Display of all traces throughout the scenario. Most occupants in the right central corridor did not escape the building.
Edward Swing, Vision Systems & Technology, Inc.
Student Team: NO
The Prajna Project is a Java toolkit designed to provide various capabilities for visualization, knowledge representation, geographic displays, semantic reasoning, and data fusion. Rather than attempt to recreate the significant capabilities provided in other tools, Prajna instead provides software bridges to incorporate other toolkits where appropriate. Prajna will be released to the Open Source community in the near future.
For this challenge, I developed custom readers to read the traces and RFID assignments. I used the Prajna grid utilities to read and display the building data, while the trace data was stored as a collection of dynamic tracks over grid coordinates. Custom user interface components provided the scenario-specific controls. These controls allowed playing the scenario in an animated fashion, toggling the paths of the building occupants, and querying a particular path or person.
In addition, the Prajna software bridge to JFreeCharts provided the charting capability for statistical displays. JFreeChart is an open source project that provides a Java toolkit for displaying graphical charts. This charting capability was used to display the cumulative movement of the various individuals.
VSTI is evaluating various elements of the custom application designed for this challenge for use in immersive building visualizations.
The Prajna Project is a toolkit developed by Edward Swing. The custom application was built at VSTI. Other VSTI programs have since incorporated some of the new components that were developed for this contest.
66x29
By displaying the routes of the evacuees, we notice that personel on the left side of the building evacuated normally. However, on the right side of the building, there were a number of individuals who did not escape the building by the end of the scenario.
Display of all traces throughout the scenario. Most occupants in the right central corridor did not escape the building.
Upon examination of the course followed by these individuals, we notice that those near the central horizontal corridor stopped moving fairly soon after the explosion. Their movement traces did not enter into the central corridor, suggesting that the individuals in that area were trapped, injured or killed in the explosion. Therefore, we concluded the device was detonated in this corridor.
Ramon Katalanow pauses to set the bomb. His location, and the location of the bomb, is highlighted with a red hand-drawn overlay
Viewing the activities of the prime suspect, we notice he paused for a short time at 66x29.5. Since this corresponded to a location where nearby individuals did not exit following the explosion, we believe this location is where he set the device.
Suspects: 21, 28, 29, 56
Direct Witnesses: 18, 21, 50, 56
Other Witnesses: 1, 27, 76
Since any suspects would likely have foreknowledge of the event, we examine individuals who were moving prior to the explosion. We used the Dynamic Track Display and the Cumulative Distance Chart to identify five suspects who had significant movements prior to the event. Of these, ID 21 moved through the area of the explosion prior to the event, so he is our prime suspect. We discounted 44, who did not show suspicious behavior.
Chart of Cumulative Distance Traveled for all individuals. Suspects were identified as those moving before the explosion.
Suspects prior to blast. Their locations are indicated by a red overlay.
We also used the Dynamic Track Display to identify witnesses. Direct witnesses include everyone who had a line of sight to the detonation, while other witnesses may have witnessed suspicious behavior. 18, 21, 50, and 56 had a line of sight to the explosion. 76 was an office-mate of 21, thus she might have witnessed unusual activity. While evacuating, 27 briefly turned toward 21; 21 then moved away from 27. 1 also witnessed 21's escape route.
Suspects: 21, 28, 29, 44
Direct Witnesses: 21
Other Witnesses: 1, 27
We have already identified the suspects in this event. Using the Dynamic Trace Display, we see that 21, 28, 29, and 44 all escaped the building successfully. 56 did not.
Location of individuals at the time of the explosion, with the bomb location highlighted in red. Track of Cleveland Jimenez (RFID 56) shows that he failed to exit the building
As far as other witnesses to the event, RFIDs 18 and 50 had a direct line of site at the time of the explosion. However, they may not have survived, as they were still inside the building at the end of the scenario. 1 and 27 were not directly affected by the explosion, and able to escape.
18, 19, 39, 47, 50, 56, 59, 60, 65, 69, 76, 78
Using the Cumulative Distance Chart, I reviewed those tracks which did not move significantly after the detonation. These individuals, visible as a horizontal line, were fairly easy to identify. I then cross-checked the locations of these individuals and determined which were still inside the building at the end of the scenario. This identified the first group of casualties.
Chart of Cumulative Distance Moved, zoomed in to show tracks which were mostly stationary after the blast.
While viewing the locations of the individuals still inside at the end of the scenario, I noticed a second group of casualties. These employees attempted to exit the building in the upper right corner of the display. These individuals were active through much of the evacuation, but became stationary near the building exit. Either they were caught in a secondary collapse or trapped inside. A final casualty (RFID 59) occurred at the bottom of the building. Her movements seemed to indicate panic or confusion.
Locations of casualties at the end of the scenario. Casualties highlighted in red.
By using the trace of personnel movements and the chart of their cumulative distance travelled, we note that the bomb exploded around T=375. Prior to this time, the vast majority of the building occupants were stationary. Following the explosion, the occupants began to move toward the nearest exit.
Certain individuals were moving prior to the event. These individuals were our primary suspects, as someone with fore-knowledge of the event would likely be trying to escape before the bomb detonated. By observing the movement patterns of these individuals with the Dynamic Trace Display, we attempted to identify what role those individuals played in the scenario. We notice several distinct behaviors. These individuals, and our observations, are as follows:
Cleveland Jimenez stops at the exact same point as Maxwell Lopez did earlier. Cecil Denison (trace visible) wanders by later, then leaves.
One curious behavior we noticed in our traces is that both Maxwell Lopez and Cleveland Jimenez stopped at 32.48,33 at a wall location. Furthermore, Cecil Denison walked by that point, then reversed his movements. Maxwell Lopez may have left a message or signal that Denison and Jimenez reacted to. It would benefit the investigators to find out what was at that location.
Again, we use the Dynamic Trace Display to monitor the movements of all occupants. Once the detonation occurred, we see that the employees on the left side of the building evacuated in a relatively orderly fashion. Most of the occupants either near the top or bottom of the building on the right side also evacuated without incident. A few individuals seemed to behave in an erratic fashion, possibly panicking briefly. A few others paused before exiting, possibly collecting personal items. We then examined the movements of those individuals still in the building at the end of the scenario in an effort to determine their ultimate fate.
Individuals trapped or killed in offices near the explosion.
Several individuals in the rooms near the explosion began to move after the blast. However, they did not escape the building. These individuals are Gale Welsh (RFID 18), Max Valdez (RFID 19), Lottie Staley (RFID 50), and Fawn Sparks (RFID 76). Cleveland Jimenez (RFID 56) was also caught in this area.
Lavon Lockhart (trace visible) turns back from exit, possibly to help other evacuees.
A second group of individuals were moving towards the exit at the upper right of the display. These individuals seemed to be following another group, but were unable to exit the building. One of them actually reached the exit (Lavon Lockheart, RFID 60), but then turned back, possibly to help other employees. The movements of a few of this group, Dian Crum (RFID 65) and Morton Kilgore (RFID 69) suggest the path may have been blocked. Phil Marin (RFID 39), Rosario Oakley (RFID 47) and Cleveland Hutchison (RFID 78) were also among this group of casualties.
Olive Palmer (trace visible, location indicated in red) turns away from the exit during the evacuation, possibly in a panic.
Another employee, Olive Palmer (RFID 59) inexplicably turned away from the exit at the lower left corner of the display. Her movements seemed erratic, so she might have been in a panic. Two other employees follow her briefly, then return to the exit.
At this point, it is unclear whether these casualties were trapped, injured, or dead.
While observing the movements of our prime suspect, we noticed another interesting interaction. After witnessing the explosion, Ramon Katalanow moved toward the bottom right corner of the building. He paused again at 73.75,24.75 until he saw Anton Knapp (RFID 27), or vice-versa. Katalanow and Knapp then moved directly away from one another to different exits.
Anton Knapp (trace visible) turns towards Ramon Katalanow (circled in red) during the evacuation. Katalanow then moves away from Knapp to other exit.
We noticed one other person of interest. Francisco Salter (RFID 36) remained almost completely stationary during the entire scenario. There are several possible explanations for this. Salter may have left his badge at his desk, either intentionally or accidentally. The tracking system may not have been working for his RFID. Less likely was that Salter was incapacitated prior to the scenario. Abel Snow (RFID 73) moved past Salter during the evacuation, so he might be able to provide information about Salter. It might be important to learn what Salter's role in the company was - if he were a security officer who was supposed to monitor the building, his actual location and fate may be very important.