Modeling the Stochastic State Matrix of a Production Line for Optimize its Operational Reliability Using Reinforcement Learning
Year: 2025
Type: article
Abstract: The development of a production line state determination model aims to create a universal tool for evaluating and optimizing industrial systems. The proposed approach enables real-time analysis of equipment states, prediction of potential failures, and enhancement of overall operational efficiency. The use of Markov chains allows for precise modeling of the sequence of production line states and the probabilities of transitions between them. This stochastic approach improves adaptability to real-world manufacturing conditions, surpassing the capabilities of traditional deterministic methods. The formation of a stochastic state matrix optimizes production processes through advanced data analytics and AI integration. This enables manufacturers to minimize downtime, enhance resource allocation, and improve overall productivity while maintaining operational stability. Transition probability estimation is based on both historical databases and real-time sensor measurements, allowing the model to adapt to various equipment types and operating conditions. AI-driven optimization enhances failure prediction accuracy, ensuring the production line remains efficient under diverse scenarios. By integrating Markov chains with data-driven insights, the approach supports proactive failure prevention and strategic resource management, ultimately improving the reliability and performance of industrial systems.
Language: English
Subfield: Safety, Risk, Reliability and Quality
Field: Engineering
Domain: Physical Sciences
Open Access status: bronze
OpenAlex ID: W4411385390