Machine learning prediction model for the sewing thread consumption
Fayyaz Ahmad, Sheheryar Mohsin Qureshi, Ayesha Younas, Muhammad Babar Ramzan, Gulam Fareed · 2026
Read Machine learning prediction model for the sewing thread consumption by Fayyaz Ahmad, Sheheryar Mohsin Qureshi, Ayesha Younas, Muhammad Babar Ramzan, Gulam Fareed (2026) in the Tower of Babel, a free online library of public-domain works preserved in clean, searchable text. No sign-up. The full text is served as a plain-text file at a predictable URL, where text is present: https://antarctic-labs.com/tower-of-babel/plos-pone-0356367.txt.
Description
Fayyaz Ahmad and four colleagues apply machine learning to a humble factory problem in this 2026 study: predicting how much sewing thread a garment will consume. Ordering too much thread wastes money; too little stalls production. The team focused on a common industrial stitch class. They compared six algorithms, including tree ensembles, support vector machines, and a neural network, on predicting consumption from simple inputs such as the number of fabric layers and stitch length.
An ensemble method called Extra Trees won clearly, predicting thread use with an average error of only a few centimeters and explaining the great bulk of the variation. Repeated cross-validation showed the model was stable rather than overfitted. The authors propose their prediction formula as a ready tool for garment manufacturers who want to order precisely and cut stock waste.
Catalog record
| CREATOR | Fayyaz Ahmad, Sheheryar Mohsin Qureshi, Ayesha Younas, Muhammad Babar Ramzan, Gulam Fareed |
|---|---|
| YEAR | 2026 |
| COLLECTION | PAPERS |
| FORMAT | TXT |
| FILE SIZE | 30996 |
| LICENSE | CC_BY |
| RIGHTS | CC BY |
| SOURCE | PLOS |