Stock-outs announce themselves weeks in advance — in the consumption trends, open orders and supplier lead times your system already holds. This demo shows TEKAI reading those signals in live SAP Business One data and naming the items that will short, and when, while there is still time to act.
Reorder points are static; demand is not. The item that doubled its run rate last quarter still carries last year’s reorder level, so the system stays quiet right up until the moment the first person to notice the problem is the one who cannot ship. The prediction problem was never data scarcity — SAP Business One holds the consumption history, the open sales orders, the open purchase orders and the supplier lead times. It is that nobody reads them together. Each lives in its own report, and the shortage only becomes visible when someone lines all four up for the same item — which, across a full item master, nobody does by hand. The cost of the miss is asymmetric, too: the stock-out costs expedited freight, broken delivery promises and sometimes the customer, while noticing three weeks earlier costs one purchase order raised on time. Static parameters cut both ways as well — the item whose demand faded still carries a reorder point sized for its glory days, quietly overstocking one shelf while another empties.
On a demonstration database: TEKAI projects forward from consumption patterns against current stock, incoming supply and lead times, and returns the shortage list — item by item, with the projected short date and the driving factor named: a demand spike, a delayed PO, lead-time drift on a supplier. Follow-up questions go item-deep, the way a planning conversation actually runs: which sales orders are exposed if this item shorts, which open PO would close the gap if expedited. The list arrives ranked and readable — a watchlist for stores and purchasing, not a model output that needs a data scientist to interpret. Every figure in the demo comes from a demonstration database, not customer data; on your own system the watchlist is built from your own history.
A forecast that ends at “these items look risky” just moves the anxiety earlier. Each flagged item in the demo carries its lever: expedite the open PO that is already running late; raise the order that the run rate now justifies; rebalance stock from another warehouse that holds more than it needs; or — when the shortage cannot be prevented — inform the customer early, which is the cheapest of all mitigations when it is done in advance instead of at the shipping dock. The AI names the item, the date and the option; stores and purchasing decide. The driving factor earns its place here too: when the same supplier’s lead-time drift keeps appearing behind flag after flag, that is a conversation to have with the supplier — with data in hand. The watchlist replaces the surprise, and the decision stays human.
SAP Business One (SQL or HANA) with sales and purchase history · TEKAI at ₹5,000/company/month India, $150 global · your own AI subscription (Claude, ChatGPT or Perplexity) · permission-scoped access, nothing trained on your data (/tekai/security/). If the history is already in SAP Business One, the model has what it needs from day one — no data project comes first.
As far as your lead times and demand patterns support — the demo works on the horizon your own data justifies, stated with its confidence.
Patterns in your history, including seasonal ones, are exactly what it reads.
MRP plans from parameters; this reads reality against the plan and flags where they diverge.
Yes — shortage flags ride the same WhatsApp exception rail as other TEKAI alerts.
Your existing SAP Business One history — no new data capture. The five-minute Fit Test is the quickest way to check the fit before connecting.
New to the category? Start with the overview: AI for SAP Business One, explained.