Building Smart Agriculture Applications Using Low-Code Tools: The Case for DisCoPar

Abstract

Modern farming is experiencing increased adoption of mobile and cloud computing applications as efforts are being made to automate farming processes. In this context, the mobile and cloud computing applications, that we refer to as smart agriculture applications (SAAs), can be used in data collection that can be entered directly into the applications by end-users (i.e., farmers) or via sensors. Implementing SAAs is often done using text-based approaches that require advanced skills and experience in software programming. There are low-code development tools (LCDTs) that farmers with less programming experience can use to implement their own SAAs. The LCDTs offer different features and techniques for application development e.g., some employ form-based application specification while others use graphical drag-and-drop techniques. As such, the different LCDTs are best suited for different specific tasks. For instance, a farmer may need to implement an application that connects to sensors to receive data and generate timely notifications when set thresholds are exceeded. However, to the farmer, it can be difficult to know which kind of LCDTs to choose and what category of these tools are best suited for the task. In this paper, we contrast different LCDTs and show how to use DisCoPar to develop SAAs by non-expert programmers. As a contribution, this paper presents properties for LCDTs that can be beneficial to farmers and demonstrates how DisCoPar can be used in developing SAAs.