City Research Online

Abstract

Guaranteeing a prescribed level of reliability for a complex multi-component system is the problem considered in this thesis. For this, optimal maintenance and inspection policies are derived, taking into account the different deteriorations the components in the system are subject to. These degradations are described with the use of continuous stochastic processes in time and are assumed not to be directly observable. Maintenance decisions are based on a performance measure defined by a functional acting on the system state process. The transient property of the performance measure enables a modified control limit rule, based on last exit times, to be considered. For this a critical level is defined and the probability of never returning below it is used in the decision making approach. A condition based maintenance policy is investigated with the use of a bijective function of the performance measure’s value, that determines the required amount of repair. Both periodic and non-periodic inspections are studied. The nonperiodic approach is handled with the use of an inspection scheduling function which assigns the amount of time between two consecutive inspections to the value of the performance measure at inspection. Two main types of models are proposed. The first type considers one threshold and focuses in guaranteeing a prescribed level of reliability for systems where crossing of a critical boundary does not cause immediate failure but will require action later. Examples include physical infrastructures such as roads. Models of the second type take failure of the system into account with the incorporation of a second threshold. Examples include aeroplanes where safety regulations imply regular inspections and repairs. Occurrence of infrequent catastrophic failures must however be considered.

Publication Type:	Thesis (Doctoral)
Subjects:	H Social Sciences > HA Statistics Q Science > QA Mathematics T Technology > TA Engineering (General). Civil engineering (General)
Departments:	School of Science & Technology > Mathematics School of Science & Technology > School of Science & Technology Doctoral Theses Doctoral Theses

Export

Downloads