Estimating labor resource requirements in construction projects using machine learning
Bibliographic record
Abstract
Purpose Existing labor estimation models typically consider only certain construction project types or specific influencing factors. These models are focused on quantifying the total labor hours required, while the utilization rate of the labor during the project is not usually accounted for. This study aims to develop a novel machine learning model to predict the time series of labor resource utilization rate at the work package level. Design/methodology/approach More than 250 construction work packages collected over a two-year period are used to identify the main contributing factors affecting labor resource requirements. Also, a novel machine learning algorithm – Recurrent Neural Network (RNN) – is adopted to develop a forecasting model that can predict the utilization of labor resources over time. Findings This paper presents a robust machine learning approach for predicting labor resources’ utilization rates in construction projects based on the identified contributing factors. The machine learning approach is found to result in a reliable time series forecasting model that uses the RNN algorithm. The proposed model indicates the capability of machine learning algorithms in facilitating the traditional challenges in construction industry. Originality/value The findings point to the suitability of state-of-the-art machine learning techniques for developing predictive models to forecast the utilization rate of labor resources in construction projects, as well as for supporting project managers by providing forecasting tool for labor estimations at the work package level before detailed activity schedules have been generated. Accordingly, the proposed approach facilitates resource allocation and enables prioritization of available resources to enhance the overall performance of projects.
Fetched live from OpenAlex and de-inverted. Abstracts are not stored in this database: the inverted indexes are 8.6 GB of the frame’s 9.3 GB of text, and the host has 13 GB free.
How this classification was reachedexpand
Full frame machine prediction
Teacher imitationNot calibrated prevalence, not ground truth. Human validation pending. The Gemma side is a direct model label for every work in the frame, read from the title-only record. The Codex side is a classifier learned from the 10,348 direct Codex labels and calibrated to design-weighted sample rates; fields without enough sample support carry no Codex call. Candidate is the union of the two sides; consensus is their intersection. These outputs are machine_predicted_unvalidated and are not human labels.
Distilled classifier scores by category (both heads)
| Category | Codex | Gemma |
|---|---|---|
| Metaresearch | 0.001 | 0.006 |
| Meta-epidemiology (narrow) | 0.001 | 0.000 |
| Meta-epidemiology (broad) | 0.000 | 0.000 |
| Bibliometrics | 0.002 | 0.002 |
| Science and technology studies | 0.000 | 0.000 |
| Scholarly communication | 0.001 | 0.001 |
| Open science | 0.001 | 0.000 |
| Research integrity | 0.000 | 0.000 |
| Insufficient payload (model declined to judge) | 0.001 | 0.000 |
Machine scores (provisional)
The two teacher heads of the student model, read on this work. A score orders the frame for review; it never asserts a category, and the validation status ships verbatim with every row.
Baseline scores from an immature model (maturity gate not passed, 7 training rounds). Scores rank; they never assert a category.
score_only:v0-immature-baseline · verbatim from the scoring run: score_only means the number may rank works, and no category label ships from itClassification
machine, unvalidatedMachine predicted; a candidate call from one source (direct Gemma or distilled Codex), not a consensus.
How this classification was reached, model by model and score by score, is at the end of the page under "How this classification was reached".