Dynamic Pricing and Matching for Online Marketplaces

Explore a technical lecture examining dynamic pricing and matching mechanisms in online marketplaces and the gig economy through a bipartite matching network analysis. Learn about a proposed two-price policy and max-weight matching policy that achieves a η1/3 optimality rate with scaled arrival rates, while discovering how max-weight matching advantages vary with server and customer types through state space collapse. Delve into the special case analysis of single customer and server type scenarios, understanding queue length distribution in heavy traffic and the unique phase transition characteristics of matching queue limiting distribution. Drawing from queueing theory, game theory, and revenue management applications, gain insights from Georgia Tech Ph.D. candidate Sushil Varma's research on ride-hailing, load balancing, and stochastic processing networks.