Combining Bayesian Networks and MCDA methods to maximise information gain during reconnaissance in emergency situations
In the immediacy of an event that disrupts the operation of an infrastructure, the time between its occurrence and the arrival of qualified personnel for emergency response can be valuable. For example, it can be used for gathering information about the status of the infrastructure by using automate...
Saved in:
| Main Author: | |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
KeAi Communications Co., Ltd.
2025-03-01
|
| Series: | Journal of Safety Science and Resilience |
| Subjects: | |
| Online Access: | http://www.sciencedirect.com/science/article/pii/S2666449624000525 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1841553791182176256 |
|---|---|
| author | Daniel Lichte |
| author_facet | Daniel Lichte |
| author_sort | Daniel Lichte |
| collection | DOAJ |
| description | In the immediacy of an event that disrupts the operation of an infrastructure, the time between its occurrence and the arrival of qualified personnel for emergency response can be valuable. For example, it can be used for gathering information about the status of the infrastructure by using automated reconnaissance devices. In an operation that precedes the intervention of human first responders, such devices can gather information about the situation, providing knowledge about the locations of stressors (e.g. fire), the inaccessibility of parts of the infrastructure or the presence of hazardous materials. In this study, we show how a Bayesian Networks can be used for knowledge representation and how it can be combined with methods from the realm of Multi-Criteria Decision Analysis (MCDA) for situation reconnaissance and route-optimisation in emergency situations, where different criteria (current belief about the location of zones of special interest, such as emergency exits, distance to the next point of interest, etc.) can be considered. As an example, we consider the case of an outbreak of a fire in a building. A pedantic check of all rooms by an automated reconnaissance device would take too long and thus delay intervention. Due to the limited time in which the building can be explored, the route is optimised to gather the greatest possible amount of information in the available time window. Results show how it is possible to maximise the information collected in a limited time window. This is done by discovering the location of fire and any hazardous materials through causal inferences automatically calculated by the Bayesian network. Route optimisation is facilitated by sequential MCDA using a parameter selection that meets the priorities of the specific application example. |
| format | Article |
| id | doaj-art-046b142b789f4d209ca1c5c72e97d65b |
| institution | Kabale University |
| issn | 2666-4496 |
| language | English |
| publishDate | 2025-03-01 |
| publisher | KeAi Communications Co., Ltd. |
| record_format | Article |
| series | Journal of Safety Science and Resilience |
| spelling | doaj-art-046b142b789f4d209ca1c5c72e97d65b2025-01-09T06:14:47ZengKeAi Communications Co., Ltd.Journal of Safety Science and Resilience2666-44962025-03-01613847Combining Bayesian Networks and MCDA methods to maximise information gain during reconnaissance in emergency situationsDaniel Lichte0German Aerospace Center, Institute for the Protection of Terrestrial Infrastructures, Rathausallee 12, 53757 Sankt Augustin, GermanyIn the immediacy of an event that disrupts the operation of an infrastructure, the time between its occurrence and the arrival of qualified personnel for emergency response can be valuable. For example, it can be used for gathering information about the status of the infrastructure by using automated reconnaissance devices. In an operation that precedes the intervention of human first responders, such devices can gather information about the situation, providing knowledge about the locations of stressors (e.g. fire), the inaccessibility of parts of the infrastructure or the presence of hazardous materials. In this study, we show how a Bayesian Networks can be used for knowledge representation and how it can be combined with methods from the realm of Multi-Criteria Decision Analysis (MCDA) for situation reconnaissance and route-optimisation in emergency situations, where different criteria (current belief about the location of zones of special interest, such as emergency exits, distance to the next point of interest, etc.) can be considered. As an example, we consider the case of an outbreak of a fire in a building. A pedantic check of all rooms by an automated reconnaissance device would take too long and thus delay intervention. Due to the limited time in which the building can be explored, the route is optimised to gather the greatest possible amount of information in the available time window. Results show how it is possible to maximise the information collected in a limited time window. This is done by discovering the location of fire and any hazardous materials through causal inferences automatically calculated by the Bayesian network. Route optimisation is facilitated by sequential MCDA using a parameter selection that meets the priorities of the specific application example.http://www.sciencedirect.com/science/article/pii/S2666449624000525Emergency surveillanceBayesian networksMulti-criteria decision analysisInformation collectionResilience |
| spellingShingle | Daniel Lichte Combining Bayesian Networks and MCDA methods to maximise information gain during reconnaissance in emergency situations Journal of Safety Science and Resilience Emergency surveillance Bayesian networks Multi-criteria decision analysis Information collection Resilience |
| title | Combining Bayesian Networks and MCDA methods to maximise information gain during reconnaissance in emergency situations |
| title_full | Combining Bayesian Networks and MCDA methods to maximise information gain during reconnaissance in emergency situations |
| title_fullStr | Combining Bayesian Networks and MCDA methods to maximise information gain during reconnaissance in emergency situations |
| title_full_unstemmed | Combining Bayesian Networks and MCDA methods to maximise information gain during reconnaissance in emergency situations |
| title_short | Combining Bayesian Networks and MCDA methods to maximise information gain during reconnaissance in emergency situations |
| title_sort | combining bayesian networks and mcda methods to maximise information gain during reconnaissance in emergency situations |
| topic | Emergency surveillance Bayesian networks Multi-criteria decision analysis Information collection Resilience |
| url | http://www.sciencedirect.com/science/article/pii/S2666449624000525 |
| work_keys_str_mv | AT daniellichte combiningbayesiannetworksandmcdamethodstomaximiseinformationgainduringreconnaissanceinemergencysituations |