Literature Review: Global Criticality Assessment Based on Feature Surrogates at the PCBA Levels

Reliability testing is essential in the PCBA (Printed Circuit Board Assembly) manufacturing process since it helps to identify potential issues before they become significant in the future. This shall ensure the quality and long-term performance of the product. However, the challenge is that the experimental reliability test during manufacturing is quite time-consuming and cost-consuming, and finding a new method to accelerate the test itself is very difficult. Currently, virtual performance predictors such as the finite element method (FEM) form a key for assessing the criticality of a PCBA product, but they are still limited due to their costly computational power if the problem is geometrically complex or difficult to handle, as electronic devices become more compact and sophisticated. With the rapid development of artificial intelligence, novel modelling techniques involving machine learning (ML) and deep learning (DL) have gained extensive attention from engineers and this is especially the case in the field of electronic system reliability. By building an appropriate surrogate model, designers need to input only the component geometry of the newly designed electronic product to immediately obtain a lifetime estimation of the specific product. The given work reviews the state-of-the-art surrogate model-based approaches used in electronic systems, specifically in the field of PCBA. The challenges and future development of surrogate modelling in support of global criticality assessment at the PCBA level are also analysed and discussed.