مسار بولونيا

حسام كامل عيسى

السنة عنوان البحث نشر البحث
2025 A comparative statistical study for triple modular redundancy by using Fréchet distribution 2nd International Conference on Emerging Trends and Applications in Artificial Intelligence (ICETAI 2024)
The Triple Modular Redundancy (TMR) system is a powerful technique that works to reduce errors and failures that occur in the system because it works to increase the reliability of the systems, these systems consist of three components. These components are exposed to higher stress than the stress they face under the normal conditions. Some mathematical properties of the system distribution were derived, such as the probability density function (PDF), the Cumulative Density Function (CDF), the reliability of the system and the risk function. This study focuses on the comparison between two prediction methods (maximum likelihood and least squares) for the “TMR” system under the (partial life test) with gradual stress, which follows the (Frechete distribution). The comparison standard was the modified mean square error, and a simulation utilizing the (Monte Carlo) method in the (R) program was studied. This study aims to find out the most efficient method for this tripartite standard system. One of the most important conclusions of this study, the maximum likelihood method is better in prediction for the parameters and is more efficient because of the least mean square error rate compared to the other method, which reduces the risk of unexpected failure and economic losses.