Predicting Project’s Uncertainty Risk in the Bidding Process by Integrating Unstructured Text Data and Structured Numerical Data Using Text Mining

Abstract

Abstract:The construction bidding process takes place in the early stages of a construction project. The bidding process involves many uncertainties since the construction risk factors are predicted and reflected in the bid price before construction commences. Therefore, bidders should thoroughly understand the uncertainty factors of the project before make bidding decisions. The uncertainty risk of construction projects is determined by the content that is contained in the bidding document. If the information provided in the bidding document is not accurate and is unclear, the uncertainty of the projects increases. Thus, this study is performed to predict risks in the bidding process of construction projects by analyzing the uncertainty of the bidding document and using it as factors to predict a project's bidding risk. To achieve this, bidding risk prediction modeling was conducted using the pre-bid clarification information of each project. In addition, text mining on pre-bid RFI documents, which are in an unstructured text data format, was performed and the results of text mining were used as major influencing factors for the risk prediction models. As a result, the accuracy of the risk prediction model including text data was improved (72.92%) when compared to the prediction model using only numeric data (52.08%). The results of this study are expected to strengthen the possibility of further similar studies in the future since it enhances the predictive accuracy by incorporating the uncertainty of the bidding document, which is rarely considered in previous studies.

Using Text Mining to Estimate Schedule Delay Risk of 13 Offshore Oil and Gas EPC Case Studies During the Bidding Process

“…Arif-Uz-Zaman et al [32] built a keyword dictionary from work orders and downtime reports to analyze the failure time of industrial production plants. Lee and Yi [33] conducted a study to predict the risk involved in the bidding process of a construction project. They analyzed the bidding documents to assess the most important or risk-intense documentation.…”

“…These two indexes are typically used by the project owner and EPC contractor to quantify the level of basic and detailed engineering maturity to determine whether the project can move to the next development stage by mapping PDRI scores and DECRIS score to project performances of schedule delay and cost overrun in the EPC execution stage. Finally, a recent study by Lee and Yi [33] used a similar approach to the SDRI model for the identification and assessment of risk and uncertainty in bidding documents using TDM and NLP.…”

“…The regression model was estimated under the basic assumption of residuals having the characteristics of normality, independence, and homoscedasticity, which was all verified with following methods [33]. Concerning normality, the central limit theorem states that given a sufficiently large sample size the distribution of means will approach normality.…”

“…For example, the term "escape tunnel" likely has a greater impact on the schedule than "nozzle". While this remains a limitation, using count data to develop risk scoring has been used historically in the literature [20,21,33,44].…”

Analysis of Document Pre-Processing Effects in Text and Opinion Mining

“…For example, the classification rate can be employed to evaluate the effect of pre-processing on the accuracy of the classification of texts and opinions [4]. It is important to note that in this paper, we focus on texts and opinions because they are tasks widely discussed in the literature, but there are other subjects that would be benefited by the proposed approach, such as risk prediction modeling [5], patent analysis [6] and sales reports [7].…”