Estimation at Completion Simulation Using the Potential of Soft Computing Models: Case Study of Construction Engineering Projects

Abstract

“Estimation at completion” (EAC) is a manager's projection of a project's total cost at its completion. It is an important tool for monitoring a project's performance and risk. Executives usually make high-level decisions on a project, but they may have gaps in the technical knowledge which may cause errors in their decisions. In this current study, the authors implemented new coupled intelligence models, namely global harmony search (GHS) and brute force (BF) integrated with extreme learning machine (ELM) for modeling the project construction estimation at completion. GHS and BF were used to abstract the substantial influential attributes toward the EAC dependent variable, whereas the effectiveness of ELM as a novel predictive model for the investigated application was demonstrated. As a benchmark model, a classical artificial neural network (ANN) was developed to validate the new ELM model in terms of the prediction accuracy. The predictive models were applied using historical information related to construction projects gathered from the United Arab Emirates (UAE). The study investigated the application of the proposed coupled model in determining the EAC and calculated the tendency of a change in the forecast model monitor. The main goal of the investigated model was to produce a reliable trend of EAC estimates which can aid project managers in improving the effectiveness of project costs control. The results demonstrated a noticeable implementation of the GHS-ELM and BF-ELM over the classical and hybridized ANN models.

Artificial Intelligence Models for Predicting Budget Expenditures

“…Table 1 shows that the values of all three index calculations differ, confirming their purpose. For example, we discuss Equation (1), which shows the best results. Still, this Equation only considers the project's current cost, summing it up with the remaining cost of the work until completion.…”

“…A study by Alhares et al (2019) examines the application of coupled intelligent models to improve project completion forecasting in the construction industry. The researchers used a twostep approach, combining global harmony and brute force algorithms with an extreme learning machine to make more reliable predictions.…”

Improving Project Estimates at Completion through Progress-Based Performance Factors

“…Several reviews of AI applications in project monitoring are available, including [41][42][43]. Algorithms include linear regression [44], support vector machine [45][46][47], tree-based methods [48,49], k-nearest neighbors [50], ensemble methods [51,52], and artificial neural networks [53][54][55][56][57][58][59]. The difficulty in using AI models lies in collecting the data and in the procedures required to prevent underfitting and overfitting.…”

Comprehensive Project Management Framework using Machine Learning

“…However, if the project management models are implemented with machine learning models for predicting the on-going conditions and its relative impact on project schedule, it can help in more accurate planning and course correction measures as required. In [12], the study has focused upon usage of the questionnaire-based data garnered about delays in the construction project of a company and used it as training dataset for a machine learning model. Using the WEKA software, the other project related delays for the same company are tested, which indicates a potential and accurate model for development.…”