According to the New York Times, there have been numerous close plane incidents in the US over the past 12 months.
Flying has long been considered one of the safest modes of transportation. Researcher Arnold Barnett states that if an individual takes one flight per day, it would take approximately 55,000 years of daily travel before encountering a dangerous crash, as reported by ABC News.
Despite this reassuring statistic, recent events have revealed a troubling trend in aviation safety.
In July, a Southwest Airlines pilot narrowly averted a potential disaster, and just nine days later, multiple jet planes were reported to be in extremely close proximity in San Francisco. Within that same month, there were at least 46 reported incidents of near misses with commercial airlines.
During the Federal Aviation Association (FAA) Summit in March of the previous year, U.S. Secretary of Transportation Pete Buttigieg emphasized the importance of proactive safety measures, stating, “We cannot wait for the next catastrophic event to recognize the warning signs of today, identify the contributing factors, and promptly address them.”
This proactive approach is mirrored in the efforts of graduate students from the USC Viterbi School of Engineering.
At the 2023 FAA Data Challenge in Washington, D.C., a team of USC students presented their research findings.
As part of their Applied Data Science master’s program, Vivian Lin, Andrew Bruneel, and Tiffany Hoi Ching Wong delved into the realm of aviation safety after learning about the FAA Data Challenge a year prior. Guided by Yolanda Gil, a research professor of computer science at USC Viterbi, the students embarked on a journey to leverage AI solutions for aviation-related problems. Their focus was on addressing airport safety issues, particularly runway incursions and excursions, under the mentorship of Professor Najmedin Meshkati, an expert in safety culture and aviation safety.
The students encountered challenges in accessing relevant data, eventually utilizing the publicly available Aviation Safety Reporting System (ASRS) supported by NASA. Through their analysis, they identified key factors contributing to runway incidents, emphasizing the importance of effective communication, adherence to protocols, and pilot proficiency in ensuring airport safety.
Their research underscored the need for improved data standardization and policy enhancements within the aviation industry to mitigate risks effectively. By proposing machine learning models to evaluate safety factors, the team aimed to drive meaningful changes in airport safety practices.
Although their project did not secure the top prize, the USC team’s efforts were commended for shedding light on critical data collection and reporting issues in aviation safety. Their innovative approach and dedication to enhancing safety protocols exemplified a proactive stance towards addressing industry challenges.
In a field dominated by technological advancements, the USC students’ foray into applied data science not only broadened their knowledge but also highlighted the importance of data integrity and quality assurance in driving impactful solutions.
Furthermore, USC’s participation in the FAA Data Challenge 2023 showcased the university’s commitment to fostering innovation and addressing real-world aviation challenges through multidisciplinary collaboration.
The pursuit of excellence in aviation safety remains a continuous journey, with each research endeavor contributing to a safer and more efficient air travel landscape.
