Simulation Based Fleet Relocation System

Popularity of the e-hailing taxi models radically changes the character of the city and city traffic. In the past few years alone, the number of daily taxi rides order through the app in Moscow has reached over a million per day, that is almost five times more than the volume of traditional taxi hailing services. Now, in almost every country new taxi service models are squeezing classic taxi models with offline fleet parks and phone operators. The rapid spread of the model of e-hailing transportation and the presence of complex problems at the intersection of economics and computer science attract many researchers to study and model the market for taxi services. However, despite the objective popularity of research on the effectiveness of platforms, there is still much less studies on optimization of strategies for agents in the system — taxi drivers. In my work, I would like to formalize the problem of developing a driver behavior strategy that maximizes their expected income.