications. The industrial domain of inhibitors was found to be the least challenging with inhibitors such as undefined regulations and laws around customer privacy and data collection. The approach of addressing exhaustively the motivators and inhibitors related to the development and deployment of IoT applications and comparing their relevance led to the insight that innovation for the IoT is not only about developing technology and over- coming privacy regulations, as is often discussed in academic literature, but also about developing and deploying successful IoT applications. The challenges relevant to this end at the enterprise level are not mainly about technology or regulations but about busi- ness and enterprise organisation. Business as well as enterprise organisation are driven by human behaviour and, therefore, deserve the increased attention of non-technical research fields, as is happening already in the field of innovation management (e.g. IoT business models). The potential of IoT applications in industrial manufacturing enter- prises is not yet fully exploited. The extended value-chain model used in this study could help to identify novel IoT applications other than the well-known ones, such as predictive maintenance. The results of this study imply that the identified inhibitors can be used by governments or industry associations interested in fostering IoT-based innovations or by enterprises operating in and innovating during the era of technology-driven digital transformation to refine policy and decision-making. Especially, governments and industry associations can define their supportive role for a future digital economy—as proposed by Hanna (2018)— based on the learnings gained from this study. Two possible directions for future work can be derived from this study. The first is research on the tools and methods to overcome the inhibitors identified herein. The unpredictability of demand and revenues and the corresponding high degree of uncertainty could be addressed by using agile development methods, which facilitate rapid incorporation of user feedback. The challenge associated with that approach is managing the different paces of iteration cycles for hardware and software development. To help enterprises to overcome the lack of in-house expertise and skills, methods that allow organisations to acquire new knowledge quickly must be inves- tigated. Second, the extended value-chain model can be investigated as a tool not only for allocating motivators but also for systematically searching for novel IoT applications along the entire value-chain. Methods General approach The design of our empirical study is based on the sequential explorative research design described in Teddlie and Tashakkori (2006). This approach is favoured because academic literature covering the motivators and inhibitors of the development and deployment of IoT applications integrally is non-existent, and the possible range of results must be defined first. Two main working steps were taken to answer the RQs. First, the liter- ature was reviewed to collect a broad spectrum of possible motivators and inhibitors of IoT application development and deployment and to develop a conceptual model (2018) 7:10 Heinis et al. Journal of Innovation and Entrepreneurship Page 19 of 22 which allows us to cover the inhibitors and motivators exhaustively. Second, based on the first step, a survey was designed and used to collect quantitative data for validating and measuring the relevance of the identified motivators and inhibitors and to test the formulated hypotheses. Data measurement A structured survey questionnaire containing four sections relevant to this study was designed and deployed: demographics, interest and engagement, motivators and inhibitors. The questions were designed to cover a wide range of answers because this work is an explorative empirical study on the topic, and the aim is to provide a general overview. Most questions were close-ended, except for the first question in the section pertaining to motivators and inhibitors. There, an open-ended question was used to allow participants to mention motivators and inhibitors without being biased by the answer options of the close-ended questions. The statements from the open-ended questions were coded man- ually into categories. Participants with no intentions to develop IoT applications (Table 5, no interest and no plans) were not asked to provide any answers on inhibitors because their insights were not expected to be valuable. In the section pertaining to interest and engagement, participants were asked to select the statement best describing their situ- ation (single option selection). The close-ended question on motivators presented a set Table 16 Participants’ demographic profile Measure Items # selections % of N = 109 Title Executive 36 33% Department head 29 27% Staff 23 21% Unit head 12 11% Other 9 8% Functional area Management 31 28% Research and development 34 31% Information technology 6 6% Production 5 5% Quality engineering 2 2% Marketing and communication 4 4% Sales 10 9% Other 17 16% MEM industry domain Mechanical engineering 34 41% Electrical engineering/electronics 24 29% Precision instruments, apparatus and devices 16 20% Metals 6 7% Vehicles 2 2% Other 27 33% Product/service category Power engineering transmission 21 19% Assembly and factory automation 17 16% Machine tools and manufacturing technology 16 15% Process engineering equipment 15 14% Precision tools 11 10% Remaining/other 139 128% Multiple selections allowed (2018) 7:10 Heinis et al. Journal of Innovation and Entrepreneurship Page 20 of 22 of motivators to the participants from which they could select multiple options. The rel- evance of a motivator was determined based on its overall selection frequency. In the section on inhibitors, participants were asked to select the relevant inhibitors and rank the selected ones. For the inhibitor domains, participants were asked to state the rank of each domain. The overall rank of an inhibitor or domain was calculated based on the ranking score (Hillmer 2017). Sample The target population of the survey comprised small-, medium- and large-sized enter- prises from the MEM industries in Switzerland. The estimated size of the target popu- lation was 4000 enterprises (Swissmem 2016). The survey was accessible online between 21 February 2017 and 13 April 2017, and it was available in the three languages, namely, English, German, and French. To limit survey access to the target population, the survey was distributed through organisations related closely to the MEM industries. Links to the survey were shared through newsletters or mailing lists of the organisations, for exam- ple, INNOVATION NETWORK,SWISS ENGINEERING and INDUSTRIE2025. In addition, the online survey was distributed directly to members of the Zurich IoT Meetup Group and members of SWISSMEM. Non-probability, convenience sampling was used to generate the sample. The number of complete survey responses and the resulting sample size of the study amounted to 109 enterprises. The enterprises ranged in size up to 350,000 full-time employees (FTEs) (M = 6913, Mdn = 220, SD = 37867). Of all enterprises, 55% were SMEs with up to 250 FTEs (M = 81, Mdn = 215, SD = 81). Forty-five percent of the enterprises with over 250 employees were LEs (M = 15278, Mdn = 1500, SD = 55646). More than 20% of SMEs did not have a research and development (R&D) department, com- pared to only 2% of the LEs. On average, there were 15 FTEs in the R&D departments of the SMEs. LEs had larger R&D departments with an average of 600 FTEs. More than 70% of the survey respondents reported that their positions were at the executive or managerial level (Table 16). The majority of the respondents worked in R&D (31%) or management (28%). Abbreviations FTE: Full-time employees; H: Hypotheses; ICT: Information and communication technology; IoT: Internet of things; LEs: Large enterprises; MEM: Metal, electrical, and machine; RQ: Research question; R&D: Research and development; SMEs: Small- and medium-sized enterprises Funding This research project is funded by ETH Zurich (federally funded) and ETH Zurich Foundation. The ETH Zurich Foundation is an independent, non-profit organisation under private law with the aim of promoting teaching and research at ETH Zurich. The ETH Zurich Foundation awards funds to selected projects within the key strategic areas set by the ETH Zurich Executive Board. Availability of data and materials The survey questionnaire supporting the conclusions of this article is included in the additional files supplied with the article. The data supporting the conclusions of this article will not be shared owing to confidentiality agreements with the participating enterprises. Authors’ contributions TBH is the main author of the article and is responsible for the conceptual architecture, data evaluation, conclusions and writing. JH helped with the questionnaire design and operative execution of the survey. MM co-designed the conceptual architecture. All authors reviewed and approved the final manuscript. Competing interests The authors declare that they have no competing interests. (2018) 7:10 Heinis et al. Journal of Innovation and Entrepreneurship Page 21 of 22 Publisher’s Note Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations. Author details 1 2 ETH Zurich, Product Development Group Zurich, Leonhardstrasse 21, 8092 Zurich, Switzerland. University of Zurich, Rämistrasse 71, 8006 Zurich, Switzerland. Received: 19 April 2018 Accepted: 23 August 2018 References Adrodegari, F., Alghisi, A., Saccani, N. (2014). Towards usage-oriented business models: an assessment of European capital goods manufacturers, In Proceeding of 21st EurOMA conference, Palermo (ITA). Ashton, K. (2009). That “internet of things” thing. RFiD Journal, 22(7), 97–114. http://www.itrco.jp/libraries/RFIDjournal- ThatInternetofThingsThing.pdf. Accessed 14 June 2016. Atzori, L., Iera, A., Morabito, G. (2010). The Internet of Things: a survey. Computer Networks, 54(15), 2787–2805. https://doi. org/10.1016/j.comnet.2010.05.010. Accessed 8 June 2015. Bandyopadhyay, D., & Sen, J. (2011). Internet of Things: applications and challenges in technology and standardization. Wireless Personal Communications, 58(1), 49–69. https://doi.org/10.1007/s11277-011-0288-5. Accessed 27 Sep 2016. Brynjolfsson, E., & McAfee, A. (2014). The second machine age: work, progress, and prosperity in a time of brilliant technologies. New York: Norton. Burton, B., & Walker, M.J. (2015). Hype cycle for emerging technologies. Retrieved October 13, 2016, from https://www. gartner.com/doc/3100227. Accessed 13 Oct 2016. Curran, C., Puthiyamadam, T., Sviokla, J., Verweij, G. (2015). 2015 Global Digital IQ Survey: Lessons from digital leaders. Technical Report.PwC. https://www.pwc.ch/de/publications/2016/digital_iq_survey_2015.pdf. Accessed 7 Aug 2017. Dijkman, R., Sprenkels, B., Peeters, T., Janssen, A. (2015). Business models for the Internet of Things. International Journal of Information Management, 35(6), 672–678. http://linkinghub.elsevier.com/retrieve/pii/S0268401215000766. Fleisch, E., Weinberger, M., Wortmann, F. (2015). Business models and the Internet of Things (extended abstract), (pp. 6–10): Springer International Publishing, Cham. https://doi.org/10.1007/978-3-319-16546-2_2. Geissbauer, R., Vedso, J., Schrauf, S. (2016). Industry 4.0: building the digital enterprise. Technical Report. https://www.pwc. com/gx/en/industries/industries-4.0/landing-page/industry-4.0-building-your-digital-enterprise-april-2016.pdf. Gepp, M., Gölzer, P., Grobholz, B. (2015). Engineer-to-order companies are reserved on adoption of current engineering trends - an empirical study, In 2015 IEEE International Conference on Industrial Engineering and Engineering Management (IEEM). https://doi.org/10.1109/IEEM.2015.7385902 (pp. 1525–1530). Greif, H., Kühnis, N., Warnking, P. (2016). Digital transformation: How mature are Swiss SMEs? http://www.pwc.ch/en/ publications/2016/pwc_digital_transform_how_mature_are_swiss_smes_survey_16_en.pdf. Accessed 27 July 2017. Gubbi, J., Buyya, R., Marusic, S., Palaniswami, M. (2013). Internet of Things (IoT): a vision, architectural elements, and future directions. Future Generation Computer Systems, 29(7), 1645–1660. https://doi.org/10.1016/j.future.2013.01.010. Haller, S., Karnouskos, S., Schroth, C. (2009). The Internet of Things in an enterprise context, (pp. 14–28). Berlin. https://doi. org/10.1007/978-3-642-00985-3_2. Hanna, N. (2018). A role for the state in the digital age. Journal of Innovation and Entrepreneurship, 7(1), 5. https://doi.org/ 10.1186/s13731-018-0086-3. Hausman, A. (2005). Industrial Marketing Management. Innovativeness among small businesses: theory and propositions for future research, 34(8), 773–782. http://linkinghub.elsevier.com/retrieve/pii/S0019850105000039. Heck, J. (2017). Creating mmentum and a positive long-term impact on the Innovation Capability of Swiss SMEs. PhD thesis, ETH, Zürich. https://doi.org/10.3929/ETHZ-A-010898455. Heinis, T., Gomes Martinho, C., Meboldt, M. (2017). Fundamental challenges in developing Internet of Things applications for engineers and product designers, In DS 87-5 Proceedings of the 21st International Conference on Engineering Design (ICED 17) Vol 5: Design for X, Design to X, Vancouver, Canada, 21-25.08 (pp. 279–88). Hillmer, B. (2017). Calculation of weighted ranking score. https://help.surveygizmo.com/help/rank-score. Hsu, C.L., & Lin, J.C.C. (2016). An empirical examination of consumer adoption of Internet of Things services: network externalities and concern for information privacy perspectives. Computers in Human Behavior, 62, 516–527. http:// linkinghub.elsevier.com/retrieve/pii/S0747563216302990. Hui, G. (2014). So You Want to Build an Internet of Things Business: Harvard Business Review. https://hbr.org/2014/01/so- you-want-to-build-an-internet-of-things-business. Accessed 3 June 2016. ICSA Labs (2017). IoT Security & Privacy. Retrieved August 8, 2017, from https://www.icsalabs.com/technology-program/ iot-testing. Accessed: 8 Aug 2017. IoT Security Foundation (2017). IoT is vast and has many security related issues − how do we go about addressing them? Retrieved August 8, 2017, from https://iotsecurityfoundation.org/working-groups/. Accessed 8 Aug 2017. Kaleka, A. (2002). Resources and capabilities driving competitive advantage in export markets: guidelines for industrial exporters. Industrial Marketing Management, 31(3), 273–283. https://www.sciencedirect.com/science/article/pii/ S0019850100001486. Khan, R., Khan, S.U., Zaheer, R., Khan, S. (2012). Future internet: the internet of things architecture, possible applications and key challenges, In 2012 10th International Conference on Frontiers of Information Technology. (pp. 257–260). IEEE. https://doi.org/10.1109/FIT.2012.53. Kinkel, S., Rahn, J., Rieder, B., Lerch, C., Jäger, A. (2016). Digital-Vernetztes Denken In Der Produktion. Technical Report. https://doi.org/10.13140/RG.2.2.23720.37129. Laperche, B., & Liu, Z. (2013). SMEs and knowledge-capital formation in innovation networks: a review of literature. Journal of Innovation and Entrepreneurship, 2(1), 21. https://doi.org/10.1186/2192-5372-2-21. Lee, I., & Lee, K. (2015). The Internet of Things (IoT): applications, investments, and challenges for enterprises. Business Horizons, 58(4), 431–440. https://doi.org/10.1016/j.bushor.2015.03.008. (2018) 7:10 Heinis et al. Journal of Innovation and Entrepreneurship Page 22 of 22 Li, S., Da Xu, L., Zhao, S. (2014). The internet of things: a survey. Information Systems Frontiers, 17(2), 243–259. Massa, S., & Testa, S. (2008). Innovation and SMEs: misaligned perspectives and goals among entrepreneurs, academics, and policy makers. Technovation, 28(7), 393–407. https://doi.org/10.1016/j.technovation.2008.01.002. http:// linkinghub.elsevier.com/retrieve/pii/S0166497208000126. McPhee, W., & Wheeler, D. (2006). Making the case for the added-value chain. Strategy & Leadership, 34(4), 39–46. https:// doi.org/10.1108/10878570610676873. Miorandi, D., Sicari, S., De Pellegrini F., Chlamtac I. (2012). Internet of things: vision, applications and research challenges. Ad Hoc Networks, 10(7), 1497–1516. http://www.sciencedirect.com/science/article/pii/S1570870512000674. Pech, RM. (2016). Achieving the innovative edge in technology, engineering design, and entrepreneurship. Journal of Innovation and Entrepreneurship, 5(1), 6. https://doi.org/10.1186/s13731-016-0035-y. Perera, C., Liu, C.H., Jayawardena, S., Chen, M. (2014). A survey on Internet of Things from industrial market perspective. IEEE Access, 2, 1660–1679. https://doi.org/10.1109/ACCESS.2015.2389854. Porter, M.E. (1985). Competitive advantage: Creating and sustaining, superior performance. New York: The Free Press. Porter, M.E., & Heppelmann, J.E. (2014). How smart, connected products are transforming competition. Harvard Business Review, 92(11), 64–88. Porter, M.E., & Heppelmann, J.E. (2015). How smart, connected products are transforming companies. Harvard Business Review, 93(10), 96–114. Rajkumar, R.R., Lee, I., Sha, L., Stankovic, J. (2010). Cyber-physical systems: the next computing revolution, In Proceedings of the 47th Design Automation Conference. http://doi.acm.org/10.1145/1837274.1837461 (pp. 731–736). New York: ACM. Raymond, L., Uwizeyemungu, S., Fabi, B., St-Pierre, J. (2018). IT capabilities for product innovation in SMEs: a configurational approach. Information Technology and Management, 19(1), 75–87. https://doi.org/10.1007/s10799-017-0276-x. Russo, G., Marsigalia, B., Evangelista, F., Palmaccio, M., Maggioni, M. (2015). Exploring regulations and scope of the Internet of Things in contemporary companies: a first literature analysis. Journal of Innovation and Entrepreneurship, 4(1), 11. https://doi.org/10.1186/s13731-015-0025-5. Saarikko, T., Westergren, U.H., Blomquist, T. (2017). The Internet of Things: are you ready for what’s coming?. Business Horizons, 60(5), 667–676. http://www.sciencedirect.com/science/article/pii/S000768131730068X?via%3Dihub. SAS Institute Inc (2016). Internet of Things visualise the impact. http://www.sas.com/de_ch/insights/big-data/internet- of-things.html. Schwab, K. (2015). The Fourth Industrial Revolution: what it means and how to respond. Foreign Affairs, https://www. foreignaffairs.com/articles/2015-12-12/fourth-industrial-revolution. Scozzi, B., Garavelli, C., Crowston, K. (2005). Methods for modeling and supporting innovation processes in SMEs. European Journal of Innovation Management, 8(1), 120–137. http://www.emeraldinsight.com/doi.org/10.1108/ Skinner, T. (2016). IoT Outlook 2016. Retrieved August 9, 2017, from http://telecoms.com/intelligence/telecoms-com-iot- outlook-2016/. Sundmaeker, H., Guillemin, P., Friess, P., Woelfflé, S. (2010). Vision and challenges for realising the Internet of Things. Cluster of European Research Projects on the Internet of Things. European Commision, 3(3), 34–36. https://doi.org/10. 2759/26127. Swissmem (2016). Panorama 2016 Zahlen und Fakten. Retrieved August 10, 2016, from https://www.swissmem.ch/ fileadmin/user_upload/Swissmem/Publikationen/Panorama/2016_Panorama_DE.pdf. Teddlie, C., & Tashakkori, A. (2006). A general typology of research designs featuring mixed methods. Research in the Schools, 13(1), 12–28. Tobler, A., Grieder, P., Heck, R. (2013). Die Klaviatur für eine wettbewerbsfähige Schweizer MEM-Industrie. Zurich: McKinsey & Company Schweiz. Retrieved July 11, 2016, from https://www.ub.unibas.ch/digi/a125/sachdok/2014/ BAU_1_6218937.pdf. Twentyman, J., & Swabey, P. (2015). Developing smart products. Tech. rep., The Economist Intelligence Unit. Retrieved July 25, 2016, from https://www.eiuperspectives.economist.com/sites/default/files/EIU-Cognizant- Developingsmartproducts.pdf. Underwriters Laboratories (2016). UL launches cybersecurity assurance program. http://www.ul.com/newsroom/ pressreleases/ul-launches-cybersecurity-assurance-program/. Weber, R.H. (2010). Internet of Things - new security and privacy challenges. Computer Law & Security Review, 26(1), 23–30. https://doi.org/10.1016/j.clsr.2009.11.008. Weiss, A., Schade, C., Riedl, M., Matt, D. (2016). Present and future of digitalization in South Tyrolean SMEs. Paper presented at the 1-14. Retrieved from https://search.proquest.com/docview/1860981961?accountid=27229.

Journal

Journal of Innovation and EntrepreneurshipSpringer Journals

Published: Sep 25, 2018

There are no references for this article.