What If Your Local Kirana Store Could Predict Tomorrow's Demand? Meet DukaanDost

Description and use

DukaanDost turns a small shop’s billing CSV into demand forecasts and restocking recommendations, enabling better inventory decisions.

Problem

Shop owners rely on billing exports that show past sales but not guidance on what to buy next, in which pack size, or how to avoid stockouts.

Outcome

The shop owner used the forecast to restock candles and oils, reducing guesswork and aligning inventory with upcoming festivals.

Workflow

• Export billing CSV from shop system
• Preview and map columns, identify bad rows
• Generate features: lags, rolling pace, calendar
• Run TabPFN v2 for demand forecast (fallback to baseline if unavailable)
• Calculate reorder quantity: lead‑time demand + safety stock, rounded to pack size
• Generate explanation via Gemma or fallback template
• Export inventory, forecasts, recommendations as CSV
• Use output to place restocking orders

Evidence

DukaanDost takes a billing CSV and writes item-level demand, a reorder quantity, and a short explanation.

Source

He used it. He read the forecast against his sales and the festivals coming up, and he restocked candles and oils, because the prediction put those items in demand.

Source

TabPFN reads the billing table, and Gemma writes the explanation.

Source