Performance and reliability assessment of cloud-based machine repair with mixed standby under fuzzy environment

Purpose This study investigates the reliability and availability characteristics of a cloud-based machine repair system composed of two active machines, two warm-standby units and two heterogeneous repairmen. To realistically represent uncertainty arising from operational variability, the system failure and repair rates are modeled under a fuzzy environment. Design/methodology/approach The system is formulated as a continuous-time Markov process. The Chapman–Kolmogorov differential equations are used to derive expressions for mean time to failure (MTTF) and steady-state availability. The failure and repair rates are represented using trapezoidal and triangular fuzzy numbers, and Yager’s ranking method is employed to defuzzify the performance measures and obtain their crisp equivalents. Findings Results show that incorporating warm-standby redundancy and heterogeneous repairmen enhances system reliability. Fuzzy modeling successfully captures parameter imprecision, and numerical analysis demonstrates how varying degrees of uncertainty influence MTTF and availability. Research limitations/implications The model assumes exponential distributions for failure and repair times and does not consider switching delays or priority-based job scheduling policies. Future work may extend the model to non-exponential or data-driven distributions. Practical implications The findings guide maintenance planning and resource allocation in cloud and distributed computing environments, where uninterrupted service and efficient repair strategies are critical. Social implications Enhanced reliability in cloud-based systems supports continuous service delivery for sectors such as healthcare, finance, education and digital governance, improving overall user trust and operational resilience. Originality/value This work provides a novel fuzzy reliability framework that integrates warm-standby redundancy and heterogeneous repair capabilities. Additionally, both trapezoidal and triangular fuzzy number representations are employed to analyze and compare fuzzy MTTF and system availability, offering more realistic insights under uncertainty.