Interactive demo
Two demos, two architectures
One runs entirely in your browser because it can. The other runs on a server because it has to. The contrast is the point.
1 · CPIM formula engine — client-side
27 formulas · 7 modules
Deterministic planning maths — EOQ, safety stock, MRP netting, forecast error, capacity. Every result is computed in your browser from@aliqizalbash/cpim-formulas, the same package that has its own unit-test suite. Nothing is sent anywhere.
Economic Order Quantity (EOQ)
The order size that minimises the sum of annual ordering cost and annual carrying cost, assuming steady demand.
EOQ = √(2 · D · S / H)
Result
774.6units
- Orders per year
- 15.49 orders/yr
- Total annual cost
- 2,323.79 $
- Cycle stock (avg)
- 387.3 units
Order about 775 units at a time. At the EOQ, annual ordering cost and annual carrying cost are equal.
2 · Monte-Carlo safety-stock simulator — Python, hosted
This one draws tens of thousands of random demand and lead-time samples, runs a full inventory simulation for each, and reports the distribution of stockouts and fill rate against a chosen reorder policy. That's a genuine numerical workload with real statistical libraries behind it — numpy, scipy — not something to fake with a bit of client-side JavaScript.
So it isn't another widget on this page. It's a standaloneStreamlit app in its own Docker container, on its own subdomain, reverse-proxied by the same web server. That separation is exactly the pattern this site is built to support — Python apps living alongside the static site, each isolated and independently deployable.