A Microsimulation Model for Assessing Urine Flows in Urban Wastewater Management

We report on a microsimulation model under development whose purpose it is to construct scenarios for the assessment of a novel wastewater management technology: the separate collection, transport, and handling of human urine. Levelling the urine flow to the treatment plant or removing it from conventional treatment altogether has many benefits for wastewater management. For an effective technology implementation, it is useful to know the spatial and temporal distribution of urine generation. This can be achieved with microsimulation modelling which depicts the objects under study on a one-to-one basis and thus allows capturing the co-variation of different variables across these microunits. In our case, microunits are buildings, apartments, toilets, and people; technology impacts are driven by physical characteristics of the building infrastructure and sociodemographic characteristics of people, including their mobility behavior. We are modelling a Swiss region comprised of 18 small municipalities that form the catchment area of a wastewater treatment plant. Swiss census data provides us with information on residential buildings and their approximate geographical location, apartments, and residents and their sociodemographic and employment profile. Using event-based simulation, we set up a daily agenda for every person in our model, sending them to work and on other trips and letting them urinate in different locations. In work to date, we could replicate the typical morning peak in urine generation and identify the areas with the greatest urine density. In future work, we will construct scenarios describing the diffusion of the technology. Diffusion can be modelled endogenously as we depict the behavior of microunits and can let them react to geographical and social information contagion. We find microsimulation advantageous because it captures the heterogenity of microunits, facilitates the linking of data from different sources, and allows many different questions to be addressed in one and the same flexible modelling framework.