Heat map visualisation of fire incidents based on transformed sigmoid risk model. (October 2019)

• Record Type: Journal Article
• Authors: Liu, Dingli; Xu, Zhisheng; Zhou, Yang; Fan, Chuangang
• Is Part Of: Fire safety journal. Volume 109(2019)
• Publication Date: 2019-10
• Subjects: Transformed sigmoid risk model -- Fire incident -- Data visualisation -- Heat map -- Linear risk model -- Density clustering; Fire prevention -- Periodicals; Incendies -- Prévention -- Recherche -- Périodiques; Fire prevention -- Research; Periodicals; 628.92205
• DOI: 10.1016/j.firesaf.2019.102863
• Languages: English
• ISSNs: 0379-7112

Abstract

Abstract: Fire is one of the most frequent disasters that threaten public safety and ecological balance. Heat maps have been used as a common visualisation method in many scientific fields but rarely in fire risk analysis. By using the transformed sigmoid function to optimise linear gradation, a new heat map method was proposed based on the classical heat map method; this was referred to as the transformed sigmoid heat map method. Combining the classical heat map method with fire risk theory, a linear risk model (LRM) was proposed. Similarly, a transformed sigmoid risk model (TSRM) was created using the transformed sigmoid heat map method. A dataset of 2002 fire incidents in Loudi City (located in South China) from 2014 to 2016 was used as the case study. The heat map visualisation of fire incidents showed that, compared to the TSRM, the LRM led to the overgeneralisation of results more easily. High- and medium-risks were mainly distributed in densely populated areas. In addition, human activity had a significant influence on the fire risk distribution in different time ranges. As a result of human activity, the number of fires during 00:00–06:00 were 57.30% of that during 12:00–18:00. These visual analysis results are useful for deploying firefighters and rescue forces depending on the fire risk in different areas and at different times. In conclusion, an online and interactive fire-risk-analysing software based on TSRM should be developed that can be applied to fire risk analysis in other regions. Highlights: Optimising the classical heat map method. Bringing forward transformed sigmoid risk model (TSRM) and line risk model (LRM). Heat map visualisation of 2002 fire incidents based on TSRM and LRM. LRM leads to overgeneralisation more easily than TSRM. Urban areas have a higher fire risk than rural areas.