Data-driven cycle time prediction of fitting and welding stations in steel fabrication
Bibliographic record
Abstract
The construction industry's lack of materials, resources, and financial assets streamlined a shift toward using digital lean principles to obtain precise management over the limited resources. Steel fabrication companies rely heavily upon the enormous equipment to get promising results. However, implementing lean principles in the fabrication process is not straightforward due to the non-repetitive nature of steel construction products. Hence, the time-based modeling for such a process lacks accuracy and reliability, especially for manual steel fabrication processes. Accordingly, the current study aims to achieve a practical and accurate estimation of fabrication time aspects. This study targets modeling manual steel fabrication processes (fitting and welding workstations) in terms of processing times (cycle time and value-added time). The proposed approach builds a machine learning (ML) model to estimate the identified processing time aspects. For performance assessment, the typical correlation analysis and linear regression (LR) approach was used as a benchmark to quantify the ML model's pros and cons in terms of practicality and accuracy. The required data source for this study is a steel fabrication industry partner. The results of this study show ML superiority in accuracy over LR processing time predictive models, particularly when predictive parameters increase ML presents a 13.2 % improvement in mean squared error compared to the LR predictive model. LR models need fewer data and are not computationally expensive like ML models, making them more practical. Additionally, the study introduces a precise and practical time estimation approach. Such an approach provides precious input for simulation models which support evidence-based decisions and benefits quantification of plans.
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 | 0.004 |
| Meta-epidemiology (narrow) | 0.001 | 0.000 |
| Meta-epidemiology (broad) | 0.000 | 0.001 |
| Bibliometrics | 0.001 | 0.001 |
| Science and technology studies | 0.000 | 0.000 |
| Scholarly communication | 0.001 | 0.001 |
| Open science | 0.001 | 0.000 |
| Research integrity | 0.001 | 0.001 |
| 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".