Development of a Hybrid Data-Driven Decision Framework for the Adoption of Smart Farming

Abstract

Smart farming brings together digital tools and data-enabled services—such as mobile agricultural applications, weather information, sensors, Internet of Things (IoT) devices, remote sensing, precision input technologies, and digital advisory platforms—to support farm-management decisions. However, technology adoption is not determined by technical availability alone. Farmers may weigh expected usefulness against affordability, digital literacy, connectivity, trust in providers, perceived risk, access to training, and the availability of repair and after-sales support. This paper presents a pilot-study design intended to refine a questionnaire and inform the development of a hybrid data-driven decision framework for matching smart-farming options to farmers’ needs, resources, and operating conditions. To demonstrate the planned analysis, a practice dataset of 50 simulated records was analysed using respondent-profile frequencies, construct-level descriptive statistics, and Cronbach’s alpha for eight three-item constructs. In this illustrative dataset, 36 of 50 records were coded as using at least one smart-farming tool, perceived usefulness had the highest mean score (M = 3.53, SD = 0.90), while training and extension support had the lowest (M = 2.53, SD = 0.92). Illustrative alpha coefficients ranged from 0.797 to 0.885 across the eight constructs. These values demonstrate reporting procedures only and do not estimate reliability among actual farmers. The paper details the research problem, conceptual grounding, objectives, questions, proposed hypotheses, methodology, illustrative results, and a transparent multi-criteria decision framework. The next stage is genuine pilot fieldwork, instrument refinement, and subsequent validation using an appropriately designed main study.