Machine Learning-Based Control of Dual-Sourcing Inventory Systems
Bibliographic record
Abstract
We examine a data-driven approach to dual-sourcing inventory systems under periodic review with uncertain demand, utilizing historical data. Different from conventional fixed-ordering policies, we advocate for a machine learning methodology to capture the dynamic behaviour of demand. We train machine learning models to predict demand for single-period and multi-period horizons. Subsequently, we utilize these predictions to manage and mitigate supply chain risk by ordering from two suppliers. One supplier, identified as the regular supplier, offers lower unit prices but longer delivery times. The other is the emergency supplier, which provides same-day delivery at a higher unit price. Through an extensive numerical study utilizing actual data from Rossmann drug stores and various machine learning models, we systematically assess the performance of prediction models. Further, we enhance the learning process of the machine learning models by incorporating inventory cost parameters into the training objective function. Specifically, we customize the learning objective, switching from mean squared error (i.e., non-guided models) to a single-period inventory cost function (i.e., guided models). This modification results in significant cost reduction in the dual-sourcing inventory system. Guiding the models offers robust data-driven solutions for complex sequential systems where fully integrating learning and decision-making is not feasible.
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 | 0.003 | 0.006 |
| Meta-epidemiology (narrow) | 0.001 | 0.001 |
| Meta-epidemiology (broad) | 0.001 | 0.001 |
| Bibliometrics | 0.001 | 0.001 |
| Science and technology studies | 0.001 | 0.001 |
| Scholarly communication | 0.002 | 0.001 |
| Open science | 0.002 | 0.001 |
| Research integrity | 0.001 | 0.002 |
| Insufficient payload (model declined to judge) | 0.002 | 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".