Fresh questions.
Practical perspectives.
Explore short planning briefs on demand, inventory, operations and forecasting. For a deeper explanation, follow a field guide or course lesson.
Browse the resources below. Search loads when this page opens.
Multi-echelon stochastic inventory design
The advanced question in multi-echelon stochastic inventory design is how the model behaves when assumptions fail. A sound design makes uncertainty, constraints and fallback actions explicit.
Explore →ArticleDistributionally robust safety stock
A deployable approach must explain what it optimizes, what it leaves uncertain and when it should defer. Those questions guide this brief on distributionally robust safety stock.
Explore →ArticleChance-constrained service planning
Model complexity should earn its place through better decisions, not novelty. This article explores chance-constrained service planning and the baselines needed to establish its incremental value.
Explore →ArticleTwo-stage stochastic procurement
Advanced planning methods are only useful when they improve a real decision under real constraints. Here, two-stage stochastic procurement is framed as a testable system-design choice.
Explore →ArticleScenario reduction for large planning models
A method that performs well on average may still fail at the tail. This note examines scenario reduction for large planning models with attention to stress cases, feasibility and operational control.
Explore →ArticleRisk pooling under correlated demand
This technical brief treats risk pooling under correlated demand as a decision system: define its state, objective, feasible actions and evidence before comparing algorithms.
Explore →ArticleGraph neural supply imbalance prediction
The advanced question in graph neural supply imbalance prediction is how the model behaves when assumptions fail. A sound design makes uncertainty, constraints and fallback actions explicit.
Explore →ArticleTemporal knowledge graphs for supplier risk
The advanced question in temporal knowledge graphs for supplier risk is how the model behaves when assumptions fail. A sound design makes uncertainty, constraints and fallback actions explicit.
Explore →ArticleCausal graph discovery for disruptions
A deployable approach must explain what it optimizes, what it leaves uncertain and when it should defer. Those questions guide this brief on causal graph discovery for disruptions.
Explore →ArticleBayesian networks for supply risk propagation
Model complexity should earn its place through better decisions, not novelty. This article explores bayesian networks for supply risk propagation and the baselines needed to establish its incremental value.
Explore →ArticleRobust optimization with uncertain lead times
Advanced planning methods are only useful when they improve a real decision under real constraints. Here, robust optimization with uncertain lead times is framed as a testable system-design choice.
Explore →ArticleModel predictive control for replenishment
A method that performs well on average may still fail at the tail. This note examines model predictive control for replenishment with attention to stress cases, feasibility and operational control.
Explore →ArticleRolling-horizon production optimization
This technical brief treats rolling-horizon production optimization as a decision system: define its state, objective, feasible actions and evidence before comparing algorithms.
Explore →ArticleBenders decomposition for network planning
The advanced question in benders decomposition for network planning is how the model behaves when assumptions fail. A sound design makes uncertainty, constraints and fallback actions explicit.
Explore →ArticleColumn generation for transportation design
The advanced question in column generation for transportation design is how the model behaves when assumptions fail. A sound design makes uncertainty, constraints and fallback actions explicit.
Explore →ArticleLagrangian relaxation in capacity allocation
A deployable approach must explain what it optimizes, what it leaves uncertain and when it should defer. Those questions guide this brief on lagrangian relaxation in capacity allocation.
Explore →ArticleMixed-integer lot-sizing under setup costs
Model complexity should earn its place through better decisions, not novelty. This article explores mixed-integer lot-sizing under setup costs and the baselines needed to establish its incremental value.
Explore →ArticlePiecewise-linear approximation for planning
Advanced planning methods are only useful when they improve a real decision under real constraints. Here, piecewise-linear approximation for planning is framed as a testable system-design choice.
Explore →