Etd

An Integrated System Design and Safety Framework for Model-based Safety Assessment

Public

Downloadable Content

open in viewer

The growing autonomy and complexity of modern-engineering systems has introduced novel challenges to assessing their safety. Traditionally, safety assessment is performed using a combination of safety analyses, safety verification and testing at various stages in the life cycle. While testing occurs in the later stages of the life cycle, safety analyses and safety verification can be performed early in the life cycle on the available design models and offer potential solutions to the need for early identification of safety-related design issues. However, the manual nature of performing safety analysis and safety verification can introduce inconsistencies between the current system design model, the results from the safety analyses, and the results from safety verification. Additionally, due to the increased autonomy and complexity, small deviations caused by component degradation or unforeseen environment disturbances can lead to unanticipated system behavior. Current industry practice for safety assessment relies heavily on field-testing. However, field-testing is not a feasible solution to observe these rare events. Simulation testing offers a potential solution to observe such faulty system behavior caused by component degradation without relying on field-testing. While several approaches have adopted Model-based Development (MBD) for safety assessment, there is currently no framework or method that allows for feedback from safety analyses and safety verification to the system design model while also using simulation testing to observe faulty system behavior caused by degraded components or environmental disturbances. This dissertation presents a model-based safety assessment framework, called the Integrated System Design and Safety (ISDS) framework, for assessing the safety of system design models early in the life cycle. The proposed framework combines a model-based safety analysis approach with a model-based safety verification approach to complete the safety assessment. The objective of the proposed framework is to eliminate sources of inconsistencies in the safety assessment process as well as to improve the safety of the system by using simulation testing to observe faulty system behaviors caused by component degradation. The ISDS framework is applied to a case study involving the development and safety assessment of a Forward Collision Warning (FCW) system in an autonomous vehicle. Results show that the feedback mechanism of the ISDS framework can eliminate the manual tasks that introduce inconsistencies in the safety assessment process as well identify a wider range of faulty system behavior compared to the current state-of-the-art. The key contributions of this research are, a) the feedback mechanism of the ISDS framework, which eliminated the need to manually update the system design model with the results from the safety analyses and safety verification, and b) the fault injection engine of the ISDS framework, which enables the use of simulation testing to identify faulty system behaviors caused by component degradation.

Creator
Contributors
Degree
Unit
Publisher
Identifier
  • etd-42846
Keyword
Advisor
Committee
Defense date
Year
  • 2021
Date created
  • 2021-12-13
Resource type
Rights statement

Relations

In Collection:

Items

Items

Permanent link to this page: https://digital.wpi.edu/show/2227ms79d