Emerging Ecosystems Empowered by AI and IoT Technologies

Introduction

During the recent years, quite a few technological achievements that were once deemed as fiction now exist, either as laboratory prototypes, or as products with high technology readiness levels. Examples include autonomous vehicles and mobile robots of advanced capability, home and personal assistants that simplify a number of every-day processes, machines that are by far faster than humans in solving specific problems, and so on. Although such ideas have already been introduced during the past few decades (Kurzweil, 1992), their manifestation into real-world products and solutions was made possible only recently, mainly due to advancements in two broad domains of electronics and computer science, the Internet of Things (IoT), and Artificial Intelligence (AI).

Internet of Things is a result of breakthroughs in a multitude of fields, e.g. electronics and telecommunications, embedded systems, software engineering, web and cloud services, as well as finance and marketing (Ibarra-Esquer et al., 2012). Nevertheless, the main market requirement that provides a boost for IoT adoption is the fact that it enables enterprises to gather and make effective use of huge amounts of data originating from the real, physical world. This collected data is then turned into usable information and actionable knowledge regarding improvements in products and services, market analysis and various predictions, and can be employed for the optimization of a number of business and production processes within the enterprise or organization (Erevelles et al., 2016).

However, very large amounts of collected measurements and calculated indices cannot be easily processed and analyzed by the human brain. Thus, in parallel to the outspread of IoT technology adoption, the requirement for efficient manipulation and processing of the available data has also appeared. For this purpose, scientists, engineers, and decision makers turn their attention mainly to AI. AI is far from a new term and notion, as it has been conceptualized from the middle of the last century as computational methods that simulate the human brain’s operations with respect to learning and decision making (Russel & Norvig, 2019). Occasionally being in and out of researchers’ spotlights, AI is currently an “umbrella” term covering multiple sub-fields, such as natural language processing, machine learning, symbolic computation, intelligent agents, and multi-agent systems, among others.

Generally, such technologies can be considered as innovation enablers, e.g. concepts of the fourth industrial revolution can be made possible with the advent of 5G communications (Gundall et al., 2018), and adaptive/personalizable mechanisms can be improved by using machine learning techniques (Vermesan, 2017). Furthermore, these approaches are also characterized as disrupting, introducing this way the need for novelties in business model design and assessment as well (Amshoff et al., 2015), (Renda, 2019). However, this disruption is regarded by business and industries differently in each case, according to the application domain and the strategy that each party decides to adopt, and vary from partial, to full integration of such novel technologies (Laudien & Daxböck, 2016).

Now, IoT and AI, like other technologies, follow the hype cycle, and after a “Media exposure” period, when increased public attention is given, comes the “Peak of inflated Expectations” (Hahanov, 2018). Currently, most SMEs or even larger organizations do not have a complete and realistic perspective of what such technologies and related products and services are actually capable of, and how they can be integrated into their processes to their benefit one hand, and that of society in general on the other.