Optimal Arrangement of Linear Connected-(1,2)-or-(2,1)-out-of-(2,<i>n</i>):F System
Optimal Arrangement of Linear Connected-(1,2)-or-(2,1)-out-of-(2,<i>n</i>):F System
复制标题
线性连通-(1,2)-或-(2,1)-out-of-(2,<i>n</i>):F系统的优化排列
DOI：
10.11221/jima.72.295
复制
发表时间：
2022
影响因子：
--
通讯作者：
Tomoaki Akiba
中科院分区：
文献类型：
--
作者：
Taishin Nakamura;Hisashi Yamamoto;Tomoaki Akiba
现代社会现有系统的稳定运行至关重要。组件分配问题（CAP）对于通过有效使用组件来提高系统的可靠性非常重要。 CAP 涉及寻求一种安排，在给定系统中包含的组件的情况下最大限度地提高系统可靠性。为了解决这个问题，我们使用代表真实系统的模型作为系统模型，例如connected-X-out-of-(𝑚, 𝑛): F系统。连接-(𝑟, 𝑠)-out-of-(𝑚, 𝑛): F 系统的最佳排列是一种连接-X-out-of-(𝑚, 𝑛): F 系统，之前已被研究过。然而，线性连通-(1, 2)-或-(2, 1)-out-of-(𝑚, 𝑛): F系统尚未被研究，因为无法使用传统方法证明必要条件。因此，在本研究中，我们关注线性连接的-(1, 2)-或-(2, 1)-out-of-(2, 𝑛): F系统，并提出一种有效的最优排列搜索方法。为了有效地寻求该系统的最优布置，我们提出交替布置作为最优布置的必要条件以及应用该条件的算法。我们提出并证明证明定理所必需的引理。此外，我们将所提出的算法与枚举算法和不排除非交替排列的算法进行比较，并确认其有效性。
Stable operation of the systems existing in modern society is crucial. The component assignment problem (CAP) is important for improving the reliability of a system through efficient use of the components. The CAP involves the seeking of an arrangement that maximizes the system reliability given the components included in the system. To solve this problem, we use a model that represents a real system as a system model, such as the connected-X-out-of-(𝑚, 𝑛): F system. The optimal arrangement for a connected-(𝑟, 𝑠)-out-of-(𝑚, 𝑛): F system, which is a type of connected-X-out-of-(𝑚, 𝑛): F system, has been previously investigated. However, linear connected-(1, 2)-or-(2, 1)-out-of-(𝑚, 𝑛): F systems have not been studied because the necessary conditions cannot be proved using conventional methods. Therefore, in this study, we focus on a linear connected-(1, 2)-or-(2, 1)-out-of-(2, 𝑛): F system and propose an efficient search method for the optimal arrangement. To efficiently seek the optimal arrangement of this system, we propose alternating arrangements as the necessary condition for optimal arrangement along with an algorithm that applies this condition. We propose and prove lemmas necessary for proving the theorem. In addition, we compare the proposed algorithm with an enumeration algorithm and an algorithm that does not exclude nonalternating arrangements and confirm its effectiveness.