Bus Dwell Time Modeling Using Gene Expression Programming

Abstract

This article proposes a gene expression programming (GEP)‐based approach to model and estimate bus dwell time (BDT) as an alternative to the more commonly used multiple linear regression (MLR) model. The proposed model is calibrated and validated using the data collected from 22 bus stops in Auckland and compared against the MLR model based on five different performance measures namely: mean error, mean absolute error, root mean square error, mean absolute percentage error, and R2 value. The proposed GEP‐based approach shows prospects to estimate BDT more accurately and overcome some of the issues associated with the MLR method, including the restrictions to stick with a predefined model form and the need to satisfy assumptions made on multicollinearity, homoscedasticity, and the normality of random error, which are often difficult to satisfy in real world applications.

A Classification-Based Surrogate-Assisted Evolutionary Algorithm for Expensive Many-Objective Optimization

“…One grand challenge in solving many real-world MaOPs is that one single fitness evaluation (FE) is computationally and/or financially very expensive, since it requires timeconsuming computer simulations or physical experiments [20], [21], e.g., in aerodynamic design optimization [22], drug design [23] or flowshop scheduling problems in [24]. Take a ten-job and five-machine flowshop scheduling problem as an example [24], it will take over 200 days for a conventional MOEA if a total of 10,000 FEs is used, which is impractical.…”

Lightweight parametric design optimization for 4D printed parts

“…The optimization algorithm implemented in this paper is based on genetic algorithms (GAs) and FEA. There are several metaheuristic methods that are usually applied in optimization problems such as GAs [3,27], Differential Evolution (DE) [41], Particle Swarm Optimization (PSO) [40], Gene Expression Programming (GEP) [38] or Ant Colony Optimization (ACO). These techniques have been applied in different fields, such as shape and topology optimization for structures [1,14,19,20].…”

Physics‐based search and optimization: Inspirations from nature

“…where x i ( t ) represents a candidate solution at iteration t , f i is the fitness function value of x i ( t ), and N is the population size. This operator is different from selection operator used in many other optimization algorithms such as genetic algorithms (GA); (Lee, & Arditi, ; Paris, Pedrino, & Nicoletti, ; Bolourchi, Masri, & Aldraihem, ; Park, Oh, & Park, ), genetic programming (Rashidi & Ranjitkar, ; Mesejo et al, ), or particle swarm optimization (PSO) (Zeng, Xu, Wu, & Shen, ; Shabbir & Omenzetter, ).…”