A farm management information system with task-specific, collaborative mobile apps and cloud storage services
Modern production agriculture is beginning to advance beyond deterministic, scheduled operations between relatively few people to larger scale, information-driven efficiency in order to respond to the challenges of field variability and meet the needs of a growing population. Since no two farms are the same with respect to information and management structure, a specialized farm management information system (FMIS) which is tailored to the realities on the ground of individual farms is likely to be more effective than generalized FMIS available today. ^ This thesis presents the design of a FMIS using proven user-centered design principles. This approach resulted in the creation of the OpenAgToolkit (OpenATK), and a suite of task-specific, collaborative Android apps. The OpenATK system architecture enabled apps to share data between apps on a device with shared local databases, and across devices on the farm using Trello application programming interface (API). Five Android apps, Rock App, Tillage App, Trello Sync App, Field Notebook App, and Planting App, were developed in the proposed architecture. Other apps such as the Anhydrous App, and Spraying App were discussed with respect to their role in the OpenATK FMIS. The OpenATK approach proved to be technically viable with current, consumer-grade technologies including free cloud storage, Wi-Fi, and task-specific, collaborative Android apps running on tablet devices. The Tillage and Trello Sync Apps were used to generate artificial records for one year on a 404 ha (1,000 acre) farm to evaluate data storage needs. The total amount of data generated for the six field operations on the 13 fields was 260 kilobytes. ^ Four OpenATK Apps were evaluated individually by four evaluators using the personal interview procedure for interface usability. The heuristic evaluation method was an appropriate evaluation method for the goals of this project as it enabled the observer to easily identify two critical interface usability problems: the long-hold method to move rock marker icons on the map and the method to draw field boundaries. Solutions to improve the usability problems were proposed, and recommendations were given for future research.^
Dennis Buckmaster, Purdue University.
Information Technology|Engineering, Agricultural
Off-Campus Purdue Users:
To access this dissertation, please log in to our