Skip to main navigation Skip to search Skip to main content

Representing Collected Road Condition Data with Chernoff Faces for Evaluation of Pavement Conditions

  • École de technologie supérieure
  • Technical Faculty of Structural Engineering

Research output: Contribution to Book/Report types › Contribution to conference proceedings › peer-review

Abstract

Maintenance engineers have a problem because of the great quantity of data they must consider when evaluating kilometres of roads and because road roughness data measured with IRI (or possibly PCI) cannot be considered in isolation from other data such as Construction and Workmanship; Mix and Structural Design; Traffic; Climate (during and after construction). Incorporating these five datapoints into a single conventional graph would require a graph in five dimensions and such a graph would lose much of the benefits that a graph is meant to give (i.e. speed of data comprehension and readability). The Chernoff Face Method (CFM) is an iconic visualization technique used to code multivariate data to simplify two-dimensional line drawings of faces so as to create a representation of the data that is more intuitively comprehensible to humans. The theory behind CFM is that humans have a special sensitivity to details in a facial representation that they do not have with data representations such as conventional graphic displays (e.g. line and bar graphs). The objective is to find an optimum balance between the volume of the data represented and the comprehensibility of the data. In particular, humans can discern eye size and eyebrows slant in faces with finer distinctions than they would be able to intuitively understand when reading another graph. This promises to be a means of displaying fine grained data for analysis by mapping data points to possible display variables such as: eye size, face shape, eyebrows slant angle (inward, outward, neutral). In order to map the IRI, construction, design, traffic and climate data to the CFM, a Multi-Criteria Analysis (MCA) model was used to determine the optimal number for the scale for each data category. The total number of Chernoff Faces for this problem was found to be 243, so that all five categories of data could be represented. Using CFM in conjunction with this data allows better evaluation of pavement distress.

Original languageEnglish
Title of host publicationTesting and Characterization of Asphalt Materials and Pavement Structures - Proceedings of the 5th GeoChina International Conference 2018 – Civil Infrastructures Confronting Severe Weathers and Climate Changes
Subtitle of host publicationFrom Failure to Sustainability, 2018 in HangZhou, China
EditorsKun Zhang, Rongqiao Xu, Shih-Huang Chen
PublisherSpringer Science and Business Media B.V.
Pages136-150
Number of pages15
ISBN (Print)9783319957883
DOIs
Publication statusPublished - 2019
Event5th GeoChina International Conference on Civil Infrastructures Confronting Severe Weathers and Climate Changes: From Failure to Sustainability, 2018 - HangZhou, China
Duration: 23 Jul 2018 → 25 Jul 2018

Publication series

NameSustainable Civil Infrastructures
ISSN (Print)2366-3405
ISSN (Electronic)2366-3413

Conference

Conference5th GeoChina International Conference on Civil Infrastructures Confronting Severe Weathers and Climate Changes: From Failure to Sustainability, 2018
Country/TerritoryChina
CityHangZhou
Period23/07/18 → 25/07/18

Keywords

  • Chernoff face
  • Climate zone
  • International roughness index
  • Material & construction
  • Multi-criteria analysis
  • Structural design
  • Traffic level

ASJC Scopus subject areas

  • Computational Mechanics
  • Environmental Engineering
  • Civil and Structural Engineering
  • Geotechnical Engineering and Engineering Geology

Fingerprint

Dive into the research topics of 'Representing Collected Road Condition Data with Chernoff Faces for Evaluation of Pavement Conditions'. These topics are generated from the title and abstract of the publication. Together, they form a unique fingerprint.

Cite this