International Journal on Advanced Science, Engineering and Information Technology, Vol. 13 (2023) No. 5, pages: 1663-1671, DOI:10.18517/ijaseit.13.5.18983

Bush Fire Simulation through Emotion-based BDI Methodology

Celine Haren Paschal, Muhammad Asyraf bin Khairuddin, Cheah Wai Shiang, Mohamad Nazri bin Khairuddin Yap


This paper introduces an emotion-based BDI (Belief, desire, intention) methodology to model decision-making during fire evacuation simulations while considering human emotions. The methodology is designed to represent human decision-making processes in graphical representations, which can be simply translated for the implementation phase to simulate various case studies. The methodology utilizes the Belief, Desire, and Intention architecture and the OCEAN Openness, Conscientiousness, Extraversion, Agreeableness, and Neuroticism) personality behavior to represent decision-making processes graphically, making it easy to translate into a simulation. The methodology aims to create a more realistic simulation closer to real human behavior by incorporating emotions that affect decision-making. In this paper, we validate the emotion-based BDI methodology by replicating the bushfire Australia case study and benchmarking with the previous work on BDI fire evacuation. From the comparison, we found that both results share almost similar patterns. The results show "dead while still unaware" (0% vs. 0%), "dead while deciding what to do" (69% vs. 48%), "dead while defending" (6% vs. 8%), and "dead while preparing to defend" (6% vs 28%), "dead while preparing to escape" (4% vs 0%) and "dead while escaping" (15% vs 20%). The results show that in our Simulation, there is a death related to preparing to escape (4% vs 0%). However, the other causes of death have an almost similar percentage of death causes. Hence, based on the comparison, supporting and validating our emotion-oriented simulation model is considered adequate. Therefore, this emotion-based BDI methodology can systematically reproduce human cognition and emotion.


Emotion based modelling; BDI; multi-agent

Viewed: 136 times (since abstract online)

cite this paper     download