第一回研究会(通算第十九回)

日本信頼性学会 春季シンポジウムにおける
システム信頼性研究部会オーガナイズドセッション

日時:2017年5月31日(水)13:00から(開催プログラムはこちら
場所:一般財団法人 日本科学技術連盟 本部(西新宿2-7-1小田急第一生命ビル

第一回研究会では,日本信頼性学会主催の春季シンポジウムにて,オーガナイズドセッションを企画し,以下の4件の講演を予定しております.みなさまのご参加をお待ちしております.春季シンポジウムへの参加登録方法などはこちらをご覧ください(リンク

(1)
Title: A note on a statistical detection method of cascading failure for parallel systems
Authors: Shuhei OTA and Mitsuhiro KIMURA (Hosei University)
Abstract: Cascading failure, one kind of dependent failures, is a phenomenon in which the failure occurrence of one component triggers other failures. The trigger component can cause a huge number of failures after it fails if other components strongly rely on the trigger component. For example, this phenomenon is observed in complex network systems such as blackouts of power transmission systems. As a result, the cascading failure causes the reliability deterioration of the systems. From the viewpoint of reliability management, the factors of the cascading failure should be ideally eliminated before the utilization of the system. In this study, we propose a statistical detection method of the cascading failure occurrence in n-component parallel systems. This research contributes to the cause analysis of the cascading failure occurrence in the n-component parallel system. The cascading failure is modeled by hazard rate switching mechanism. We assume that the lifetime distribution of the remaining components changes if such a trigger component actually exists and fails. The cascading failure occurrence can be detected by identifying the change of the distribution. The performance of the proposed method is demonstrated by simulation studies.

(2)
Title: Conditions for the optimal arrangement of a connected-(m-1,s)-out-of-(m,n):F lattice system
Authors: Taishin NAKAMURA, Hisashi YAMAMOTO, Sayaka MATSUO, Xiao XIAO (Tokyo Metropolitan University), and Tomoaki AKIBA (Chiba Institute of Technology)
Abstract: A connected-(r,s)-out-of-(m,n):F lattice system consists of m×n components arranged as an (m,n) matrix, and fails if and only if the system has an (r,s) sub-matrix where all components are failed. One of the most important problems in reliability theory is the component arrangement problem (CAP) on the assumption that component reliabilities are given, and components are interchangeable. The CAP is to find the optimal arrangement of components to maximize system reliability. By taking CAP into account, we can make the best use of limited resources and maximize the performance of the system.
In this study, we provide necessary conditions for the optimal arrangement of the connected-(r,s)-out-of-(m,n):F lattice system in the case of r=m-1. Since we only need to calculate the reliability of the systems corresponding to the arrangements satisfying the necessary conditions, our proposal necessary conditions can considerably reduce the search space for the patterns of component arrangements. We evaluate the performance of the proposed algorithm by numerical experiments.

(3)
Title: Compromised life test plan for accelerated degradation test of three factor stress level
Author: Takenori SAKUMURA (Chuo University)
Abstract: Accelerated degradation tests (ADT) are often used to quickly find the life of materials such as electrical insulators. An ADT consists of several high levels of stress and the number of samples at that level. Each sample is subjected to the stress for a certain evaluation time and the degradation rate at that time is measured. The time when the degradation reaches a certain threshold is regarded as the failure time. The lifetime can be predicted from the obtained failure time data under the assumption that the physical law is empirically established between the stress and the lifetime and the logarithmic lifetime follows a specific probability distribution under a certain stress. In this research, focusing on finding the optimum sample allocation number, consider the case where the number of stress levels is three. In order to consider the reality, we use the mathematical model obtained from actual experimental data and its parameter value.

(4)
Title: Analysis for the Conditions of Component Assignment for the Optimal Arrangement of a k-Window System
Author: Tomoaki Akiba (Chiba Institute of Technology)
Abstract: k-window system in this study is an application system of the consecutive-k-out-of-n:F system. This system can be expressed to the occurring of a failure in the system when a certain extent intensive components failure happens in the system. One of the most important problems for this kind of system is the optimal arrangement problem. In the previous study, we confirmed by the simulation approaches that optimal arrangement of k-window system. However, this component arrangement depends on the system parameters k, r and components reliabilities. In this study, we provide necessary conditions of component assignment for the optimal arrangement of k-window system by the mathematical analysis.

広告
カテゴリー: ご案内 | コメントをどうぞ

第十八回信頼性研究会

日時:2016年12月17日(土) 10:00~12:00
場所:マホロバマインズ三浦(神奈川県三浦市南下浦町上宮田3231)
ホテルのwebサイトはこちら

2016年度最後のOR学会信頼性研究会は, 上記の通り神奈川県三浦市で開催する予定です. 前日の16日(金)には電子情報通信学会信頼性研究会が同じ場所で開催されますので(発表申込:10月17日), 併せてご案内申し上げます.

講師と題目
(1)
講師:北川智大 氏(防衛大学校)
題目:ランダム作業時間に従事する船舶搭載システムの保全方策について
概要:船舶がある確率分布に従うランダムな長さの航海に従事する場合,それに搭載されているシステムについても同じ時間だけ作業することが求められる.このような船舶搭載システムの保全は,一般的に航海終了後に岸壁やドックで行われるため,修理等の遅れ時間を加味する必要がある.本発表は2部構成となっており,第1部では,システムの故障が船内の予備アイテム等により即座に修理可能な場合と,航海終了後に修理される場合とに確率的に区別できる問題を扱う.この確率が予備アイテム等の搭載コストと既知の関係があると仮定し,ある平均アベイラビリティを満たしつつ,平均コストレートを最小にする最適な確率について議論する.これは,船舶に搭載する予備アイテム等にかけるコストを最適化することと同等である.第2部では,ランダムな長さの航海時間が事前に付与される状況を想定する.システムの故障に対して,航海中に緊急的に修理するか,航海終了後に修理するか選択できるとし,故障発生時刻に応じた合理的な修理時期の選択ついて議論する.

(2)
講師:太田修平 氏(法政大学大学院)
題目:依存故障するn素子直並列システムの信頼性解析~EFGMコピュラを用いて~
概要:互いに依存して故障する素子によって構成される,コヒーレントなシステムの故障の特性を解析することは,信頼性工学において,非常に重要な課題である.本講演では,複数素子の寿命時間が互いに確率的に依存している,n素子直並列システムの信頼性解析について議論する.具体的には,n素子直並列システムを,同時分布関数のEFGMコピュラによる表現を用いてモデル化し,そのシステムの信頼性評価を行う.特に,n変量EFGMコピュラがもつ,依存パラメータの制約に注意しながら,システムのMTTFを解析的に導出し,システムの寿命と,素子間の依存性の強さの関係について議論する.また,いくつかの数値例とともに,モデルの特性について述べる.

(3)
講師:Wong Young Yun 氏(Pusan National University, Korea)
題目:How to determine maintenance units in modularized systems?
概要:In this paper, we consider an optimization problem in which we determine the maintenance units in modularized systems. The modularized system has a hierarchical structure in design phase and has several modules. Each module also consists of some components. When some components fail, then the system fails and we should maintain correctively the system. We can replace the failed components or the module with the failed components. We want to determine the maintenance units in the modularized system with hierarchical structure. The system availability and maintenance cost are used as optimization criteria. Numerical examples are also studied.

懇親会(研究部会忘年会):研究会終了後, 横須賀市内で開催を予定しています.

【注意】研究会への参加費は無料ですが,資料等の準備の関係上,聴講を希望する方は 12月9日(金)までに幹事までお知らせください.

問合せ先 法政大学 田村信幸 E-mail: sigrelwebmaster@gmail.com
信頼性研究部会HP(10月までの情報はこちら

カテゴリー: ご案内 | タグ:

第十七回信頼性研究会

日時:2016年10月15日(土) 12:00~19:00 (情報交換会: 17:15~19:00)
場所:常翔学園大阪センター
大阪市北区梅田3-4-5 毎日インテシオ3F
(研究会場:301+302セミナー室、情報交換会場:ラウンジ翔)

本会は以下4研究部会合同での開催となりました.

主催:
「確率モデルとその応用」研究部会
「信頼性」研究部会
「待ち行列」研究部会
「不確実性環境下の意思決定モデリング」研究部会

当信頼性研究部会からは,以下の講演者を推薦し,ご登壇頂きました.

講演者: 早川 有(早稲田大学)
講演題目: Delayed reporting of faults in warranty claims
共同研究者: Richard Arnold, Stefanka Chukova and Yu Hayakawa
講演概要: When a complex system is operated, it may experience multiple faults. If the system is operating under warranty these faults may be claimed for and repaired at zero or minimal cost to the consumer. However if the faults do not lead to system failure the user may find it inconvenient to claim for each repair as it occurs, and may instead delay making a report or claim until a sufficiently large number of faults has accumulated. In this talk, the speaker presented a model for the delayed reporting of faults: multiple non-fatal faults are accumulated and then simultaneously reported and repaired. The reporting process is modelled as a stochastic process dependent on the underlying stochastic process generating the faults. The joint distribution of the reporting times and numbers of reported faults is derived.

その他,詳細はリンク先をご覧ください(リンク

カテゴリー: ご案内