AI demand forecasting for Tet and peak season in F&B
By LOOP Editorial
Last updated:
AI demand forecasting for Tet and peak season in F&B
Tet is the single most expensive forecasting mistake a Vietnamese F&B operator can make. Over-prep and you write off raw beef on Feb 14. Under-prep and you 86 your bestseller on the highest-traffic night of the year. The good news: this is exactly the problem AI demand forecasting was built for.
Why traditional forecasting breaks at Tet
Most chains forecast Tet by taking last year''s number and multiplying by a growth factor. That works in a normal week. It catastrophically fails when:
- Tet falls on a different solar date each year (the lunar calendar shifts demand by up to 3 weeks).
- A new branch opened mid-2025, so last-year baseline is missing.
- A nearby competitor closed (or opened), changing your draw radius.
- Weather forecast shifts a rainy day into the long weekend.
Excel models cannot ingest these signals. Traditional ERP forecasting modules can, in theory, but require a data scientist to maintain. AI forecasting closes that gap.
What an AI forecast actually does
A modern forecasting model — we use a gradient-boosted ensemble plus a small transformer for sequence effects — eats five inputs:
- Per-SKU historical sales (3+ years, hourly granularity)
- Calendar features (lunar date, school holiday, payday, weather)
- Local events (concerts, sports, public holidays — scraped from city event feeds)
- Branch-level traffic (foot traffic from in-store sensors, delivery aggregator order volume)
- Promotion and pricing history (your own and competitors'')
Output: a per-SKU, per-hour demand forecast for the next 14 days, with confidence intervals. Operators see a dashboard. The POS uses the numbers to auto-suggest prep quantities and trigger purchase orders.
McKinsey''s research on AI in food retail puts the typical waste reduction at 20–30% when forecasting moves from rule-based to AI.
A real Tet 2025 example
One of our coffee chain customers in HCMC has 14 branches. In Tet 2024 they pre-ordered 800kg of arabica based on last-year-times-1.2. They ended Tet with 180kg of stale beans (a $4,200 write-off) and ran out at three branches on Mùng 3.
For Tet 2025 we ran the LOOP forecaster against:
- Their own 2022-2024 hourly sales
- Lunar calendar phase (Tet fell 11 days earlier than 2024)
- Local data: a major district 1 mall opened a new food court on Jan 15
- Weather: forecast cold snap on Mùng 1-2 (lower iced drink demand)
The model predicted 720kg total, with 60% concentrated in the four days Mùng 4–7 (post-family-visit window when people resume social outings). Branch-level allocation was uneven — district 2 needed 30% less, district 1 needed 25% more.
Result: 78kg leftover (vs 180kg the year before), zero 86s, $5,800 net margin improvement vs prior year. See Statista''s coffee market data for Vietnam for context on the underlying market growth.
What to look for in an AI forecasting feature
- Per-SKU, not per-category. Aggregated forecasts mask the items that actually waste.
- Branch-level allocation. Network-wide forecasts are useless if you cannot split them.
- Confidence intervals. A point estimate hides the risk. A 90% interval tells the chef whether to prep aggressively or conservatively.
- Auto-PO integration. The forecast is only useful if it triggers a purchase order, not just a chart.
- Override + learn. The chef has context the model lacks. Let her override, then learn from the override.
How LOOP handles it
LOOP forecasts every published SKU at every branch every night at 2am local time. The result feeds:
- Tomorrow''s prep checklist (printed at open)
- The auto-PO engine (orders cut to suppliers by 9am)
- The voice query layer ("hey Loop, what''s my forecast for tomorrow lunch?")
We backtest every model weekly against the prior 14 days and surface MAPE in the admin dashboard. Anything above 18% triggers a review.
FAQ
How much history do I need? 12 months minimum, 24+ months recommended. New branches inherit a transferred model from the closest demographic match for the first 90 days.
What if I run a flash promotion? The promotion calendar feeds the model. A 24-hour heads-up is enough to re-forecast.
Does it work for a single branch? Yes. Single-branch forecasts are actually easier — less variance to disentangle.
Can it forecast new menu items? Yes, using a similarity model against existing items + the launch promotion plan. Accuracy is lower (~70%) until 4 weeks of real data come in.
Related reading
- AI forecasting for Tet peak demand: the 6-week playbook for Vietnamese F&B
- AI fraud detection at the POS: voids, refunds, ghost orders
- AI fraud detection for voids and refunds on POS: catching ₫30–80M/month leaks
Why this matters in 2026
Multi-outlet F&B operators across Vietnam and Southeast Asia are running into the same wall in 2026: aggregator commissions compress margins, food-cost drift compounds across outlets, labour cost climbs faster than ticket size, and a traditional POS only surfaces the damage at month-end when the only response left is firefighting. Operators who win in 2026 close the loop in hours, not weeks — variance flags before the next shift, demand forecasts before purchasing, daypart promos drafted automatically for slow slots, and a single morning brief instead of five dashboards. That is the bar this guide is written against, and the reason LOOP exists. The cost of a missed signal is no longer a single bad week — it is the difference between a chain that compounds outlet-level profitability and a chain that opens new outlets to mask the leaks at the old ones.
The SEA F&B operator landscape in 2026 also looks materially different from 2023. Aggregator commissions in Vietnam have settled in the 22–28% band; Thailand and the Philippines run higher, Singapore lower. Labour minimums have moved twice in eighteen months in Vietnam. E-invoice (TT78) is now non-negotiable and enforced. Loyalty has shifted from punch cards to messaging-native (Zalo OA, LINE, WhatsApp, Messenger) — and the chains that ride that shift are seeing repeat visits double inside ninety days. None of that lands as an upgrade on a legacy POS; it lands as a different operating model.
SEA benchmarks (2026)
- Median food cost across SEA QSR chains: 30–34% in 2026.
- Median labour cost across SEA F&B chains: 22–28% in 2026.
- Repeat-visit rate for loyalty-enabled cafés: 38–46% in 2026.
- Average ticket time for SEA QSR in peak: 6.8–9.2 minutes in 2026.
- Aggregator commission band in VN: 22–28% per order in 2026.
- AI demand forecast MAPE on LOOP cohorts: 14–22% per outlet in 2026.
- VAT e-invoice (TT78) compliance among LOOP outlets: 100% by 2026.
- Average POS uptime LOOP cohorts: 99.92% rolling-90-day in 2026.
Operator playbook — first 30 days on LOOP
Week 1 — Foundations. Import menu, recipes, modifiers, customers, loyalty balances and 24 months of sales via CSV. Connect aggregators (GrabFood, ShopeeFood, Be, foodpanda, Gojek). Configure e-invoice provider (MISA / Viettel / VNPT). Confirm payment rails (VietQR for VN; PromptPay / QRIS / DuitNow / PayNow / QR Ph for the rest of SEA). Train two staff per outlet on voice and text commands; the rest pick it up by observation in days 4–7.
Week 2 — Variance and forecast online. Switch demand forecasting on at daypart level. Set variance alert thresholds (default: food-cost ±3pp, labour ±2pp, void rate ±0.5pp). Let the system run a full week without intervention so the baseline calibrates. Review the morning brief each day; ignore the urge to override — by day 10 the forecast typically holds within MAPE 18% and stays there.
Week 3 — Promo and loyalty loop. Turn on daypart promo drafting for the two slowest hours per outlet. Connect Zalo OA / LINE / WhatsApp for delivery; start with a single segment (e.g. lapsed-30-day) and a single offer. Measure incremental visits, not coupon redemptions.
Week 4 — Compound. Roll the same flow to a second outlet, then a third. The operating model is the same at outlet 2 as outlet 20 — that is the point of LOOP.
KPI table — what to watch
| KPI | Target band 2026 | LOOP signal |
|---|---|---|
| Food cost % | 30–34% (QSR), 27–32% (café) | Variance alert within 6 hours of shift close |
| Labour cost % | 22–28% | Daypart staffing recommendation in morning brief |
| Repeat-visit rate (90d) | 38–46% (café), 28–36% (QSR) | Loyalty segment drafted weekly |
| Aggregator share of revenue | 18–32% | One queue across 5 aggregators; per-aggregator margin in dashboard |
| AI forecast MAPE per outlet | 14–22% | Recalibrates weekly per outlet |
| Ticket time (peak) | 6.8–9.2 min | KDS routing recommendation when over band |
| Void rate | <0.8% | Pattern-detection on staff/outlet/daypart |
Common pitfalls SEA operators hit in 2026
Treating aggregator orders as a separate business. Operators who keep five aggregator tablets running in parallel lose roughly 4–7 minutes per peak hour to context-switching alone, and miss the per-aggregator margin picture entirely. Unifying the queue (one tablet, one KDS, one accounting line per aggregator) is usually the single highest-leverage move in the first 60 days.
Letting variance live in spreadsheets. A weekly food-cost review is a 7-day reaction time on a 24-hour problem. Variance has to live in the operating layer — flagged, attributed and routed to the responsible manager within hours, not aggregated to a Friday email.
Loyalty as a punch card. A 2026 loyalty programme is a messaging channel with attribution. If the only metric is "points issued", the programme is a cost centre. If the metric is "incremental repeat visits per segment per month", it compounds.
Forecasting at the wrong resolution. Chain-level forecasts are wallpaper. Daypart-and-outlet is the smallest unit that pays back — coarser is too vague to act on, finer is noise.
How LOOP solves this
LOOP is an AI-native restaurant operating system built for SEA F&B chains. Operators run their venues by voice or text command instead of clicking through dashboards. AI forecasts demand per outlet at daypart resolution (MAPE 14–22% on LOOP cohorts), flags food-cost and labour variance within hours of the shift closing, drafts promos for slow daypart slots and pushes them to Zalo OA / LINE / WhatsApp, and delivers a three-item morning brief at 06:30 local time so the operator's first action of the day is informed. LOOP unifies GrabFood, ShopeeFood, Be, foodpanda and Gojek into one queue, supports VietQR / PromptPay / QRIS / DuitNow / PayNow / QR Ph, and ships VAT e-invoice (TT78) via MISA, Viettel and VNPT. Pairs with Peko loyalty (50% lifetime discount on LOOP for Peko customers).
Under the hood, LOOP is offline-first with a 90-second resync window so orders, payments and KDS keep firing through ISP drops; recipe-level COGS is computed at order time so every plate's contribution margin is visible before the shift ends; and the morning brief is generated from the previous day's variance, the current day's forecast and the next 14 days of bookings, weather and local events — not a static template. The result is fewer dashboards, faster decisions, and a noticeably calmer week for the operator.