Highly reliable military microelectronic components are the cornerstone of quality in aerospace and defense weaponry.Design and manufacturing are the origins of reliability，while screening tests serve as a critical means to ensure it.Extensive screening，however，also increases the cost of component usage，making it urgently needed to carry out optimization research on screening tests oriented towards cost reduction.Currently，numerous studies have been conducted on the correlation between screening tests and manufacturing defects at home and abroad.The American Solid State Technology Association has even established the screening optimization standards for tailoring and optimizing later-stage screening tests by controlling the manufacturing defects.In this paper，the U.S.screening optimization standards are analyzed，and the manufacturing defects related to burn-in are examined.With the production testing data from 1 479 domestically produced 4-to-6-line decoders，the conditions required for optimizing burn-in test duration are determined.Additionally，the electrical parameters of both failed and normal devices under dynamic burn-in conditions are analyzed.The results reveal that，compared with the 240-h dynamic burn-in during the primary screening，the 40-h dynamic burn-in in the secondary screening induces a greater change in the input high-level current of the components.Furthermore，the 40-h dynamic burn-in in the secondary screening significantly reduces the static current，whereas the 240-h dynamic burn-in in the primary screening notably increases it.This indicates that the relationship between the burn-in duration and parameter drift is nonlinear，i.e.，longer burn-in time does not necessarily exert greater influence on the components.These findings demonstrate the feasibility of optimizing the duration of burn-in tests.