INTERCITY MODE CHOICE STUDY IN AKURE, NIGERIA: BINOMIAL LOGISTIC REGRESSION AND ARTIFICIAL NEURAL NETWORK APPROACH

Summary

This study examines the factors influencing intercity mode choice behavior among travelers in Akure, Nigeria,
with the aim of identifying key determinants, developing predictive models, and providing policy
recommendations for improved transportation systems. Using both the Binomial Logistic Regression (BNL)
model and the Radial Basis Function (RBF) neural network, data were obtained through a structured
questionnaire survey targeting travelers along Akure’s three main intercity corridors—Akure-Owo, AkureOndo, and Akure-Osun. The survey captured trip characteristics, personal socio-demographics, and transport
facility attributes from a stratified random sample of 1,360 respondents. Analysis revealed that income, cost
of transportation, gender, duration of stay, purpose of trip, trip destination, age, and travel distance
significantly influenced mode choice. The RBF neural network model achieved higher predictive accuracy (R²
= 0.991, MAPE < 0.05) compared to the BNL model (R² = 0.669, MAPE < 0.08). Findings indicate that
private transportation dominates (54.7%) due to comfort and flexibility, while public transport is preferred
by lower-income groups due to affordability. It is recommended that policymakers invest in efficient,
affordable, and reliable public transport to reduce private vehicle dependency, congestion, and associated
environmental impacts. This research contributes by providing empirical evidence on intercity travel behavior
in Akure, validating the superiority of neural network approaches in travel demand modeling, and offering
practical strategies for sustainable mobility planning.

Index Terms

Browse additional articles authored by one or more contributors to this paper.