Sequence-oriented sensitive analysis for PM2.5 exposure and risk assessment using interactive process mining

Area of Science:

• Environmental Science and Public Health
• Data Science and Process Mining

Background:

• Air pollution is a major global environmental and health challenge, impacting mortality risk.
• Effective air quality monitoring and climate change mitigation are crucial, as highlighted by the Paris Agreement.
• Existing methods for analyzing exposure data and proposing mitigation strategies require enhancement.

Purpose of the Study:

• To propose a novel methodology for sequence-oriented sensitive analysis to support air pollution personal exposure mitigation.
• To develop a human-oriented approach integrating environmental experts' knowledge with Interactive Process Mining.
• To identify optimal activities and parameters for effective mitigation policies.

Main Methods:

• Application of Interactive Process Mining for personal air pollution exposure mitigation, a first in this field.
• Development of a sequence-oriented sensitive analysis technique.
• Testing the methodology with synthetic data generated using demographic and epidemiological statistics from Valencia, Spain.

Main Results:

• The methodology successfully identified target activities for mitigation through sequence-oriented sensitive analysis.
• The proposed scenarios demonstrated significant improvements in Key Performance Indicators (KPIs).
• In the best-case scenario, population exposure was reduced by 18%, and relative risk decreased by 12%.

Conclusions:

• The proposed Interactive Process Mining-based methodology is a viable and innovative tool for air pollution mitigation.
• This approach reduces computational costs and time compared to traditional sensitive analysis.
• The methodology's human-oriented nature and proven effectiveness with synthetic data support its future application with real-world measurements for environmental improvement.