row1
Abstract
Public schools in the United States are mandated by State legislatures to conduct periodic school bus evacuation training for the benefit of students. The approach to this training varies widely among school systems and there is rarely documentation describing the manner of testing. Typically, there is only a record that the training itself was conducted. Radio-frequency identification (RFID) is becoming more prevalent in many applications related to the safe transport of children on school buses. The purpose of this study was to explore the use of RFID to measure evacuation times compared to video and/or stopwatch measurements, for thirty-four passengers on a Type “D” school bus. The results obtained over three evacuation scenarios (front door only; rear door only; both doors simultaneously) revealed no statistically significant difference between those times recorded by RFID and those observed by video analysis for all trials. Based on these findings, RFID has the potential to provide a fast, cheap, non-invasive way to record school bus evacuation times, eliminating lengthy video analysis. Elimination of stopwatch and/or videotape analysis could lead to numerous schools uploading evacuation training times into a central repository, building a reliable source for benchmarking and potential design changes for emergency exit systems.
Summary
Periodic school bus evacuation trainings are conducted all over U.S. as mandated by Public School systems for safety. The approach, however, varies widely across school systems and the potential of utilizing RFID passive tags for measuring evacuation times and maintaining a central repository. The objectives of this experiment were to compare video analysis and RFID passive tags by measuring the evacuation times and determining the flowrate(s) (passengers per minute) of adult subjects through different door configurations. The various configurations include front door only, rear door only and both doors simultaneously.
A type D school bus was utilized for the experiment and Auburn University Engineering students were chosen as participants along with a pre-event brief to cover the purpose of the study. RFID readers Zebra FX7500-42325A50-WR) and antennas (Alien ALR 8697) were placed close to each door and power was adjusted to recognize the RFID tag as close as possible to the participant completely off the bus. Video cameras were placed as well for validation purposes. Passive RFID tags namely, Smartrac Dogbone RFID (MONZA R6-P) Wet Inlay were adhered to name badges on lanyards worn around participant necks chosen randomly.
A 61% accuracy was achieved for RFID times versus video analysis and for 21% the tags recognized the evacuees faster than video analysis. Potential solutions to eliminate the errors can be shifting the position of the readers and utilizing more than 1 tag to read RFID times. Also, other items such as backpacks or loose clothing and other items, even a single piece of paper can affect RFID readings.
To the authors’ knowledge, this is the first study that attempted utilizing RFID tags and video analysis and reliable accuracy was achieved. The purpose of this study is to help build a reliable standardized source of data rather than stopwatch and video analysis for benchmarking and potential design changes for emergency exit systems. Future studies can include the limitations of this study including larger sample sizes, using age-appropriate participants, full capacity of school bus, and simulating the intensity of an actual emergency evacuation.
