Content deleted Content added
m Typo/quotemark fixes, replaced: ’s → 's, aquired → acquired |
Citation bot (talk | contribs) Altered author4. Add: bibcode, article-number. Removed URL that duplicated identifier. Removed parameters. | Use this bot. Report bugs. | Suggested by Headbomb | Linked from Wikipedia:WikiProject_Academic_Journals/Journals_cited_by_Wikipedia/Sandbox | #UCB_webform_linked 261/967 |
||
| (One intermediate revision by one other user not shown) | |||
Line 34:
== Data-driven reliability predictions ==
Data-driven models for reliability prediction utilise data acquired from tests to failure on electronic components by establishing relationships between the different variables presented in the data. As such relationships can be complex, data-driven models often require computations in high dimensions, which means that a large dataset is needed to optimize the output of the model.<ref>{{cite conference |last1=Ghrabli |first1=Mehdi|last2=Bouarroudj |first2=Mounira | author3=Chamoin, Ludovic|author4=Aldea,
== Physics-based reliability predictions ==
Physics based reliability predictions use physical equations and formulae to determine failure. This approach requires precise knowledge of the degradation process and the physical properties to ensure accuracy. These models often utilise numerical simulations to infer the quantities needed by the model.<ref>{{ cite journal | title=Physics-informed Markov chains for remaining useful life prediction of wire bonds in power electronic modules | journal=Microelectronics Reliability | year=2025 | last1=Ghrabli | author2=Bouarroudj, Mounira | author3=Chamoin, Ludovic|author4=Aldea,
| |||