Alex Davila-Frias and Om Prakash Yadav*
Estimating the All-Terminal Network Reliability (ATNR) by using Artificial Neural Networks (ANNs) has emerged as a promissory alternative to classical exact NP- hard algorithms. Approaches based on traditional ANNs have usually considered the network reliability upper bound as part of the inputs, which implies additional time-consuming calculations during both training and testing phases. This paper briefly reviews and compares the results of our recent work on advanced neural networks for ATNR, which dispense with upper bound input need and offer improved performance. The results are compared with traditional ANNs in terms of features such as the error (RMSE), execution time, or the ability to relax the perfects nodes assumption, among others. A quick discussion highlights the fact that modern neural networks outperform traditional ANN; however, there are trade-offs in the performance of advanced neural networks. Such trade-offs provide an opportunity for future research efforts as, suggested in this paper as well.
HTML PDFShare this article
