A two stage statistical framework for cold start spare part demand forecasting
Sendhil Nathan B, Veera Siva Reddy B, Chandrasekhara Sastry C, Sachin Salunkhe, Robert Cep, Santi Jitpichitchai · 2026
Read A two stage statistical framework for cold start spare part demand forecasting by Sendhil Nathan B, Veera Siva Reddy B, Chandrasekhara Sastry C, Sachin Salunkhe, Robert Cep, Santi Jitpichitchai (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/PMC13271490.txt.
Description
Sendhil Nathan B is lead author of a 2026 journal article on forecasting demand for spare parts that have never been ordered before, the cold-start problem. Such demand is sparse, often zero, and has no history, so standard forecasting tools that lean on past data fail badly.
The authors propose a two-stage framework. First it predicts whether demand will occur at all. Then it predicts its size with a distribution suited to skewed amounts. Finally it simulates many futures to give a full probability picture. On an industrial data set, with strict testing on unseen parts, their method beat strong single-stage and deep learning rivals. It cut point forecast error by 6.4 percent over the best baseline and scored far better on probability accuracy.
Catalog record
| CREATOR | Sendhil Nathan B, Veera Siva Reddy B, Chandrasekhara Sastry C, Sachin Salunkhe, Robert Cep, Santi Jitpichitchai |
|---|---|
| YEAR | 2026 |
| COLLECTION | PAPERS |
| FORMAT | TXT |
| FILE SIZE | 104509 |
| LICENSE | https://creativecommons.org/licenses/by/4.0 |
| RIGHTS | CC BY |
| SOURCE | Europe PMC |