The Value of Flexibility in Robust Supply Chain Network Design
Bibliographic record
Abstract
A supply chain network design problem (SCNDP) involves making long-term, irreversible strategic decisions whose cost efficiency depends on effectively leveraging flexibility and demand information. After analyzing the interaction of these factors, we propose five distinct policies for addressing the SCNDP. Starting with a localized production model where demand is satisfied locally, the study extends to scenarios where production capacity at one location serves other nodes (Policy II). Further flexibility is introduced by enabling capacity sharing among facilities with prearranged links (Policy III). Unlike these policies, which optimize capacities as proxies for production decisions prior to demand realization, Policies IV and V defer production decisions until demand is realized. The robustness and resilience of these policies are evaluated under varying levels of demand uncertainty and risks of supply disruptions. To provide actionable insights, we develop robust two-stage optimization frameworks for the proposed policies and design efficient methods to address varying uncertainty budgets for both supply and demand risks. Our results reveal that under demand uncertainty alone: (i) Capacity-sharing links among facilities yield the highest cost savings across all uncertainty budgets due to the pooling effect and significantly reduce shortage probability (Policy III), and (ii) production postponement offers only marginal benefits, highlighting the greater importance of upstream capacity-sharing over postponing production, particularly for moderate uncertainty budgets. Under simultaneous demand and supply risks, we demonstrate that (iii) capacity-sharing retains its value, while flexibility from production postponement deteriorates performance, favoring partially flexible networks with only capacity-sharing links over fully flexible ones with both sources of flexibility. To contextualize these findings, we apply the most effective policies to a real-world case study, quantifying their impact and providing design recommendations. Finally, we extend our models to an event-wise ambiguity set and demonstrate that by (iv) leveraging the supermodular structure of the second-stage problems, one can solve instances up to eight times larger.
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 distilled prediction
Teacher imitationNot calibrated prevalence, not ground truth. Human validation pending. Learned from the 10,348 direct Codex labels and 10,348 direct Gemma labels. Candidate is the union of thresholded teacher heads; consensus is their intersection. These outputs are machine_predicted_unvalidated and are not human labels or direct frontier model labels.
Codex and Gemma teacher scores by category
| Category | Codex | Gemma |
|---|---|---|
| Metaresearch | 0.001 | 0.000 |
| Meta-epidemiology (narrow) | 0.000 | 0.000 |
| Meta-epidemiology (broad) | 0.000 | 0.000 |
| Bibliometrics | 0.000 | 0.001 |
| Science and technology studies | 0.001 | 0.000 |
| Scholarly communication | 0.000 | 0.000 |
| Open science | 0.000 | 0.000 |
| Research integrity | 0.000 | 0.000 |
| Insufficient payload (model declined to judge) | 0.000 | 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 teacher head, 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".