Abstract

abstract

Maintaining industrial system in  high availability is particularly crucial and significant in production processes, which in turn affect profitability. This paper will develop a new Markov Decision Process model for system availability optimization under budget constraints. The primary objective will be to optimize long-run availability, meaning maximization of expected system uptime at minimum cost of operation per unit time. Given the dual objectives, our model will focus on availability and consider the cost as one of its constraints. Such a model provides practical maintenance decision-making by integrating real-world data within the MDP framework. We solve the proposed model using linear programming, and results show a balance between system performance and financial sustainability. Further challenges remain in the estimation of the parameters of the system accurately. In this respect, future work is needed to enhance the robustness and adaptability of the model to handle such complex systems in real-time environments.

Keywords
1. Markov Decision Process, 2. Availability, 3. Maintenance, 4. Budget Constraint, 5. Sustainability in Maintenance.