MoT: A Model-Driven Low-Code Approach for Simplifying Cloud-of-Things Application Development

TL;DR

Cloud-of-Things (CoT) development is complex and requires specialized expertise. The authors propose MoT, a model-driven, low-code approach with a dedicated UML profile to streamline IoT-cloud integration and automate deployment. They validate feasibility via a hospital monitoring case study and assess user acceptance through a TAM survey, reporting positive perceptions of usefulness and ease of use. The work delivers a UML-based MoT.Profile, a seven-step workflow, and empirical evidence suggesting broader accessibility and faster CoT development, with future work aimed at full cloud deployment and broader evaluation.

Abstract

The integration of cloud computing and the Internet of Things (IoT) is essential for scalable, intelligent systems. However, developing cloud-of-things (CoT) applications remains challenging. It requires significant technical expertise and lacks standardized, model-driven methodologies. Current approaches fail to ensure interoperability, automation, and efficiency. This study introduces the Model of Things (MoT), a model-based approach that incorporates low-code principles to simplify CoT development. MoT reduces technical barriers by providing a custom UML profile designed for IoT and cloud services. To evaluate MoT, we conducted a case study and a Technology Acceptance Model (TAM) questionnaire. The results confirmed MoT's feasibility, demonstrating that it streamlines CoT application development and deployment. Users found MoT accessible, even with limited IoT experience, and reported high perceived ease of use and usefulness. Qualitative feedback highlighted MoT's ability to reduce complexity and speed up development. MoT offers a promising, model-driven solution for CoT application development. By lowering entry barriers and promoting automation, it enhances both efficiency and flexibility. This study represents a step toward a more user-friendly framework, enabling broader adoption of CoT technologies.

Figures (23)

• Figure 1: A Node-Red interface that shows an illustrative Node-RED workflow using typical input and output components.
• Figure 2: An overview of the MoT process. It presents a seven-step model-based software development process, from modeling and transformation to deployment and execution, using specific artifacts and tools to ensure a structured and efficient workflow.
• Figure 3: An overview of the supported modeling process. Developers create UML use case diagrams (Step 1) and apply MoT stereotypes manually (Step 2). After the MoT approach generates an XML code of the annotated diagram (Step 3).
• Figure 4: An overview of the model transformation process supported. The MoT's model transformation tool receives a use case diagram as input (Step 1) and uses transformation rules (Step 2) to generate abstract application components as output artifacts (Step 3).
• Figure 5: Depicting an architecture where an API client communicates via API Gateway to an AWS Lambda running Node-RED, which retrieves flow configurations from S3, enabling remote workflow execution and management.