Vocavio has just concluded a project with newly founded data science company, B&P Consulting based in Ontario, Canada. The project brief was initially focused on mapping OBs (observable behaviours) annotated by flight instructors to signal data generated by Vocavio’s voice analysis. This sprint project quickly became about the sample size available, what models might support such analysis and what else can the data share with us?
In summary, here’s what the data did tell, and some insights might just surprise you.
- Top instructors observed 82% of what the voice analytics sensor observed in relation to workload during a 20 min LOFT exercise. Average instructors observed 40% and lower yield was at 16%.
- Lag between instructors observing and inputting on e-grading is to be expected, averaging at 46seconds. The extreme lag was at 160 seconds between behaviours observed and data annotated into the e-grading solution. This lag impacts on model fit by reducing the validity of the instructor signal tied to more ambiguous workload/stress events.
Women in Aviation Training (WIAT) at EATS 2023:
Talent emerges in many places. For Product Director at Vocavio, Conor McKenna, it was at WATS in Orlando, Florida, earlier this year where he listened to an all-female panel. One speaker, Naila Ayala, is an emerging talent in data science. She shared some valuable research on the role of eye-tracking in solo pilot flight training from a project at WISA.
Ayala has since incorporated as B&P Consulting Inc and Vocavio was interested in determining if data models used in eye tracking would translate to signal data from voice analysis, another biosignal capability that can add value in pilot training.
“Where data models have more input from instructors, we can build better trust in new bio-signal technologies like voice analysis and eye tracking to further augment pilot training and debriefing of crew,” says Naila Ayala, Cognitive Neuroscientist, Data Scientist, and PhD Candidate.
Using a small dataset, several insights were achieved around correlation of instructor OB data and signal data, and importantly strong levels of data model fit.
“This short project with Naila exceeded my expectations. We have known that mapping OBs to signal data isn’t without its challenges, but the early work is showing significant promise and insights for the pilot training audience,” said Conor McKenna. “And yes, we are delighted that Halldale is expanding its ‘women in aviation’ agenda, big problems need diverse teams, comfort with data experimentation and alternative perspectives on potential solutions.”
This sprint project demonstrated the value of collaboration and sprint projects, where budgets are accessible, fast to scope and can be delivered in a timely manner.
To find out more about women in aviation training and to join the WIAT forum, click here.