Project R-1991

Analysis of personal exposure using sensor networks (Research)

Society is becoming increasingly sensitive to individual risks. For many general risks, such as road accidents and air pollution, individuals are now more risk averse than before. However, policy choices will (eg regarding the environment) often change the human exposure to pollution and change the individual health risk. These guidelines are usually only at the population level studied and assessed. Historically the exposure of the population (especially regarding air pollution) has always been calculated by simply multiplying the calculated concentrations of pollutants with the population density at the same place. Measuring the pollution for a long time in many places, is usually associated with high costs and is complicated by the complex and large measuring equipment. Recently, Hasselt University and VITO demonstrated that dynamic analysis better allows to predict the exposure of urban populations than the traditional static approach. As people throughout the day travel from one place to another, their exposure will be determined by the background concentrations in each of those locations, plus local sources (indoor, school, car, ...). We formulate the hypothesis that the movement between micro-environments is the main determinant of differences in exposure between people who live and work within the same geographical unit. Policies that could intervene in the daily routine activities may be an effective tool to change the exposure (reduce). To test this hypothesis, three complementary methods: 1. the use of a time-use model to predict the times and locations of exposure 2. newly developed low-cost sensors for air pollution are deployed in a network so data can be gathered concerning the exposure of specific groups in the population 3. Data mining to manage and analyze large amounts of data from a distributed sensor network

15 November 2009 - 14 November 2013