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

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).

Connectors of smart design and smart systems

“…The current research results still show an existing limit line at designing smart embedded systems for dependability (Srivastava and Singh, 2009). The issues related to the evaluation of the dependability of CPSs during the design period are recognized (Gheraibia et al , 2019), but providing any generic solution is difficult due to the increasing aggregate complexity, the operational and environmental dynamics, and the consequences of self-adaptation (Sondermann-Wölke et al , 2010). The dependability of complex smart systems is ontologically emergent and compositionality also plays a crucial role in it (Hartmann, 2014).…”

Noise Resistant Multidimensional Data Fusion via Quasi-Cliques on Hypergraphs

“…Value represent the total number of positives accepted by the statistical test (from all possible paired permutations) on the right and the number of unique columns from the first dataset with at least one match on the left. To test the algorithms, we have run them over two pairs of relations from the KEEL regression datasets [28], the training and test catalog from the Euclid photometric-redshift challenge [29], and a set of sensor measurements from an aircraft fuel distribution system [30]. Some statistics about these datasets are summarized in table 5.…”

A Data Mining Approach for Transformer Failure Rate Modeling Based on Daily Oil Chromatographic Data

“…Suppose that the maintenance is performed on the transformer at time t 3 in a warning state, which results in a significant decrease in the failure rate at time t3. The SDs in state 2 at time t1 and t2 are equal, and the failure rates at time t1 and t2 calculated by formula (16) and are also the same. The reason for equal SDs at times t1 and t2 is that the Markov model always assumes that the transformer can return to the initial state after the maintenance is performed at time t3 [22].…”

“…The total failure rate at time t can be modified from (15) and (16) by replacing the parameter ESD (27) Combining (26) and (27), the analytical failure rate model proposed in this paper is established.…”

“…[31] established a conceptual framework to incorporate complex basic events in hierarchically performed hazard origin and propagation studies (HiP-HOPS), which can guarantee the modelling capability on complex failures and the efficiency of Model-based safety analysis (MBSA) effectively. [16] gave the detailed classification for safety models associated with machine learning (ML) and further proposed a novel approach based on ML and real-time operational data to reduce the difficulty in modeling and evaluation these complex safety-critical systems. In [32], the state analysis method is introduced for generic power system reliability assessment.…”