How jet lag could potentially be reduced using machine learning

How jet lag could potentially be reduced using machine learning

How jet lag could potentially be reduced using machine learning

Affiliation Dassault Aviation Hackathon

Project Date 2017

Role Project Lead / Designer

Collaborators Mainak Jas Ramsi Ferkous

icon
This weekend project was done during the Boost the Falcon Experience Hackathon by Dassault Aviation.

image

Context

Jet lag has deleterious effects on health and can even impair cognitive processes in the long run [6]. Studies have shown that getting exposed to varying degrees of light at the right time before and after the flight can minimize its effects [1, 3, 4].

Alodia is an intelligent system that learns the clients’ preferences and adjusts the lighting and temperature accordingly to suit their comfort level. Athletes, for example, sleep better at 18℃ or lower compared to average passengers due to their bodily needs [7]. Changing the light exposure has also been effective [2]. Boeing’s mood lighting system gradually changes the light ambiance between phases of the flight to ease the change in time zones [5]. This, however, is deployed in commercial aircraft and not customized to individual passengers. Alodia, on the other hand, extracts patterns from the clients’ data and learns a decision tree. It is targeted towards VIPs flying on business where the entire cabin could be customized to their needs.

Initial pitch

Solution

Alodia first learns the passengers through natural conversations with them. One solution is to integrate it to Messenger but this may well be any other chat platforms already available or coded from scratch. The machine learning algorithm learns a decision tree from this data. The parameters of which are time zones crossed, age, travel direction, and sleep quality1. If passengers didn't have a good night sleep prior to the flight, Alodia could tailor the cabin ready for resting while it gradually acclimatizes them as they near their destination. It then spits out the color value that should be set: warm to help them drowse off and cool to wake them up. Alodia could eventually remind passengers when to take their meals, to exercise, and to rehydrate and could also adjust the temperature level. The more it converses with the clients, the better its predictions become.

Prototype

While the prototype is very rough, it’s a good start to test out how machine learning could potentially disrupt business aviation by providing clients with exacting taste a more personalized experience.

Tools

Recast AI , Python, HTML, CSS, Code on Github

Footnote

  1. We limited the model to only consider these parameters due to time constraints.

References

  1. Eastman, C. & Burgess, H. How to Travel the World without Jet Lag. Sleep Medicine Clinics. 2009 June. https://www.ncbi.nlm.nih.gov/pmc/articles/PMC2829880/
  2. Entrain Yourself. http://entrain.math.lsa.umich.edu/
  3. Forger, D. & Serkh, K. Optimal Schedules of Light Exposure for Rapidly Correcting Circadian Misalignment. PLOS Computational Biology. 2014 April. https://doi.org/10.1371/journal.pcbi.1003523
  4. Gray, R. How Flying Seriously Messes with your Mind. http://www.bbc.com/future/story/20170919-how-flying-seriously-messes-with-your-mind
  5. Kundu, S. Combatting Jet Lag with All Colors of the Rainbow. https://www.forbes.com/sites/sujatakundu/2016/08/31/manipulating-your-melanopsin-with-mood-lighting-combatting-jet-lag-with-all-colours-of-the-rainbow/#7e9c930149be
  6. The 2017 Nobel Prize in Physiology or Medicine - Press Release". Nobelprize.org. Nobel Media AB 2014. Web. 4 Oct 2017. http://www.nobelprize.org/nobel_prizes/medicine/laureates/2017/press.html
  7. Winter, C. Choosing the Best Temperature for Sleep. http://www.huffingtonpost.com/dr-christopher-winter/best-temperature-for-sleep_b_3705049.html