An enhanced GO methodology with multiple signal transmission types based on Bayesian network

GO methodology is a visual modeling approach, which can be used to analysis the reliability and safety of dynamic complex systems. However, with only one type of signal flow, traditional GO methodology cannot depict the system with multiple transmission types, which makes it hard to reflect the complex delivery relationship in system. This also leads to an incomplete expression of heat and humidity transmission. What’s more, different failure modes can result in distinct influences, which is beyond the describing scope of a single type of signal flow. In this paper, we present an enhanced GO methodology with multiple types of signal transmission, and apply Bayesian Network (BN) for calculation. By introducing new signal flow type, the model can describe multiple transmission. And the type of signal flow can be defined according to the system requirements, not restricted to default forms. Then, an algorithm based on BN is given to implement the ex-tended model. For every operator, each type of transmission is corresponding to a Conditional Probability Table (CPT), which depicts the states of current signal type. Afterwards, the interaction between CPT in the same operator is established, and the failure rate of the operator considering multiple signal is calculated. Finally, the calculation process between the operators is stipulated.