This award provides funding for the study of matching and allocation problems with applications to a variety of emerging areas, including applications spurred by the growth of the internet and electronic commerce, as well as non-standard applications such as the design of exchanges for kidney transplants, the assignment of students to schools in districts where there is school choice, etc. Particular attention will be paid to the design of mechanisms in which participants do not benefit from misreporting any private information (such as their preferences) that is relevant to the allocation problem under consideration. These problems will be studied in an axiomatic framework, which seeks to understand mechanisms that satisfy many desirable criteria, whose choice is motivated by the application under consideration. Minimal sets of properties that uniquely determine commonly-used mechanisms will be identified. Further, new mechanisms with compelling efficiency and equity properties will be designed and their use in practice will be evaluated. Several classical flow and matching models---traditionally studied in the operations research literature from the point of view of computational efficiency---will be studied, taking into account incentive constraints and equity objectives.
If successful, the project will result in new mechanisms for several real-life matching and allocation problems. It will also result in the development of new theoretical tools and techniques that may be broadly applicable in understanding the interaction between equity objectives and (both economic and computational) efficiency objectives in the presence of incentive constraints. This award will support the development of a doctoral course focusing on the applications of matching and allocation problems in practice, as well as the redesign of a course that covers O.R. methods in the public sector. The award will also support a graduate student and outreach activities such as the writing of expository articles.
