An Information Processing Theory Framework for Intelligent Fault Diagnosis and Predictive Maintenance

Divya D¹Arunkumar O N^2*

• ¹International School of Business and Research, Bengaluru, India
• ²Symbiosis Institute of Business Management, Symbiosis International University, Bengaluru, Bangalore, India

This paper formulates and applies an Information Processing Theory (IPT)-grounded model for intelligent fault diagnosis and predictive maintenance (PdM). The introduced model makes associative connections between task precursors and their respective information-processing requirements (IPR), and the technological and organizational machinery providing information-processing capacity (IPC). It also introduces a layer of translation that couples predictive intelligence with prescriptive maintenance activity. Building on digital twins, multi-sensor data fusion, federated and edge learning, and multi-agent orchestration literature, the research defines essential theoretical constructs, develops testable hypotheses, and prescribes mixed-method empirical methods for validating and measuring them. The major contributions of the paper are an exhaustive multi-layer fit framework covering the technical, organizational, and ecosystem aspects; the development of propositions on mechanism complementarity; and a set of design rules of practical use for application. The design rules focus on diagnosing IPRs prior to choosing mechanisms, building complementary modules, designing convertible interfaces, and tracking fit as a performance KPI. Collectively, the framework connects technical performance and organizational results, creating a clear guide for building adaptive, explainable, and operationally effective Predictive Maintenance (PdM) systems.