DELAY CLAIMS - A BIM AND TEXT MINING APPROACH

Abstract

The rising complexity of current construction projects led by an increasing demand from project owners to implement fast-track programmes has led to a surge in the number of claims and disputes. The significant rise in Construction records and data resulted in Claimants and Defendants to experience difficulties to provide credible evidence to substantiate their claims. The aim of this thesis proposes a two-fold process to enhance the delay claim process by introducing; 1) a new methodology for the automatic text classification of project delay claims documents that utilize the activity and Work Breakdown Structure keywords of a given path of a delay event (DE) to train and further predict unlabeled project documents, where Multinomial-Naïve Bayes (MNB) Classification is selected as the supervised learning algorithm; and 2) develop an algorithm to link the delay event related 4DBIM objects with the respective classified documents by extending the non-proprietary Industry Foundation Class (IFC) Schema of a dynamic property set to include delay-related attributes. The proposed two-fold methodology has been implemented on a series of delay claims events in a project; the implementation of the two-stage methodology enhanced the overall performance and efficiency of the delay claim assessment process