Hybrid ANN-CBR model for disputed change orders in construction projects

Jieh Haur Chen, Shu-Chien Hsu

Research output: Journal article publicationJournal articleAcademic researchpeer-review

49 Citations (Scopus)

Abstract

The purpose of this paper is to provide a method that can be used to solve potential lawsuit problems caused by change orders in construction projects. A hybrid Artificial Intelligence (AI) model, the Hybrid ANN-CBR Model (HACM), is developed utilizing the AI branches of Artificial Neural Networks (ANN) and Case Based Reasoning (CBR). The research is based on the litigation archives collected by the Supreme Courts and appellate courts in 48 states and one district of the USA. The accuracy of the HACM prediction rate reaches 84.61%, not only for predicting litigation likelihood by using the ANN approach but also utilizing the CBR approach to yield warnings and display litigation information related to past cases. After evaluating 31 cases it is confirmed that the model HACM performs well especially for those medium sized construction projects. It can be concluded that it is feasible to link ANN and CBR together to provide a tool with a relatively high rate of prediction accuracy and a conceptual model to solve potential severe disputes cased by change orders.
Original languageEnglish
Pages (from-to)56-64
Number of pages9
JournalAutomation in Construction
Volume17
Issue number1
DOIs
Publication statusPublished - 1 Nov 2007

Keywords

  • AI
  • ANN
  • CBR
  • Change orders
  • Construction management
  • Disputes
  • Litigation

ASJC Scopus subject areas

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

Cite this