Does Boeing have a statute that commands that the windscreens are changed for new irrespective of no signs of damage or fatigue, and if so, what are the intervals, for safety reasons?


First, Boeing does not provide any statues for when maintenance is performed. They provide guidelines and recommendations through the Maintenance Planning Data (MPD) manual that most airlines follow. Some of these recommendations are required to be performed by the FAA with the certification of the aircraft or later AD action.

That said, there are no recommended maintenance tasks for the 787 windshields. The reasons is windshields are certified to be able to to hold cabin pressurization even with every pane shattered. Although there is no recommended or required task to replace the windshields at a set interval, some airlines may choose to do so for operational reasons.

Mandatory part replacements based on flight hours or cycles is almost exclusively reserved for a special classification of hardware known as "Safe Life Parts" these are generally just the Landing Gear and Engine Rotors. These are required to be scraped at specific intervals due their failure mechanisms and because they can be single point catastrophic failures.

I have specifically reviewed whether or not hard time replacement of 737/757/767 and 777 windshields made sense at a very large airline. The data showed that windshield failure rates when plotted against either cycles or hours had very large standard deviations. This means it is very hard to set an appropriate hard time interval, if you set it too early you would have few failures in the field but you replaced a lot of windows that had a ton of life left on them, if you set it too late you still had many field failures and were still throwing away good windows. When doing a Wiebull analysis I was able to identify at least 8 different failure mechanisms that all had very different rates and only affected a small population of windshields. To highlight the point, we had multiple windshields that failed after less than 100 flight hours while we had many that lasted more than 60,000 flight hours (and a few that were up to 80,000 that had never been replaced).

| improve this answer | |
  • $\begingroup$ Sometimes a person with exactly the right knowledge comes along to answer a question. Thanks! $\endgroup$ – Zach Lipton Feb 23 '19 at 8:34
  • $\begingroup$ Safe Life applies to all kinds of Line Replaceable Units, generally those with a structural element in their use, like flap actuators. The life limit is usually on the root structural item that defines the LRU's serial number (like a housing), and set based on threshold crack formation during cert fatigue testing. Windshields don't go through that kind of fatigue test process and lamination failures are considered random. The result is they are maintained on condition. Units that don't crack crap out from loss of optical quality from erosion of outer glass or failures of the heating system. $\endgroup$ – John K Feb 23 '19 at 20:33
  • $\begingroup$ @JohnK that is true about windshields, but at least on Boeing Aircraft, the only Safe Life parts tend to be landing gear and engine rotors with very few exceptions. The 787 currently has some systems safe life parts, but Boeing is working to remove them with further testing/re-design. $\endgroup$ – OSUZorba Feb 23 '19 at 22:19
  • $\begingroup$ Thank you for your interesting & informative answer. As a consequence and being specific, are you saying that the Mexico - UK flight could have safely continued it's journey, with just one outer cracked yet intact windscreen, instead of aborting after approx an hour and returning to base, which resulted in a 36 hour delay ? $\endgroup$ – Peter Lynch Feb 24 '19 at 3:15
  • $\begingroup$ Delay was due to a substitute plane being dispatched from the UK along with a replacement screen and engineers. If that is correct, does it not follow that the operator cannot escape liability for delay compensation under Regulations in Europe EC 261 on the grounds that the delay was beyond their control, given the delay was caused by the deliberate decision of the pilot aborting which now appears to not have been necessary ? $\endgroup$ – Peter Lynch Feb 24 '19 at 3:21

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.