![Sample our Economics, Finance,Business & Industry journals, sign in here to start your access, latest two full volumes FREE to you for 14 days]({"type":"image" , "src" : "/sda/5931/TOC-Subject-Sample-2021-Economics-Finance-Business-Industry.jpg"})
ABSTRACT
New 4.0 technologies are spreading in the economy and society. Yet, empirical evidence lags behind conceptual reflections in the analysis of these technologies, their antecedents and the degree of continuity and/or originality with respect to ICTs (i.e. 3.0 technologies). To fill this gap, this paper proposes a large-scale analysis of the full spectrum of 4.0 technologies based on an original and methodologically innovative database of 4.0 inventions in European countries, over the period 2000–2015. The analysis highlights that 4.0 technologies do not overlap in full with 3.0 technologies. Rather, they are the outcome of recombination processes involving substantially diverse and distant technologies.
Acknowledgements
The authors would like to thank Roberto Barbieri and Gaetano Cascini (Politecnico di Milano) for suggestions and feedbacks on patent search methodology. Cooperation from all participants to the Delphi study is gratefully acknowledged.
Disclosure statement
No potential conflict of interest was reported by the author(s).
Notes
1 Industry 4.0 is a term coined in 2011 in Germany at the Hannover Trade Fair. It refers to the common initiative of the representatives of business, policy and science promoting the idea of strengthening the competitiveness of the German industry (Müller, Buliga, and Voigt Citation2018). Specifically, Industry 4.0 describes a process of increasing digitalisation, robotisation and automation of the manufacturing environment, enriched with the creation of digital value chains to enable inputs from suppliers and customers, and between business partners (Lasi et al. Citation2014).
2 A technological or industrial revolution can be identified when new technologies radically influences the behaviour of firms and people throughout the economic system. Changes occur in the way people work and communicate, express, inform and entertain themselves, and, finally, do business but also through new (formal and informal) institutional rules (Perez Citation2010).
3 We thank an anonymous referee to help us framing more clearly this point.
4 ICTs and 3.0 technologies are considered as synonymous in the paper and are used interchangeably.
5 The labels 3.0 and 4.0 technologies have been chosen in order to highlight the connection between each group of technologies and the ensuing socio-economic transformations deriving from their creation and adoption on a large scale, that is, the third and the fourth industrial revolution, whose actual periodization and numbering however is still matter of debate (see on this point Perez Citation2010).
6 IPC and CPC are, therefore, hierarchical classifications organized in hierarchical levels. For more details see: https://www.wipo.int/classifications/ipc/ipcpub/, last visited 2020/02/26, https://www.cooperativepatentclassification.org/index, last visited 2020/02/26.
7 For approaches combining both technological code and keyword analysis, see Montecchi, Russo, and Liu (Citation2013). For more complex methodology of patent search including TRIZ (Theory of Inventive Problem Solving), see Cascini and Russo (Citation2007).
8 https://ec.europa.eu/eurostat/cache/metadata/Annexes/htec_esms_an6.pdf, last visited 01/03/2021.
9 Precision of a patent search indicates the ratio between relevant and irrelevant results of the search. Patent searches traditionally aim at achieving high recall rates, sometimes at the cost of lower precision (Cascini and Russo Citation2007). The text analysis, then, was especially important to improve the precision rate in the identification of 4.0 inventions.
10 For a detailed discussion of the Delphi method, see Flostrand, Pitt, and Bridson (Citation2020) and Okoli and Pawlowski (Citation2004).
11 The Delphi method presents some advantages with respect to alternative ones; in particular, it allows providing exogenous information, independent from the nature and characteristics of the patent data and of the specific sample of patents. This would not have been the case, for instance, if the technological keywords had been extracted from a sub-sample of patents through some machine-learning techniques.
12 The groups of 3.0 and 4.0 patents are mutually exclusive, that is, if a patent is classified as 4.0, it is not considered also as a 3.0 one.
13 Each CPC code belongs to one and only one of the seven EPO’s categories but each patent is associated with one or more CPC codes; therefore, it might be the case that a patent is associated with more than one EPO’s category, according to its CPC codes. In order to overcome this problem, each patent was assigned the 4.0 category displaying the highest share of CPCs in the total number of CPCs of the patent itself. This approach enables to emphasize the most prominent technological content of each invention. Patents characterized by equal shares of CPC belonging to different categories show a less clear technological profile and, therefore, have not been considered in the remainder of the analysis.
14 Figures 3.1 and the following ones in EPO (Citation2017, page 27 and following ones) report similar trends.
15 For a debate on the notion and measurement of patent scope see Novelli (Citation2015).
16 The analysis covers the period 2000–2015 because patent applications in 4.0 technologies were negligible in previous years (EPO Citation2017, Citation2020; Capello, Laffi, and Lenzi Citation2020).
17 The results are significant with p < .0001.
18 With the exception of a slightly declining trend for 3.0 patents, the ICT scope of the other categories of patents is nearly constant in the considered period (2000–2015).
19 Temporal trends are confirmed as well: the difference in the ICT scope between 3.0 and 4.0 patents is decreasing over time for all the 4.0 categories, due to the diminishing share in 3.0 patents.
20 Appendix E reports a few examples of 4.0 patents that do not cite 3.0 patents.
21 These differences are statistically significant as shown in Table D2, Appendix D.
22 Differences are statistically significant. N = 9190, difference = 3.699, significant with p < .001, t statistics = 5.58.
23 CPC codes in the numerator are considered at 4 digit level, while the denominator considers CPC at the most disaggregate level.
24 The difference between the mean value of the radicalness indicator for 3.0 and 4.0 technologies is –0.0318, with a p-value < .001.