Abstract:Purpose The construction projects are highly subjected to uncertainties, which result in overruns in time and cost. Realistic estimates of workforce and duration are imperative for construction projects to attain their intended objectives. The aim of this study is to provide accurate labor and duration estimates for the construction projects, considering actual uncertainties. Design/methodology/approach The dataset was formulated from the information collected from 186 construction projects through direct interviews, group discussions and questionnaire methods. The actual uncertainties and exposure conditions of construction activities were recorded. The data were verified with the standard guideline to remove the outliers. The prediction model was developed using support vector regression (SVR), a machine learning (ML) method. The performance was evaluated using the widely adopted regression metrics. Further, the cross validation was made with the visualization of residuals and predicted errors, ridge regression with transformed target distribution and a Gaussian Naive Bayes (NB) regressor. Findings The prediction models predicted the duration and labor requirements with the consideration of actual uncertainties. The residual plot indicated the appropriate use of SVR to develop the prediction model. The duration (DC) and resource constraint (RC) prediction models obtained 80 and 82% accuracy, respectively. Besides, the developed model obtained better accuracy for the training and test scores than the Gaussian NB regressor. Further, the range of the explained variance score and R 2 was from 0.95 to 0.97, indicating better efficiency compared with other prediction models. Research limitations/implications The researchers will utilize the research findings to estimate the duration and labor requirements under uncertain conditions and further improve the construction project management practices. Practical implications The research findings will enable industry practitioners to accurately estimate the duration and labor requirements, considering historical uncertain conditions. A precise estimation of resources will ensure the attainment of the intended project outcomes. Social implications Delays in construction projects will be reduced by implementing the research findings, which significantly ensures the effective utilization of resources and attainment of other economic benefits. The policymakers will develop a guideline to develop a database to collect the uncertainties of the construction projects and relatively estimate the resource requirements. Originality/value This is the first study to consider the actual uncertainties of construction projects to develop RC and DC prediction models. The developed prediction models accurately estimate the duration and labor requirements with minimal computational time. The industry practitioners will be able to accurately estimate the duration and labor requirements using the developed models.

Strategies for Imputing Missing Values and Removing Outliers in the Dataset for Machine Learning-Based Construction Cost Prediction

Predicting ENR Construction Cost Index Using Machine-Learning Algorithms

Development of a Cost Prediction Model for Design Changes: Case of Korean Apartment Housing Projects

Predicting Enr'S Construction Cost Index Using The Modified K Nearest Neighbors (Knn) Algorithm

Application of machine learning techniques for predicting the consequences of construction accidents in China

A Survey of Data-Driven Construction Materials Price Forecasting

Prediction of construction accident outcomes based on an imbalanced dataset through integrated resampling techniques and machine learning methods

Software cost estimation with incomplete data

Comparison of Machine Learning Approaches for Medium-to-Long-Term Financial Distress Predictions in the Construction Industry

Machine Learning Algorithms for Construction Projects Delay Risk Prediction

A Duration Prediction Using a Material-Based Progress Management Methodology for Construction Operation Plans

Enhancing safety of construction workers in Korea: an integrated text mining and machine learning framework for predicting accident types

Indexing Cost Sensitive Prediction

Predicting the construction projects time and cost overruns using K-nearest neighbor and artificial neural network: a case study from Jordan

Predicting construction project compliance with machine learning model: case study using Portuguese procurement data

Imputation for prediction: beware of diminishing returns

Missing values handling for machine learning portfolios

On the Relation between Prediction and Imputation Accuracy under Missing Covariates

Duration and resource constraint prediction models for construction projects using regression machine learning method

A hybrid machine learning approach for early cost estimation of pile foundations

Locally linear reconstruction based missing value imputation for supervised learning