System and method for predicting emergency evacuation movement in facility
By utilizing access control systems and AI models to analyze employees' historical movement patterns, evacuation routes and time predictions are optimized, addressing the shortcomings of traditional emergency evacuation plans and achieving more efficient and accurate emergency evacuation management.
Patent Information
- Application Number
- CN202510507924.9
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Priority Date
- 2024-04-23
- Filing Date
- 2025-04-22
- Publication Date
- 2025-10-24
AI Technical Summary
Traditional emergency evacuation plans and training methods lack realism, fail to consider employee behavior and availability, leading to prolonged evacuation times, and disrupting normal work during real-time drills. They also lack effective means of emergency evacuation prediction and optimization.
By capturing access control data using the facility's access control system, analyzing employees' historical movement patterns through artificial intelligence models, predicting and optimizing evacuation routes, and continuously improving the model based on feedback from evacuation drills, evacuation time can be dynamically adjusted and optimized.
It improves the accuracy and efficiency of emergency evacuation, reduces evacuation time, minimizes disruption to normal work, and provides more targeted emergency evacuation plans.
Smart Images
Figure CN120833072A_ABST
Abstract
Description
TECHNICAL FIELD
[0001] The present disclosure relates generally to emergency evacuations of facilities, and more specifically to methods and systems for improving emergency evacuation procedures of facilities. BACKGROUND
[0002] Emergency evacuations can occur at any time and are critical to ensuring the safety of facility occupants in emergency situations. Conducting frequent fire drills, while beneficial to safety, can disrupt the workday of employees and erode their confidence. Traditional emergency evacuation training methods lack realism and fail to incorporate critical access control data, resulting in suboptimal evacuation plans and longer evacuation times. In one example, envision a scenario in which an evacuation plan for an office facility does not account for how employees typically move throughout the facility. This oversight can result in inefficient evacuation routes, leading to increased evacuation times in a real emergency, potentially endangering lives.
[0003] Traditional emergency evacuation planning and training methods typically rely on static evacuation plans and real-time drills, which can not accurately simulate real-world scenarios if not planned with expected user behavior and availability. Furthermore, conducting real-time drills can cause disruptions in the regular operations of a facility, inconveniencing employees or occupants of the facility and potentially impacting productivity or daily activities. It would be desirable for systems and methods to improve evacuation plans for a facility to reduce evacuation times, avoid overcrowding during real-time drills, account for holidays, personnel availability, and exit door status, and reduce the impact on the daily workflow of occupants of the facility. SUMMARY
[0004] The present disclosure relates generally to emergency evacuations of facilities, and more specifically to methods and systems for improving emergency evacuation procedures of facilities. In some cases, the present disclosure provides a solution that utilizes access control data captured by an access control system of a facility. This access control information can be used to model the behavior of occupants in a facility under various circumstances, such as under an evacuation event, to create reliable emergency evacuation visualizations that can be used to improve emergency evacuation procedures for evacuating the facility.
[0005] An example can reside in a method for predicting an emergency evacuation time for a facility. The illustrative method includes capturing location information for each of a plurality of people associated with the facility in the facility, and identifying one or more historical movement patterns for one or more of the plurality of people associated with the facility based at least in part on the captured location information. One or more predefined evacuation routes for the facility are identified. These predefined evacuation routes can be automatically generated or manually defined. The method includes predicting the emergency evacuation time for evacuating the facility using the one or more predefined evacuation routes in response to an evacuation event based at least in part on the one or more historical movement patterns for one or more of the plurality of people associated with the facility. In some cases, the predefined evacuation routes can be modified, resulting in one or more modified evacuation routes. Then, based at least in part on historical behavior patterns, the method can predict an evacuation time for evacuating the facility using one or more modified evacuation routes. This can be repeated until a desired predicted evacuation time is reached.
[0006] In some cases, an evacuation drill can be initiated using an evacuation route predicted to achieve a desired evacuation time. The location information for each of one or more people in the facility during the evacuation drill can be captured using an access control system for the facility. The location information captured during the evacuation drill can be used as feedback to enhance the model, resulting in more accurate subsequent predictions of an evacuation time for evacuating the facility.
[0007] In some cases, the location information for each of the plurality of people in the facility can continue to be captured over time, resulting in updated location information. Updated historical behavior patterns for people in the facility can be determined based at least in part on the updated historical behavior patterns. From this, an updated evacuation time for evacuating the facility using one or more evacuation routes can be predicted. If the updated evacuation time does not satisfy a desired predicted evacuation time, different evacuation routes can be defined, and a corresponding evacuation time can be predicted for each subsequent evacuation route using the updated historical behavior patterns until the desired predicted evacuation time is reached.
[0008] Another example can reside in an emergency evacuation prediction system for a facility. The emergency evacuation prediction facility includes an access control system to capture location information for each of a plurality of people associated with the facility, and a controller operably coupled to the access control system. The controller is configured to identify one or more historical movement patterns for one or more of the plurality of people associated with the facility based at least in part on the captured location information, and predict an emergency evacuation time for evacuating the facility using one or more predefined evacuation routes based at least in part on the one or more historical movement patterns for one or more of the plurality of people associated with the facility.
[0009] Another example can exist in a non-transitory computer-readable medium having stored thereon instructions that, when executed by one or more processors, cause the one or more processors to receive location information for each of a plurality of people associated with a facility in the facility, and identify one or more historical movement patterns of one or more of the plurality of people associated with the facility based at least in part on the captured location information. The one or more processors are caused to predict an emergency evacuation time for evacuating the facility using one or more predefined evacuation routes based at least in part on the one or more historical movement patterns of the one or more of the plurality of people associated with the facility.
[0010] The foregoing summary was provided to illustrate some of the innovative aspects of the disclosure and is not intended to be a complete description. A complete appreciation of the disclosure can be gained by considering the entire specification, claims, drawings and abstract. BRIEF DESCRIPTION OF DRAWINGS
[0011] The various examples described can be better understood with reference to the following description together with the accompanying drawings, in which:
[0012] Figure 1 is a schematic block diagram illustrating an exemplary emergency evacuation prediction system that can form part of a building management system;
[0013] Figure 2 is a flow diagram illustrating an exemplary method for predicting an emergency evacuation time for a facility;
[0014] Figure 3 is a flow diagram illustrating an exemplary method for predicting an emergency evacuation time for a facility;
[0015] Figure 4 is a flow diagram illustrating a series of exemplary steps that can be performed by one or more processors executing executable instructions; and
[0016] Figure 5 is a flow diagram providing an exemplary overview.
[0017] While the disclosure is susceptible to various modifications and alternative forms, specific details have been shown by way of example in the drawings and will be described in detail herein. It should be understood, however, that the intention is not to limit the disclosure to the particular example described. On the contrary, the intention is to cover all modifications, equivalents, and alternatives falling within the spirit and scope of the disclosure. DETAILED DESCRIPTION
[0018] The following description should be read with reference to the accompanying drawings, which are drawn in scale wherein like numerals represent like elements throughout the several figures. The drawings, which are not necessarily to scale, depict examples that are illustrative of the disclosure and are not intended to limit the scope of the disclosure. While examples of various elements are illustrated, one skilled in the art will recognize that many of the examples provided have suitable alternatives that can be utilized.
[0019] It is assumed herein that all numbers are modified in all instances by the term "about" unless otherwise specifically noted. The expression of a range of numerical values as an endpoint includes all numbers between the stated range (e.g., 1 to 5 includes 1, 1.5, 2, 2.75, 3, 3.80, 4, and 5).
[0020] As used in this specification and the appended claims, the singular forms "a," "an," and "the" include plural referents unless the content clearly dictates otherwise. As used in this specification and the appended claims, the term "or" is generally employed in its sense of "and / or" unless the content clearly dictates otherwise.
[0021] It should be noted that recitations herein of "one embodiment," "some embodiments,” "other embodiments” or the like do not mean that a feature only in a single embodiment is new. Also, these phrases are not necessarily referring to the same embodiment. In addition, when a particular feature, structure, or characteristic is described in connection with an embodiment, it is submitted that it is within the purview of one skilled in the art to effect such feature, structure, or characteristic in connection with other embodiments whether or not explicit
[0022] Figure 1 is a schematic block diagram illustrating an example emergency evacuation prediction system 10 of a facility. The emergency evacuation prediction system 10 can be part of a building management system, such as an access control system of the facility, although this is not required. In the illustrated example, the emergency evacuation prediction system 10 includes an access control system 12 configured to capture location information for each of a plurality of persons associated with the facility. In some cases, the location information for each of the plurality of persons associated with the facility can include access control data from the access control system 12. In some cases, the access control data can include one or more (or even all) of a personal identifier for each person passing through an access control point of the access control system 12, an access event time, and an access location.
[0023] In some cases, the access control system 12 may include multiple access card readers associated with each of the various doors distributed around the facility. As people use their access cards to enter and exit through specific doors and, therefore, move around the facility, the access control system 12 can track where at least most people are at any given time. In addition, the access control system 12 can provide historical data about people's movements, such as where they were at what time during previous time periods (such as the last hour, last day, last week, etc.). In some cases, the access cards may be physical access cards that are swiped on a card reader, while in other cases, the access cards may be electronic access cards, such as RFID tags, Bluetooth devices, mobile phones, etc., which can provide frictionless entry and exit through the facility's access control points. In some cases, the access control system 12 may include cameras and video analytics algorithms to identify and track people as they move around the facility. These are just examples.
[0024] The exemplary emergency evacuation prediction system 10 also includes a controller 14 operatively coupled to the access control system 12. The controller 14 is configured to identify one or more historical movement patterns of one or more individuals from among the plurality of individuals associated with the facility based at least in part on the captured location information, and to predict an emergency evacuation time for evacuating the facility using one or more predefined evacuation routes based at least in part on the one or more historical movement patterns of one or more individuals from among the plurality of individuals associated with the facility. In some cases, the controller 14 may be configured to identify which individuals from among the plurality of individuals associated with the facility are currently in the facility, and to predict an emergency evacuation time for evacuating the facility using one or more predefined evacuation routes based at least in part on the one or more historical movement patterns of the individuals identified as currently in the facility.
[0025] In some cases, the controller 14 includes an AI (artificial intelligence) model 16 for predicting emergency evacuation times. In some cases, the AI model 16 can be a trained AI model. In some cases, the controller 14 can be configured to use the AI model 16 to at least partially process the captured location information to identify a predicted emergency evacuation time based on historical location information and one or more predefined evacuation routes. In some cases, the controller 14 can be configured to train the AI model 16 to account for the number and location of people in the facility and crowd dynamics in the facility.
[0026] In some cases, the controller 14 can receive one or more modified evacuation routes from a user or can automatically derive one or more modified evacuation routes and predict an evacuation time for evacuating the facility using the one or more modified evacuation routes based at least in part on historical behavior patterns. In some cases, the controller 14 can automatically derive a plurality of modified evacuation routes and predict a corresponding evacuation time for each of the plurality of modified evacuation routes as part of a Monte Carlo simulation to determine an optimal modified evacuation route.
[0027] To help improve the accuracy of the model, an evacuation drill can be initiated. The controller 14 can capture location information for each of the plurality of people in the facility during the evacuation drill and use the location information captured during the evacuation drill as feedback to enhance the model to improve subsequent predictions of evacuation times for evacuating the facility. In some cases, the controller 14 can be configured to continue to capture location information for each of the plurality of people in the facility over time, resulting in updated location information. The controller 14 can be configured to identify updated historical behavior patterns of the people in the facility based at least in part on the updated location information and predict updated evacuation times for evacuating the facility using one or more evacuation routes based at least in part on the updated historical behavior patterns.
[0028] Figure 2 is a flowchart illustrating an exemplary method 18 for predicting an emergency evacuation time for a facility. The method 18 includes capturing location information for each of a plurality of people associated with the facility in the facility, as shown in block 20. In some cases, the location information for each of the plurality of people associated with the facility can include an entry time and an exit time at a preferred point of ingress and egress of the facility. In some cases, the location information can include access control data collected by an access control system of the facility. As an example, the access control data can include a personal identifier for each person passing through an access control point of the access control system, an access event time, and an access location. In some cases, capturing location information for each of the plurality of people associated with the facility can include capturing location information for two or more of the plurality of people associated with the facility during one or more prior evacuation events, and wherein at least one of the historical movement patterns can be based at least in part on the location information captured during the one or more prior evacuation events.
[0029] One or more historical movement patterns of one or more of the plurality of people associated with the facility are identified based at least in part on the captured location information, as indicated by block 22. One or more predefined evacuation routes of the facility are identified, as indicated by block 24. An emergency evacuation time for evacuating the facility in response to an evacuation event using the one or more predefined evacuation routes is predicted based at least in part on the one or more historical movement patterns of one or more of the plurality of people associated with the facility, as indicated by block 26.
[0030] In some cases, the method 18 can include identifying which of the plurality of people associated with the facility are currently in the facility, as indicated by block 28. The method 18 can include predicting an emergency evacuation time for evacuating the facility in response to an evacuation event using the one or more predefined evacuation routes based at least in part on one or more historical movement patterns of the people identified as currently in the facility, as indicated by block 30. In some cases, the method 18 can include determining one or more updated predefined evacuation routes of the facility by predicting an emergency evacuation time for each of a plurality of possible predefined evacuation routes, as indicated by block 32. In some cases, the method 18 can also include notifying one or more of the plurality of people associated with the facility of the one or more updated predefined evacuation routes of the facility, as indicated by block 34.
[0031] Figure 3 is a flowchart illustrating an example method 36 for predicting an emergency evacuation time of a facility. The method 36 includes capturing location information of each of a plurality of people associated with the facility in the facility, as indicated by block 38. In some cases, the location information of each of the plurality of people associated with the facility can include an entry time and an exit time at a preferred point of ingress and egress of the facility. In some cases, the location information can include ingress and egress control data collected by an ingress and egress control system of the facility. As an example, the ingress and egress control data can include a personal identifier of each person passing through an ingress and egress control point of the ingress and egress control system, an ingress and egress event time, and an ingress and egress location. In some cases, capturing the location information of each of the plurality of people associated with the facility can include capturing location information of two or more of the plurality of people associated with the facility during one or more previous evacuation events, and wherein at least one of the historical movement patterns can be based at least in part on the captured location information during the one or more previous evacuation events.
[0032] One or more historical movement patterns of one or more of the plurality of people associated with the facility are identified based at least in part on the captured location information and / or by processing the captured location information at least in part using a trained artificial intelligence (AI) model, as indicated by block 40.
[0033] One or more predefined evacuation routes for the facility are identified, as indicated by block 42. The captured location information is processed, at least in part, to identify a predicted emergency evacuation time for evacuating the facility using the one or more predefined evacuation routes in response to an evacuation event based, at least in part, on one or more historical movement patterns of one or more of the plurality of people associated with the facility and / or using a trained artificial intelligence (AI) model, as indicated by block 44. In some cases, the trained AI model can be trained to take into account the number and location of people in the facility at the time of the evacuation event and the crowd dynamics in the particular facility. In some cases, a floor plan of the facility can be provided to the AI model and the historical movement patterns can be expressed relative to the floor plan to provide spatial awareness to the trained AI model. The floor plan can include door locations, corridor locations, room locations, potential chokepoint locations, etc. In some cases, the number and location of people in the facility can be based, at least in part, on one or more of the historical movement patterns of the people in the facility. In some cases, the number and location of people in the facility can be based, at least in part, on real-time location data collected by an access control system of the facility.
[0034] In some cases, the method 36 can include training the AI model by feeding back actual emergency evacuation times after each of one or more previous evacuation events, as indicated by block 46. In some cases, the method 36 can also include feeding back actual location information of one or more of the plurality of people associated with the facility during each of the one or more previous evacuation events, as indicated by block 48. This can help improve the accuracy of the AI model in predicting future evacuation times over time.
[0035] Figure 4 A flowchart illustrating a series of exemplary steps that can be performed by one or more processors executing instructions stored on a non-transitory computer readable storage medium is shown. For example, the one or more processors can be part of the controller 14. When the one or more processors execute the instructions, the one or more processors are caused to receive location information of each of a plurality of people associated with a facility in the facility, as indicated by block 52. In some cases, the location information can include access control data collected by an access control system of the facility. The one or more processors are caused to identify one or more historical movement patterns of one or more of the plurality of people associated with the facility based, at least in part, on the captured location information, as indicated by block 54. The one or more processors are caused to predict an emergency evacuation time for evacuating the facility using one or more predefined evacuation routes based, at least in part, on the one or more historical movement patterns of the one or more of the plurality of people associated with the facility, as indicated by block 56.
[0036] In some cases, these predefined evacuation routes can be modified, resulting in one or more modified evacuation routes. Then, based at least in part on the historical behavior patterns, one or more processors can predict an evacuation time for evacuating the facility using the one or more modified evacuation routes. This can be repeated until a desired predicted evacuation time is reached.
[0037] In some cases, an evacuation drill can be initiated using an evacuation route predicted to achieve a desired evacuation time. The ingress and egress control system of the facility can be used to capture location information of each of the one or more people in the facility during the evacuation drill. The location information captured during the evacuation drill can be used as feedback to enhance the model, resulting in more accurate subsequent predictions of evacuation times for evacuating the facility.
[0038] In some cases, location information of each of the plurality of people in the facility can continue to be captured over time, resulting in updated location information. Updated historical behavior patterns of people in the facility can be determined based at least in part on the updated historical behavior patterns. From this, updated evacuation times for evacuating the facility using one or more evacuation routes can be predicted. If the updated evacuation times do not meet a desired predicted evacuation time, different evacuation routes can be defined, and corresponding evacuation times for each subsequent evacuation route can be predicted using the updated historical behavior patterns until the desired predicted evacuation time is reached.
[0039] Figure 5 is a flowchart illustrating an example overview 58. The overview 58 begins with a real-time monitoring block 60. In some cases, real-time monitoring includes integration with the ingress and egress control system of the building in order to obtain ingress and egress control event data. This enables real-time monitoring of employee movement during normal operation of the facility as well as during evacuation events. This data can include entry and exit times, preferred or usual points of ingress and egress, population distribution, etc. This data can be obtained using security cameras, access card data, mobile devices, manually entered data, and / or other sensor data that can track employee location and movement. The overview 58 includes an AI behavior pattern analysis block 62. The AI behavior pattern analysis block 62 can include using a trained AI model to analyze historical behavior patterns of employees during normal and emergency evacuations. In normal evacuations, the AI behavior pattern analysis block 62 can analyze employee ingress and egress logs and events to extract data such as times, frequencies, and locations of ingress attempts in order to sometimes identify behavior patterns for each individual employee. Data from previous evacuation events (e.g., fire drills, bomb threats, tornado drills, etc.) if available can also be used to better understand typical responses to alarms by each employee and how they navigate through the building during evacuations.
[0040] Summary 58 includes a predictive modeling block 64. Predictive modeling includes using an AI model to predict likely evacuation routes, times, and / or potential choke points based at least in part on historical behavior patterns. This can help optimize evacuation routes and reduce evacuation times. Summary 58 includes a scenario visualization block 66. Scenario visualization block 66 includes simulating different emergency scenarios prior to actually conducting an actual evacuation event (e.g., a fire drill). These simulations can be based on scenarios developed by predictive modeling block 64 and can take into account various factors such as holidays, personnel availability, and the status of various doors (locked or unlocked, blocked by fire, etc.). This can help gain insight into potential issues such as overcrowding and congestion, which can help plan more effective evacuation routes.
[0041] Summary 58 includes an optimized evacuation route block 68. Optimized evacuation route defines an optimized evacuation route. In some cases, the system can receive one or more modified evacuation routes from a user or can automatically derive one or more modified evacuation routes. To automatically derive an optimized modified evacuation route, optimized evacuation route block 68 can use a trained AI model to determine an optimized modified evacuation route. Alternatively or additionally, as part of a Monte Carlo simulation, optimized evacuation route block 68 can use each of a plurality of modified evacuation routes to predict an evacuation time for evacuating the facility and identify a modified evacuation route having a lowest evacuation time.
[0042] Summary 58 includes a feedback and improvement block 70. Feedback can be collected during an actual evacuation drill regarding the location and / or behavior of each occupant. The feedback can be used to improve the accuracy of predictive models (e.g., AI models) as well as the overall effectiveness of the evacuation process.
[0043] In some cases, the optimized evacuation route can change depending on current conditions associated with the building. For example, the optimized evacuation route can change depending on the number of people in the building, the distribution of people in the building, particular people in the building, particular passageway doors that are available, the location of an emergency event in the building (i.e., the location of a fire), etc. In some cases, digital signage can be provided throughout the building that provides the current optimized evacuation route to occupants of the building. Additionally or alternatively, the current optimized evacuation route can be sent directly to each occupant’s mobile phone. In some cases, depending on the relative locations of a first occupant and a second occupant in the building when an evacuation event occurs, the current optimized evacuation route sent to the first occupant of the building can be different than the current optimized evacuation route sent to the second occupant of the building.
[0044] While several illustrative embodiments of the disclosure have been described, it will be apparent to those skilled in the art that other embodiments can be made and utilized, within the scope of the appended claims. However, it should be understood that the disclosure is, in many respects, only illustrative. Changes can be made in the details, particularly in matters of shape, size, arrangement of parts, and order of steps, without departing from the scope of the disclosure. Accordingly, the scope of the disclosure is to be defined only by the language of the following claims.
Claims
1. A method of predicting an emergency evacuation time for a facility, the method comprising: capturing location information for each of a plurality of people associated with a facility in the facility; identifying one or more historical movement patterns for one or more of the plurality of people associated with the facility based at least in part on the captured location information; identifying one or more predefined evacuation routes for the facility; and predicting the emergency evacuation time for evacuating the facility in response to an evacuation event using the one or more predefined evacuation routes based at least in part on the one or more historical movement patterns for one or more of the plurality of people associated with the facility.
2. The method of claim 1, wherein the location information comprises access control data collected by an access control system of the facility.
3. The method of claim 2, wherein the access control data comprises a personal identifier for each person passing through an access control point of the access control system, an access event time, and an access location.
4. The method of claim 1, the method comprising: identifying which of the plurality of people associated with the facility are currently in the facility; and predicting the emergency evacuation time for evacuating the facility in response to an evacuation event using the one or more predefined evacuation routes based at least in part on the one or more historical movement patterns for the people identified as currently in the facility.
5. The method of claim 1, wherein capturing location information for each of the plurality of people associated with the facility comprises capturing location information for two or more of the plurality of people associated with the facility during one or more previous evacuation events, and wherein at least one of the historical movement patterns is based at least in part on the location information captured during the one or more previous evacuation events.
6. The method of claim 1, wherein predicting the emergency evacuation time for evacuating people from the facility during the evacuation event comprises using a trained artificial intelligence (AI) model to at least partially process the captured location information to identify the predicted emergency evacuation time, wherein the trained AI model is trained to account for a number and location of people in the facility at the time of the evacuation event and a crowd dynamics in the facility.
7. The method of claim 6, the method comprising training the AI model by one or more of: feedback of an actual emergency evacuation time after each of one or more previous evacuation events; and feedback of actual location information for one or more of the plurality of people associated with a facility during each of one or more previous evacuation events.
8. The method of claim 1, wherein the location information for each of the plurality of people associated with the facility comprises an entry time and an exit time at a preferred access point of the facility. 9. The method of claim 1, wherein identifying one or more historical movement patterns of one or more of the plurality of people associated with the facility comprises using a trained artificial intelligence (AI) model to at least partially process the captured location information to identify one or more historical movement patterns of one or more of the plurality of people associated with the facility.
10. An emergency evacuation prediction system for a facility, the emergency evacuation prediction system comprising: an access control system to capture location information for each of a plurality of people associated with the facility; a controller operatively coupled with the access control system, the controller configured to: identify one or more historical movement patterns of one or more of the plurality of people associated with the facility based at least in part on the captured location information; and predict an emergency evacuation time for evacuating the facility using one or more predefined evacuation routes based at least in part on the one or more historical movement patterns of one or more of the plurality of people associated with the facility.