Action-Oriented Programming and Automatic Agent Generation for Adaptive Data Collection in Decentralized Data Ecosystems

Root Cause Analysis (RCA) and Risk Assessment (RA) to enable informed decision-making. However, the growing complexity, volume, and decentralized nature of manufacturing data creates significant challenges for effective data collection. Conventional programming paradigms and manual data collection scripts fail to maintain adaptability when data sources, formats, and ownership are distributed across systems. This research addresses these challenges by proposing the integration of Action-Oriented Programming (AcOP) with Automatic Agent Generation (AAG) as an innovative solution. AcOP focuses on actions as core execution units, providing clear definition of system behavior. The modularity of AcOP aligns with Edge AI principles by supporting localized, independent actions while ensuring coordination across distributed components. AAG leverages large language models (LLMs) to autonomously create intelligent agents that manage these actions and conduct preliminary data analysis using automatically assigned domain-specific knowledge. A microservice application for RCA and RA was developed in three implementations: Object-Oriented Programming (OOP), AcOP standalone, and AcOP with AAG integration. Results demonstrate that AcOP enhances modularity, adaptability, and error handling in decentralized environments. The integration of AAG further improves automation, reduces manual intervention, and enables more adaptive data collection, offering a flexible solution for autonomous microservice architectures.