Random field-based probabilistic support for the assessment of the electrical conductivity of industrial floors and their operational reliability in time

Abstract In modern engineering, electro-resistant flooring plays a crucial role in preventing electrostatic discharges that may pose safety hazards to people or can damage sensitive electronic equipment. The electrical resistance of industrial floors, characterized primarily by the electric leakage resistance parameter, is often highly variable due to localized material imperfections and temporal fluctuations. Hence, enforcing strict regulatory requirements to ensure long-term reliability and serviceability of such flooring necessitates systematic in-situ testing. While these measurements provide important insights into the current state of the industrial floor, they also suffer from proprietary limitations, environmental variability, and instrument sensitivity. These issues may lead to situations where a single set of measurements fails to represent the overall performance of the flooring. Furthermore, a high economic cost of frequent testing must also be addressed. To account for these challenges, a probabilistic approach is introduced, enhancing the possibility of interpretation of the variability in the in-situ results and enabling long-term predictions of the flooring operational reliability. The methodology combines annual real-life measurements with simplified stochastic updating models. Representative correlated random fields are generated in the process, with Monte Carlo simulations and Cornell linear reliability index approach used to evaluate the failure probability and reliability change over time. By integrating real-live tests of the flooring with their simulated random field-based numerical counterparts, the proposed approach refines prediction accuracy and helps engineers make more informed maintenance decisions.