Selective Maintenance for Multi-state Systems Considering Dependence

• Author / Creator

  Dao, Cuong D

• Industrial organizations rely on the safe and efficient operation of their assets to manufacture products and provide services with high quality. As a result of widespread mechanization, maintenance costs have increased significantly over the years. Maintenance is essential to keeping engineering systems (mechanical equipment, vehicles, electrical devices, etc.) reliable at minimum costs. Due to limited resources, e.g. budget, time, it may not be possible to do all desirable maintenance actions. In this context, the maintenance manager has to decide which subsystems or components should be maintained in order to meet the requirements on system’s performance. This problem is called selective maintenance. In traditional reliability theory, the system is considered to be in two possible states of perfect functioning or failed. However, in practice, many systems can degrade and operate in an intermediate working state. As the system deteriorates, its performance may be in several states varying from perfect functioning to complete failure. Such a system is called multi-state system. When the system and its components are multi-state, the decision making becomes more complicated since imperfect maintenance actions to its intermediate levels are also possible. The maintenance manager has to decide which components to be maintained and the state levels that they should be maintained to. Existing selective maintenance models usually assume the components to work in a stable condition without considering the inter-relationships between them when maintenance activities are performed as well as when the system is running. However, components interact with each other due to several reasons. For example, the failure of a component may create fire and cause immediate failures of other components due to an induced failure mode, or the performance of a component may dictate the degradation of other components in a specific system design. This PhD research aims to study selective maintenance modeling for complex systems, specifically selective maintenance for systems with multiple working levels, while considering dependent relationships in the system. This research tackles the maintenance problem when resources are limited by providing maintenance strategies using both systems performance requirements and maintenance resources as input information. Several types of relationships of components in the systems are modelled as follows. i. When maintaining several components simultaneously, there are savings due to the sharing of resources, e.g. materials, tool, manpower, etc. ii. In a complex system structure, repairing a component requires actions on other components. iii. The current health state of a component may affect the performance and degradation of other components. iv. The degradation rate of a component may change depending on the operating conditions and its current health state. This research proposes maintenance models for multi-state systems with dependence, which is one of the most challenging topics in the field of reliability and maintainability. It has significant contributions in terms of dependence modelling, reliability analysis, and selective maintenance optimization of multi-state systems. The results can be applied in a wide range of industries where physical mechanical systems exist and maintenance for them is compulsory. The maintenance models not only provide decision makers with a set of maintenance actions to ensure reliable and cost-effective operation of the systems, but also help them wisely utilize their available resources.

• Subjects / Keywords

  • Dependence
  • Selective maintenance
  • Multi-state system

• Graduation date

  Fall 2016

• Type of Item

  Thesis

• Degree

  Doctor of Philosophy

• DOI

  https://doi.org/10.7939/R3V11VR2H

• License

  This thesis is made available by the University of Alberta Libraries with permission of the copyright owner solely for non-commercial purposes. This thesis, or any portion thereof, may not otherwise be copied or reproduced without the written consent of the copyright owner, except to the extent permitted by Canadian copyright law.