Guiding vacant taxi drivers to demand locations by taxi-calling signals: A sequential binary logistic regression modeling approach and policy implications

Abstract

Taxi-calling signals (TCSs) have appeared in many cities to reveal passenger demand at locations away from the roadside to cruising vacant taxis to reduce search times for both vacant taxi drivers and customers. This study aims to find out the factors influencing vacant taxi drivers' customer-search decisions on whether to enter or bypass recommended areas while the drivers are cruising along a road with a series of TCSs. Observational survey data were collected and analyzed to understand the travel behavior of vacant taxi drivers. A sequential binary logistic regression (SBLR) model is first proposed to examine the dynamic decision-making process of vacant taxi drivers. A simulation model and a solution procedure are then developed by adopting the intervening opportunity modeling concept to validate the SBLR model. A sensitivity analysis is consequently conducted to show that the installation of TCSs can effectively increase the number of vacant taxis entering off-road locations for picking up customers. Potential policy implications are discussed.