Cardiff University | Prifysgol Caerdydd ORCA
Online Research @ Cardiff 
WelshClear Cookie - decide language by browser settings

Modelling inspection and replacement quality for a protection system

Alberti, Alexandre R., Cavalcante, Cristiano A.V., Scarf, Philip and Silva, André L.O. 2018. Modelling inspection and replacement quality for a protection system. Reliability Engineering and System Safety 176 , pp. 145-153. 10.1016/j.ress.2018.04.002

[thumbnail of Modelling inspection and replacement quality R1 v4 AAM clean.pdf]
PDF - Accepted Post-Print Version
Available under License Creative Commons Attribution Non-commercial No Derivatives.

Download (821kB) | Preview


An inspection and replacement policy for a protection system is described by a mathematical model that incorporates multiple aspects of maintenance quality. A three-state component failure model is assumed, with a defective state preceding failure. The quality of maintenance intervention is modelled by supposing that inspections may misclassify defects (false positives and false negatives) and further that an inspection may induce a defect. The quality of replacement is modelled by supposing that a component arises from a heterogeneous population, composed of weak and strong items and with the mixing parameter determining quality. Isolation valves used in water distribution systems motivate the model development, and a case study is considered in this context. We evaluate the impact of these aspects of the quality of maintenance upon cost and production losses. Defect induction is found to be a key determinant of the cost-optimal policy. The proposed model allows us to verify conditions that justify investment in higher quality maintenance, and thus to provide guidance for prioritization of this investment.

Item Type: Article
Date Type: Publication
Status: Published
Schools: Business (Including Economics)
Publisher: Elsevier
ISSN: 0951-8320
Date of First Compliant Deposit: 7 December 2020
Date of Acceptance: 5 April 2018
Last Modified: 20 Jan 2021 16:20

Citation Data

Cited 19 times in Scopus. View in Scopus. Powered By Scopus® Data

Actions (repository staff only)

Edit Item Edit Item


Downloads per month over past year

View more statistics