• Login
• Admin

Zhang, Yang (2016) Development of BDD models for decision support in phased mission systems. PhD thesis, University of Nottingham.

PDF (BDD Models for Decision Support in Phased Mission Systems) (Thesis - as examined) - Repository staff only - Requires a PDF viewer such as GSview, Xpdf or Adobe Acrobat Reader Download (3MB) Request Thesis

Abstract

Autonomous systems are becoming increasingly commonplace, with applications either existing or suggested in many different industries. As levels of autonomy increase, the need for these systems to interpret with environments in which they operating and make decisions about their own future actions following internal failures or external threats. In the past, reliability analysis methods have been suggested as having the potential to provide information that could be used in a real-time decision support tool for autonomous systems in changing environments. Real-time support is particularly important in systems such as unmanned aerial vehicles (UAV), where any delay in making a decision following a failure occurrence or the emergence of a threat could be catastrophic.

Reliability Analysis can be used to calculate the failure probability of a mission such as that performed by a UAV by modelling the mission as a sequence of tasks known as a phased mission.

Binary Decision Diagram models have shown great potential for analysing phased mission systems since they can produce accurate mission and phase failure probabilities in reasonably short time frames. Although research to date has shown that Binary Decision Diagrams appear to have the most promise for performing the real-time analysis that would be required as an input to a decision making tool for phased mission systems, the analysis as it stands still falls some way short of being near-instant, as it must be for decisions to be made quickly when required. In common with many systems, phased mission systems can contain components that fail in multiple failure modes. It is therefore important that multiple failure modes are modelled while developing the Binary Decision Diagram tools and techniques considered in this research.

The research presented in this thesis aims to address the deficiencies seen in previous methods by investigating the Binary Decision Diagram techniques and suggesting how the techniques can be developed for use within a decision support tool where fast, accurate decision making is required. The novelty of the research is as follows:

1. Different Binary Decision Diagram models for phased mission systems are reviewed and three new Binary Decision Diagram models are proposed to improve the efficiency and accuracy of analysis for phased mission systems containing multiple failure mode components.

2. Since the size of a Binary Decision Diagram has a significant effect on the time required to quantify it and the Binary Decision Diagram size is influenced by variable ordering, nine different variable ordering schemes are investigated for phased mission systems. Eight of them are extended from fault tree analysis of single phase systems containing single failure mode components and one is newly-developed specially for use within a decision support tool.

3. Due to the potential time limitation for decision making, approximation methods are investigated to evaluate the failure probabilities in phased mission systems while trading off between accuracy and analysis efficiency. Three new approximation models are developed and their analysis efficiency advantage over the exact analysis is demonstrated testing on a large number of sample phased mission systems. A performance indicator is developed in order to facilitate the choice of approximation method taking into account accuracy and efficiency requirements.

The benefits of the developed methods are demonstrated through the consideration of a case study.

Item Type:	Thesis (University of Nottingham only) (PhD)
Supervisors:	Prescott, D. Andrews, J.
Keywords:	Binary Decision Diagrams(BDDs), Reliability analysis, Phased Mission Systems(PMS), Decision making
Subjects:	T Technology > TA Engineering (General). Civil engineering (General)
Faculties/Schools:	UK Campuses > Faculty of Engineering
Item ID:	32431
Depositing User:	Zhang, Yang
Date Deposited:	28 Jul 2016 08:32
Last Modified:	19 Oct 2017 15:37
URI:	https://eprints.nottingham.ac.uk/id/eprint/32431

Actions (Archive Staff Only)

Edit View