icccbe
2010
© Nottingham University Press
Proceedings of the International Conference on
Computing in Civil and Building Engineering
W Tizani (Editor)
Abstract
Most construction projects today experience changes at different phases due to a number of reasons.
Since contract recognition by two parties is diverse, communication between the parties is likely to be
destructive even though owners and contractors may gradually realize that the project changes are
reasonable. These changes may raise disputes between owners, contractors and subcontractors as
change orders can have adverse effects on project performance and are difficult to quantify and
manage. Many studies have sought methods to quantify the impact of change orders on project
performance, and to resolve construction disputes but only few exist involving Artificial Neural
Network (ANN) approaches for dispute resolution. The aim of this paper is to present an ANN Model
so that the influence of change orders can be estimated and probable disputes may be avoided or
resolved before litigation occurs. For this purpose, various factors that describe the adverse effects of
change orders on project performance have been identified from a background research. Based on the
survey conducted to the contractors in North Cyprus construction industry, an ANN Model has been
developed to manage change orders through all phases of a project such that construction operations
can continue with the least amount of interruption that usually results from of disputes between
different parties involved in a project.
Keywords: change orders, project performance, dispute resolution, artificial neural networks
1 Introduction
Most construction projects today experience changes at different phases due to a number of reasons.
Unplanned for changes in construction projects can cause additional work beyond that expected,
resulting in extra cost and time (Chivitello, 1987). Since contract recognition by two parties is diverse,
communication between the parties is likely to be harmful even though owners and contractors may
gradually realize that the project changes are reasonable. These changes may raise disputes between
owners, contractors and subcontractors as change orders can have adverse effects on project
performance and are difficult to quantify and manage. Change orders have long been identified to
have a negative impact on construction productivity, leading to a decline in labor efficiency and, in
some cases, sizeable loss of man hours. Change orders continue to pose serious challenge to owners
and contractors alike (Moselhi et al., 2005). Quantifying the impact of change orders on project
performance remains to be a challenging task, despite the reported findings of many studies and
documented cases (Moselhi et al. 1991; Thomas and Napolitan 1995; Ibbs 1997; Hanna et al. 1999a,b;
Moselhi et al., 2005; Ibbs 2005; Yitmen et al. 2006; Ibbs et al. 2007).
The ANN approach is an information processing technology based on simulating the human brain
and nervous system. It is usually applied to establish forecast models (Arditi et al., 1999). Among the
algorithms fitting the approach, the Back-propagation (BP) algorithm serves as the most
representative and practical, being an efficient approach for training multiple-layer artificial networks
An artificial neural network model for estimating the influence of
change orders on project performance and dispute resolution
Ibrahim Yitmen & Ebrahim Soujeri
European University of Lefke, North Cyprus