Predicting the Outcome of Construction Litigation Using an Integrated Artificial Intelligence Model

Abstract

An integrated prediction model (IPM) was used to predict the outcome of construction litigation with the hope that it will generate a higher prediction rate than the rates obtained by using artificial neural networks (ANN), case-based reasoning (CBR), and boosted decision trees (BDTs) in earlier studies. These studies were conducted by using the same 132 Illinois circuit court cases (with slight variations in the case of the BDT and IPM studies) filed between 1992 and 2000. A prediction rate of 91.15% was obtained, higher than the 66.67% in the ANN study, 83.33% in the CBR study, and 89.59% in the BDT study. IPM involves four processes, namely, data consolidation, attribute selection, prediction using hybrid classifiers, and assessment. The performance of IPM is compared with the performance of ANN, CBR, and BDT. If the outcome of construction litigation can be predicted with reasonable accuracy and reliability, all parties involved in the construction process could save considerable money and time.