This is a research prototype. The data and analyses are preliminary and not yet validated — we'd welcome your .
During a severe winter storm in December 2022, Southwest Airlines' crew scheduling software, SkySolver, failed to effectively reassign pilots and flight attendants. This failure resulted in massive operational disruptions, including the cancellation of over 13,000 flights and the stranding of thousands of passengers. The incident highlighted the airline's reliance on outdated or insufficient technology to manage its complex flight network.
Southwest Airlines left passengers stranded for days throughout the flight network when Southwest crew scheduling software repeatedly failed to recover from weather-induced flight cancellations.
Domain classification, causal taxonomy, severity scores, and national security assessments were LLM-classified and may contain errors.
AI systems that fail to perform reliably or effectively under varying conditions, exposing them to errors and failures that can have significant consequences, especially in critical applications or areas that require moral reasoning.
The incident was caused by SkySolver's lack of capability and robustness when handling a large-scale disruption, leading to system failure and illogical outputs.
AI system
Due to a decision or action made by an AI system
Unintentional
Due to an unexpected outcome from pursuing a goal
Post-deployment
Occurring after the AI model has been trained and deployed
The operational failure was caused by the SkySolver AI system failing to perform its scheduling function under extreme conditions after being deployed.
Added categories peak at 2: Minor (Psychological Harm, Indirect)
The December 2022 Southwest Airlines operational meltdown was caused by the failure of its SkySolver crew scheduling software during a severe winter storm. While the incident caused massive commercial disruption, financial loss, and affected over a million travelers, it was an unintentional legacy software failure with minor national security implications, primarily highlighting vulnerabilities in critical infrastructure IT resilience.
The report describes SkySolver as a system that suggests crew assignments and reassignments, which are then used by crew schedulers, acting as an assistant to human decision-makers.
High Risk: The system is used in the management and operation of critical transport infrastructure (commercial aviation), where its failure led to a nationwide transportation shutdown.
The primary function of SkySolver is to schedule and reassign flight crews (pilots and flight attendants) to flights, which is a core scheduling and workforce administration task.
The reports provide clear, detailed accounts of the operational failure of Southwest's SkySolver system during the December 2022 storm, including the consequences (13,000+ cancellations, financial impacts, and crew scheduling issues). The role of the AI system is explicitly discussed by airline executives and union leaders.
The incident could be interpreted primarily as a human management and infrastructure failure (failing to upgrade IT systems) rather than an inherent AI safety failure, though the software's specific failure to generate logical solutions under stress is a key driver.
Technology
SkySolver System Overwhelm
The crew scheduling software was overwhelmed by the scale of disruptions.
Outdated Scheduling Software
SkySolver was nearing end of life and lacked modern capabilities.
Flawed Algorithmic Solutions
The system produced illogical crew assignments like circuitous deadheading.
Data Inputs
Inability to Track Crew Location
The system lost track of where crew members were physically located.
Jammed Communication Channels
Jammed phone lines prevented crew from inputting availability and status.
Human Factors
Overwhelmed Crew Schedulers
Schedulers had to manually process matches when the automated system failed.
Communication Bottlenecks
Crews spent hours on hold trying to report locations and get assignments.
Process and Methods
Point-to-Point Network Structure
Unlike hub-and-spoke, the daisy-chain network propagated delays nationwide.
Manual Fallback Failure
Manual scheduling processes were too slow to handle the volume of errors.
Management
Growth Outpacing Technology
Management expanded airline operations faster than supporting IT infrastructure.
Underinvestment in Back-End IT
Prioritized consumer-facing tools over critical operational scheduling systems.
Delayed System Modernization
Management was slow to update critical IT systems despite historical warnings.
AI system involved
Identified by automated extraction; can differ from the alleged developers above.
