MIT Convective Weather Avoidance Study

Thank you for your interest in our study! More information about the study can be found below. If, after reviewing the following information, you would like to participate, please answer the questions at the bottom of the form and submit. 

Sign in to Google to save your progress. Learn more

The end goal of this work is the development of better predictive algorithms for en-route convective weather avoidance. Development of these models relies on labeled historical aircraft weather encounters. This study is being conducted to develop an algorithm for recognizing those behaviors in previous weather encounters.
Informed Consent

Participation: We expect that participation will take about 15-30 minutes, and is done entirely remotely and asynchronously (we will not need to schedule any meetings). Participation consists of categorizing images, such as the images above, of previous aircraft weather encounters as non-deviations, deviations due to weather, or deviations for other reasons.

Methods: Current weather avoidance models are based on small sets of labeled weather encounters. We seek to improve these models by leveraging machine learning models to generate larger labeled datasets. At this time, we are planning to use an artificial neural network to process our labeled data. The expert-labeled encounters collected during this study will be used to train models to label significantly larger sets of encounters. This larger set of labeled encounters will be used to build more robust weather avoidance models.

Benefits: Successful development of a predictive convective weather avoidance model may result in improved decision support tools for air traffic controllers and other personnel involved with the routing of aircraft. These tools may help to reduce flight plan deviations due to convective weather as well as delays associated with enroute convective weather. 

Privacy: The personal information that is being collected are the email addresses, the types of aircraft certified for, and the number of flying hours for each labeler. Each participant will be assigned a randomized alphanumeric code to maintain anonymity during data analysis. These assignments are kept on a separate hard drive from the data analysis.  The labels themselves will be used only in aggregate. We will not release or publish individual data for email addresses, type certificates, or flying hours, or any other personal information that is provided to us. 

Participation in this study in voluntary, and there are no adverse consequences for choosing not to participate. If you choose to participate, you may decline to answer any or all questions. You may withdraw from participation at any time, with no adverse consequences. 

If you have any other questions about this study, please contact 

I am interested in participating in the study *
Clear form
Never submit passwords through Google Forms.
This content is neither created nor endorsed by Google. Report Abuse - Terms of Service - Privacy Policy