Funded Projects

Green Infrastructure (GI) is a network of semi-natural constructions that use vegetation and engineered tools to manage stormwater in a cost-efficient way while providing several benefits to the communities in the surrounding areas. One of the many advantages of GI is the reduction of potential flooding events that may occur by controlling the water runoff of stormwater. The installation of GI is currently an endeavor mainly carried by local governments in public areas and, despite its potential benefits, it is uncommon to see such type of infrastructure in private land. Typical hurdles like the installation and maintenance costs as well as the lack of interest, knowledge, and means have prevented landowners to install GI in their property. Alas, having GI installed in private land has been proven to be highly beneficial for cities, specially in largely congested areas with no available public land suitable for GI. Several cities, like New York, Cincinnati, and Portland, have launched social programs that distribute monetary incentives to persuade local residents to voluntarily install GI in their land. However, identifying efficient policies that optimally distribute such incentives to maximize the effects of the new GI installations is still a major challenge. Among the many difficulties of designing such type of programs are the limited budget and the fact that the program must incorporate both the government social objective as well as each landowner's individual response within the incentive distribution process. That is, an efficient incentive program must be designed to anticipate the landowners response to strategically allocate the available resources and maximize the common good. We propose to tackle the problem of designing such incentive programs by using a two-stage framework that couples machine learning techniques and optimization models. The preliminary results obtained from this project will serve the development of a subsequent grant proposal.