Farming in the Era of Internet of Things

An Information  System Architecture for Smart Farming

Abstract.

The Internet of Things and associated smart products are finding  application in evermore domains. Within agriculture it is described under the  term smart farming. Using smart products allows farmers to automatically  record relevant information, monitor operating procedures or remotely control  machines. To make these capabilities usable for added value, not only smart  products but entire information systems that align to the domain and its  requirements are necessary. Within the literature, various architectures can be  found already. However, many lack a methodical foundation or an abstraction  of application cases and technology. Against this background, the article  presents an information system architecture independent of application cases or  technology and oriented toward the domain of smart farming. Starting point of  the development is a systematic literature review, based on Webster and Watson [1] in combination with Vom Brocke [2, 3], based on 18 existing  architecture approaches that are analyzed and aggregated.

Introduction

As a result of ever-increasing digitalization due to the exponential nature of  technological development, smart products and their related services are increasingly  occupying space in both business and personal life. Even historically analog-driven  domains, such as agriculture, can no longer escape the trend of digital transformation  and its associated advantages [4, 5].  

The starting point of this development includes smart or so-called intelligent products  and their capability spectrum, which can be used for everything from simple  monitoring or remote tasks [4] to dynamic, adaptive decision making [6]. The  literature describes the domain-specific use of these products in combination with  other technologies, such as cloud computing or machine learning-based data analysis,  in the context of agriculture as smart farming, or farming 4.0 [7, 8].

To process the data collected from these products and their sensors in a manner that is  timely and appropriate, and to create from these added values for the system user, information systems specially designed for the domain and its challenges are  necessary. Within the literature, some architectural approaches are currently found in  the context of smart farming ([9–12] etc.); however, in most cases, they are designed for exactly one scenario or one specific technology and are therefore not applicable to  the broader range of agricultural needs. Against this background, the following  research question is examined: How must information systems be designed from an architectural point of view for the domain of smart farming?

In the context of this publication, this research question is elaborated on the basis of a  systematic literature search after Webster and Watson [1] in combination with Vom  Brocke [2, 3] by presenting a generic information system architecture for the domain  of smart farming. The remainder of this paper is structured as follows. First,  theoretical bases for smart products, information systems, architectures, and smart  farming are established. Subsequently, the applied research methodology and the  results of the literature review are presented. The developed information system  architecture for smart farming is subsequently presented in detail. The paper  concludes with the limitations of the work and a corresponding outlook for future  research.