The business problem
Predictive analytics tells you what's likely to happen. Prescriptive analytics — optimization — tells you what to do about it. Flexport runs huge optimization problems every day: which container goes on which ship, which route minimizes cost given fuel and time-of-arrival constraints, how to rebalance equipment when trade flows go out of balance.
This case introduces optimization at the level a business analyst needs — formulating the problem, identifying the constraints, and using a solver. It's slightly outside the core seven decisions, but it's where many real "data science" jobs in supply chain actually live.
This case is harder than the others
Optimization is its own discipline with its own vocabulary (objective function, decision variables, constraints, slack). Don't expect to feel as fluent here as in classification. Take it as an invitation to a follow-on course or a self-study path.
Case kit
Everything you need
- Flexport Optimization Problems — ReadingBackground and problem statements for the case.
- Flexport NotesCompanion notes covering the optimization concepts you'll need.
- Excel Optimization TemplatesPre-built spreadsheets for using Excel Solver on these problems.
- Python Optimization NotebookSame problems solved with PuLP / scipy.optimize for those who want the code path.
Topics you'll be applying
- Decision 1 — Data Prep & EDA — formulating the optimization input data
- Decision 2 — Regression — predicted demand often becomes the input to optimization
- Decision 7 — Time Series — same — forecasts feed multi-period optimization