Convert the seasonality curve into a safety-stock recommendation. Peak-to-trough ratios above 5x drive a pre-build month and weeks-of-cover output.
Eight pool molecules cross the 5x peak-to-trough threshold and trigger Strategic Safety Stock automatically. Carfilzomib leads at 3942x peak-to-trough with a September peak and March trough, followed by tirzepatide and acalabrutinib in the 600-900x range. The output is a pre-build month, a peak month, and a weeks-of-cover target sized to the variance. Run before every annual S&OP cycle.
| # | Molecule | Peak month | PT ratio | Pre-build | Cover target |
|---|---|---|---|---|---|
| 1 | Carfilzomib | Oncology | 16 weeks of cover | Sep peak | OAI |
| 2 | Tirzepatide | GLP-1 | 14 weeks of cover | Nov peak | NAI |
| 3 | Acalabrutinib | BTK inhibitor | 14 weeks of cover | Dec peak | NAI |
| 4 | Edoxaban | Anticoagulant | 12 weeks of cover | Dec peak | VAI |
| 5 | Lenalidomide | Oncology | 10 weeks of cover | Feb peak | NAI |
A single molecule or a portfolio. The output is sized at molecule granularity but the worklist can roll up to a class or therapy area.
Peak month, trough month, and peak-to-trough ratio computed from molecule_seasonality. Lead time and target service level fold in as planning parameters.
Citation-anchored brief with the pre-build month, the peak window, and the recommended weeks-of-cover per molecule. Handoff to the S&OP system.
A flat 8-week cover policy under-protects extreme-seasonal molecules and over-funds flat ones. The sized cover follows the curve.
The AI agent names the month to start building the inventory cushion. No more guessing the lead time to the peak.
Cover sizing folds in your API lead time. A 12-week lead time on a September peak means production starts in June. The brief surfaces both.
Any molecule whose curve crosses 3x peak-to-trough triggers the safety-stock workflow. The trigger fires on every monthly refresh.
A real chat thread in PharmaDB. Type a question, the AI agent runs the tools, the answer lands as a saveable note.
pharmagraph_query Twelve pool molecules with peak-to-trough variance above the flat-policy noise band · count is recommended weeks of cover.
The safety-stock workflow is the planner-facing payload on top of the seasonality curve. PharmaDB takes the peak-to-trough ratio, folds in lead time and service level, and returns a sized worklist of pre-build months and weeks-of-cover targets. The view is molecule_seasonality; the planner-facing output is the recommended cover per molecule.
Cover = lead time + peak-to-mean ratio multiplied by months in the peak window, expressed in weeks. The default service level is 95%. Tune both the lead time and the service level on the query.
A flat policy works. The page does not recommend extra cover and the molecule falls out of the worklist. The cover saving on flat molecules typically funds the cover build on the extreme ones.
Single-year peaks are noisy. A 3942x ratio in one year can be a one-time tender; the same ratio across four years is structural. PharmaDB uses the four-year mean to filter out noise.
Yes. The same curve applies to API and FDF, with FDF often showing a different shape because of tender cycles and pack-size effects. Run both for a complete cover picture.
Seasonality forecasting returns the demand-shape curve. Safety stock targets sizes the cover decision on top: which molecules, how many weeks, starting in which month.
Trade aggregates refresh monthly. Cover sizing recomputes on every refresh. The catalog row carries the last-computed timestamp surfaced on every result page.
Bring the molecule, the lane, or the supplier you're sourcing this week. The AI agent runs it on PharmaDB in 30 minutes. You keep the brief.