Electrifying the Urban Taxi Fleet: A Data-driven Approach

TL;DR

This paper presents a comprehensive data-driven framework for electrifying urban taxi fleets, optimizing dispatching and charging strategies based on real-time data to improve efficiency and equity.

Contribution

It introduces a novel simulation framework that models electric taxi operations and dispatching, applicable to various cities and fleet types.

Findings

Accurately models electric vehicle charging behavior and dispatching.

Reduces taxi customer waiting times and increases demand fill rate.

Ensures equitable income distribution among taxi drivers.

Abstract

This paper is devoted to proposing a data-driven approach for electrifying the urban taxi fleet. Specifically, based on the gathered real-time vehicle trajectory data of 39053 taxis in Beijing, we conduct time-series simulations to derive insights on both the configuration of electric taxi fleet and dispatching strategies. The proposed simulation framework accurately models the electric vehicle charging behavior from the aspects of time window, charging demand and availability of unoccupied charges, and further incorporates a centralized and intelligent fleet dispatching platform, which is capable of handling taxi service requests and arranging electric taxis' recharging in real time. To address the impacts of the limited driving range and long battery-recharging time on the electrified fleet's operations efficiency, the dispatching platform integrates the information of customers, taxi drivers and charging stations, and adopts a rule-based approach to achieve multiple objectives, including reducing taxi customers' average waiting time, increasing the taxi demand fill rate and guaranteeing the equity of the income among taxi drivers. Although this study only examines one type of fleet in a specific city, the methodological framework is readily applicable to other cities and types of fleet with similar dataset available.