ISE Magazine

FEB 2017

Issue link: https://industrialengineer.epubxp.com/i/777540

Contents of this Issue

Navigation

Page 51 of 67

52 ISE Magazine | www.iise.org/ISEmagazine Inside IISE Journals research Interval estimation in load-sharing systems In a system with multiple components, an operating component's performance usually is affected by other components. A common example is the load-sharing system. In a load-sharing system, a constant total load is imposed on the system and shared by all components. The load is re- distributed to all surviving components if one component fails. The additional load shortens the residual lifetimes of the surviving components. Some com- mon examples are cables in a suspension bridge, parallel pipes in a piping system for oil transmission and servers in a cloud computing system. Reliability estimation for load-sharing systems is considered a challenging task due to the dependent nature of the com- ponent lifetimes. This challenge was taken up in "In- terval Estimation for k-out-of-n Load- Sharing Systems" by doctoral student Yaonan Kong and assistant professor Zhisheng Ye from National University of Singapore. They developed two inter- val procedures for the model parameters using pivotal quantities and generalized pivotal quantities. Pivotal quantities are functions of the data and model param- eters whose distribution does not depend on any unknown parameters. The paper also discussed interval estimation for im- portant reliability characteristics, includ- ing the expected lifetime and reliability of the load-sharing system. A simulation study shows that the confidence intervals produced from the proposed procedures are more accurate compared with traditional approximate interval procedures such as the large sample normal approximation and the bootstrap. The proposed procedures have been applied successfully to a deep-groove ball bearings example reported in 2006. To test the lifetime of the ball bearings, two bearings are run simultaneously on a single shaft. A bearing fails when cra- ters are developed on the running paths. The first ball-bearing failure introduces an exfoliation of material and may con- taminate the common oil supply shared by the two bearings. This could result in earlier failure of the surviving bearing in the same system. Thus, the ball bearings are dependent, and their failures could be modeled by a load-sharing process. Based on failure data from six systems, interval estimation for the model param- eters and system reliability was obtained. The confidence intervals ensure accu- rate uncertainty quantification during This month we highlight two articles from IISE Transactions describing advances in quality and reliability modeling. The first article studies reliability estimation of load-sharing systems. In a load-sharing system, a component failure redistributes the load across the remaining components, thus increasing their stress level. The second article develops a procedure for integrating knowledge about causal relationships between categorical variables into statistical process control methodology. These articles will appear in the March 2017 issue (Volume 49, No. 3). Yaonan Kong (left) and Zhisheng Ye discuss reliability of load-sharing systems.

Articles in this issue

Links on this page

Archives of this issue

view archives of ISE Magazine - FEB 2017