Safety + AI: A Novel Approach to Update Safety Models Using Artificial Intelligence

Youcef Gheraibia,Koorosh Aslansefat,Ioannis Sorokos,Yiannis Papadopoulos,Sohag Kabir

doi:10.1109/access.2019.2941566

Youcef Gheraibia, Koorosh Aslansefat + Show 3 more

Open Access

https://doi.org/10.1109/access.2019.2941566

Copy DOI

Abstract

Safety-critical systems are becoming larger and more complex to obtain a higher level of functionality. Hence, modeling and evaluation of these systems can be a difficult and error-prone task. Among existing safety models, Fault Tree Analysis (FTA) is one of the well-known methods in terms of easily understandable graphical structure. This study proposes a novel approach by using Machine Learning (ML) and real-time operational data to learn about the normal behavior of the system. Afterwards, if any abnormal situation arises with reference to the normal behavior model, the approach tries to find the explanation of the abnormality on the fault tree and then share the knowledge with the operator. If the fault tree fails to explain the situation, a number of different recommendations, including the potential repair of the fault tree, are provided based on the nature of the situation. A decision tree is utilized for this purpose. The effectiveness of the proposed approach is shown through a hypothetical example of an Aircraft Fuel Distribution System (AFDS).

Highlights

Safety critical systems are systems for which human life, environmental health, and financial assurance need to be guaranteed
To address the above-mentioned issues related to unforeseen events and misunderstanding about system behavior, in this paper, we propose a data-driven approach using machine learning to provide assistance when an unknown or unconsidered scenario encountered during system operation
THE PROPOSED APPROACH The approach proposed in this paper considers that at design time, safety analysts have knowledge about the system model and behavior, and they have already created a fault tree of the system based on their knowledge about the foreseeable failure events

Summary

INTRODUCTION

Safety critical systems are systems for which human life, environmental health, and financial assurance need to be guaranteed. The probability of either a system functioning correctly without any fault during the mission time or terminating its service(s) through a safe procedure can be called safety [5]. Another example would be aircraft emergency landing safety. Because of the criticality of the system functioning, a rigorous reliability and safety evaluation requires comprehensive and certified model(s) that are usually provided by a team of high-level experts. Basic events: in the FTA, a system can decompose to sub-systems and each sub-system can decompose to sub-sub-systems This procedure will continue to the level that no more decomposition is affordable or possible. An error in the model (e.g. FT) used for providing evidence can make the safety guarantee void, make safety certificate invalid

MOTIVATION AND CONTRIBUTIONS

AND RELATED WORKS

DECISION-MAKING PROCESS

CASE STUDY EVALUATION

Findings

CONCLUSION