veyepar: Aircraft predictive maintenance using Python/ML

Author(s):	Amar Verma
Location	Auditorium 1
Date	aug Sun 25	Days Raw Files
Start	14:20	First Raw Start	13:49
Duration	00:50:00	Offset	0:30:36
End	15:10	Last Raw End	15:19
Chapters	00:00 0:25:39 0:27:02
Total cuts_time	37 min.

raw-playlist raw-mp4-playlist encoded-files-playlist host tweet mp4 svg png

assets release.pdf aircraft_predictive_maintenance_using_pythonml.json
logs
Admin: episode episode list cut list raw files day marks day marks day image_files

State:

Locked:

clear this to unlock

Locked by:

user/process that locked.

Start:

initially scheduled time from master, adjusted to match reality

Duration:

length in hh:mm:ss

Name:

Video Title (shows in video search results)

Emails:

email(s) of the presenter(s)

Released:

has someone authorised pubication

Normalise:

Channelcopy:

m=mono, 01=copy left to right, 10=right to left, 00=ignore.

Thumbnail:

filename.png

Description:

markdown

Applied machine learning for predictive maintenance (PdM) with objectives to reduce aircraft downtime & in-workshop costs

"Aircraft engines must be serviced, overhauled and examined on a very regular basis. This is driven by several factors, such as the total flying hours, on-ground taxi time, time from last overhaul, parts with limited life cycle, FAA bulletins & advisories, as well as for issues reported by sensors, ground crew & flight deck.

In case of more than one engines on the aircraft, most possibly the engines are non-coordinated in their maintenance requirements. This is very obvious as the engines could be maintained, installed on separate ocassions.

Whenever an aircraft engine is needed to be serviced, it causes direct impact to cashflows due to some of these factors:
* plane need to be grounded if alternative engine not available
* engine maintenance is delayed due to 
* parts not available
* specifically skilled engineer not available
* workshop space not available
* Aircrafts (generally) use more than one engine, so the negative impact is amplified
* Aircraft's flying schedule can be impacted causing rebookings, schedule conflicts and direct impact to cashflows

Our solution addressed this challenging problem by combining & deriving few different algorithms for different parts of the problem:
- determine extent of the engine deterioration using a simplified variant of Kalman filtering method
- predict the optimal scheduling window using adaptation of genetic algorithm

Innovation & Engineering hacker. Expert in AI/ML, Algorithms & Big Databases

Comment:

production notes