Launch failure analysis – is Blue’s return to flight schedule realistic?

New Glenn in the factory. Credit: Blue Origin

June 9, 2026

By Caleb Henry

Blue Origin’s May 28th hot-fire explosion upended the company’s push to make 2026 the year it shifted into high gear. The open question on the space industry’s mind is how big of a setback this will be, especially for key customers Amazon, AST SpaceMobile, and the U.S. government.

The explosion drew many comparisons to SpaceX’s Amos-6 failure in 2016, which halted Falcon 9 operations for about five months and required 15.5 months to return the launch pad to service. Blue now says launches will resume by the end of the year, so within seven months. 

Analysis of past medium-to-heavy-lift rockets provides guidance for what comes next.

Throughout the 2000s, four main vehicles dominated commercial launch: Russia’s Proton, Europe’s Ariane 5, SpaceX’s Falcon 9, and Sea Launch’s Zenit.

Proton had 10 failures from 2007 to 2015, returning to flight on average 77 days later, or about 2.5 months. While failures varied in type, Russia was able to quickly return the vehicle to service thanks in part to data collected from about 400 launches dating back to the 1960s, giving engineers a deep knowledge of the rocket.

Ariane 5 had a troubled start in the 1990s, failing four times in its first five years before normalizing. The first failure, on the rocket’s maiden flight, took 513 days to recover from. Each subsequent failure required on average 39% less time to return to flight. The rocket enjoyed a clean flight history for 20 years (2003-2023) blemished once by a software glitch that placed two satellites in the wrong orbit in 2018. That anomaly took 70 days to recover from, as Ariane 5 had launched nearly 100 times and could quickly analyze the cause of the glitch.

Falcon 9 had three failures in its early days from 2012 to 2017, averaging 152 days (~5 months) to return to flight. After the 2017 failure, Falcon 9 launched for seven years without payload-destroying hiccups. The last failure was in 2024 when a Falcon 9 upper stage anomaly cost a batch of Starlink satellites, but by then the vehicle had launched more than 300 times, providing a wealth of data to study for rapid identification of the anomaly’s source. SpaceX returned the Falcon 9 to flight in just 15 days.

Sea Launch, a multinational launch provider active from 1995 to 2014, was an outlier. Already infrequently launched, the Zenit rocket averaged 285 days (9.5 months) before returning to flight, counting four failures across 13 years. Sea Launch recoveries did not improve with time. Having operated prior to the age of constellations and without an anchor government customer to weather hard times, Sea Launch ceased operations in 2014 and after an ownership shuffle, shut down sometime thereafter.

Conclusion: if Blue Origin’s early assessment of its launch pad damage holds true, precedent suggests a return to flight this year is possible. Blue Origin has the launch demand to avoid Sea Launch-style delays, and spaceport damage appears sufficiently contained to avoid an Amos-6-style setback. There is risk, however, in the paucity of New Glenn launch data to draw on, meaning Blue Origin will have more homework to do than experienced launch providers in returning to flight. The newness of methalox as a fuel source also introduces uncertainty (lightened by ULA Vulcan launch data, which shares the BE-4 engine). Ultimately, every launch failure is different, and until the investigation is completed, the space industry will have to wait.