Predicting the maintenance cost of construction equipment: Comparison between general regression neural network and Box-Jenkins time series models

Hon Lun Yip, Hongqin Fan, Yat Hung Chiang

Research output: Journal article publicationJournal articleAcademic researchpeer-review

47 Citations (Scopus)

Abstract

This paper presents a comparative study on the applications of general regression neural network (GRNN) models and conventional Box-Jenkins time series models to predict the maintenance cost of construction equipment. The comparison is based on the generic time series analysis assumption that time-sequenced observations have serial correlations within the time series and cross correlations with the explanatory time series. Both GRNN and Box-Jenkins time series models can describe the behavior and predict the maintenance costs of different equipment categories and fleets with an acceptable level of accuracy. Forecasting with multivariate GRNN models was improved significantly after incorporating parallel fuel consumption data as an explanatory time series. An accurate forecasting of equipment maintenance cost into the future can facilitate decision support tasks such as equipment budget and resource planning, equipment replacement, and determining the internal rate of charge on equipment use.
Original languageEnglish
Pages (from-to)30-38
Number of pages9
JournalAutomation in Construction
Volume38
DOIs
Publication statusPublished - 1 Mar 2014

Keywords

  • Construction equipment
  • General regression neural network
  • Maintenance management
  • Time series analysis

ASJC Scopus subject areas

  • Control and Systems Engineering
  • Civil and Structural Engineering
  • Building and Construction

Cite this