Determination of Work Zone Capacity Using ELM, MPMR and GPR

Abstract

This article examines the capability of Extreme Learning Machine (ELM), Minimax Probability Machine Regression (MPMR) and Gaussian Process Regression (GPR) for determination of Work Zone Capacity. Number of lanes, number of open lanes, work zone layout, length, lane width, percentage trucks, grade, speed, work intensity, darkness factor, and proximity of ramps have been adopted as inputs of ELM, MPMR and GPR. ELM has excellent generalization performance, rapid training speed and little human intervention. MPMR is developed based on the concept of minimax probability machine classification. It does not assume any data distribution. GPR is a probabilistic, and non-parametric model. In GPR, different kinds of prior knowledge can be applied. This article describes a comparative study between the ELM, MPMR and GPR models.