Data-driven fuzzy information granulation for predicting freight volume trends
Yunbo Gao, Xinyu Wang, Ming Niu, Jianguo Li, Liping Cai, Rui Li · 2026
Read Data-driven fuzzy information granulation for predicting freight volume trends by Yunbo Gao, Xinyu Wang, Ming Niu, Jianguo Li, Liping Cai, Rui Li (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-0348239.txt.
Description
Yunbo Gao and co-workers build a forecasting method for rail freight volumes that admits uncertainty instead of hiding it. Freight data are fuzzy and nonlinear, and point forecasts often miss badly. Their 2026 study first compresses nine years of monthly records from a Chinese freight center into low, central, and high fuzzy bands, then trains a tuned support vector model on each band.
The hybrid method posted the smallest errors of the models compared, for every band. It also drew the narrowest honest forecast interval around future volumes. An improved swarm search kept the training from settling too early on weak settings. Rail planners gain not just a number but a believable range for staffing, wagons, and warehouse space.
Catalog record
| CREATOR | Yunbo Gao, Xinyu Wang, Ming Niu, Jianguo Li, Liping Cai, Rui Li |
|---|---|
| YEAR | 2026 |
| COLLECTION | PAPERS |
| FORMAT | TXT |
| FILE SIZE | 44590 |
| LICENSE | CC_BY |
| RIGHTS | CC BY |
| SOURCE | PLOS |