Vehicle Dimensions Based Passenger Car Classification using Fuzzy and Non-Fuzzy Clustering Methods

Abstract

There has been globally continuous growth in passenger car sizes and types over the past few decades. To assess the development of vehicular specifications in this context and to evaluate changes in powertrain technologies depending on surrounding frame conditions, such as charging stations and vehicle taxation policy, we need a detailed understanding of the vehicle fleet composition. This paper aims therefore to introduce a novel mathematical approach to segment passenger vehicles based on dimensions features using a means fuzzy clustering algorithm, Fuzzy C-means (FCM), and a non-fuzzy clustering algorithm, K-means (KM). We analyze the performance of the proposed algorithms and compare them with Swiss expert segmentation. Experiments on the real data sets demonstrate that the FCM classifier has better correlation with the expert segmentation than KM. Furthermore, the outputs from FCM with five clusters show that the proposed algorithm has a superior performance for accurate vehicle categorization because of its capacity to recognize and consolidate dimension attributes from the unsupervised data set. Its performance in categorizing vehicles was promising with an average accuracy rate of 79% and an average positive predictive value of 75%.

Segment-Based CO₂ Emission Evaluations From Passenger Cars Based on Deep Learning Techniques

“…Since the division of vehicles into segments by experts is not standardized and therefore not always uniform, and some vehicle models have recently positioned themselves as "crossovers" between established vehicle categories [7][8], it has become increasingly difficult and inaccurate to segment the vehicle population using conventional classification methods. Using mathematical approaches, vehicles can be uniformly divided into segments based on similarity features.…”

A science-based approach to classifying light vehicles in Europe: methodology and case studies

“…The compilation of chosen attributes is detailed in Table 2. The selection of these features was informed by their significance as established in prior research 5,32,35 and based on the statistical evaluations presented in Additional Information 1. The selected attributes are the Maximum number of passengers, Power-to-mass ratio and Maximum cargo mass.…”

“…Some of them 6,[27][28][29] have investigated various schemes using statistical and exploratory data analysis, aiming to describe typical characteristics of each segment. However, due to the complexity of the classification problem, most of these studies 5,[30][31][32][33][34] have tackled the classification and dimensionality reduction task with advanced Machine Learning (ML) algorithms and principal component analysis techniques, respectively. These studies have demonstrated that ML techniques can successfully address the complex problem of cars classification.…”

“…These studies have demonstrated that ML techniques can successfully address the complex problem of cars classification. Among them, the series of works by Niroomand et al 5,32,35 proved able to reproduce the EU classification scheme with an accuracy of up to 85%. Nevertheless, ML models obscure the understanding of vehicles classification, so that the criteria behind the assignation of a vehicle to a certain segment remain opaque.…”

Large dataset partitioning using ensemble partition-based clustering with majority voting technique

“…A typical data clustering process starts with a group of information items and splits them into k clusters using Euclidean distance and other similarity distance metrics. Partition-based clusters could satisfy a few of following requirements: i) Each cluster should have at least one data item and ii) In non-fuzzy clustering algorithms, each object must only be present in one cluster [9].…”