0%
of food is lost between harvest and retail, before it ever reaches the consumer.
FAO / UNEP · ~USD 400 billion per year
WAREHAUS · WMS + UHF RFID · BY TECHPRO
The warehouse that warns before it hurts.
A WMS built on AI with RFID technology: full inventory counts in minutes — by hand with mobile readers or automatically with fixed UHF antennas — and alerts that end in moved pallets, not emails.
- inventory accuracy
- 99.5%inventory accuracy
- faster than manual counts
- ×10faster than manual counts
- to production
- 6 weeksto production
THE PROBLEM
Your inventory is already costing you money. Quietly.
20–0%
of a typical operation's inventory is dead stock or turns so slowly it only ties up capital and space.
NetSuite / Supply Chain Today
0.0%
of retail sales evaporate as shrink: expiry, damage, misplacement and theft.
NRF FY2022 · USD 112.1 billion in the U.S. alone
0%
of picking time is spent walking between locations, not picking product.
Public warehouse-operations benchmark
And every unit sitting in the warehouse costs 20 to 30% of its value per year just to exist: tied-up capital, space, insurance, handling and obsolescence. That's the carrying cost, and you pay it whether the item sells or not.
None of these losses requires a catastrophic event: they only require that no one looks at the data in time.
THE APPROACH
It's not a WMS with an AI module. It's a WMS built on AI.
Warehaus runs a closed loop: what the antennas capture feeds the model, the model fires alerts, every alert ends in a physical action, and every action retrains the model. Inventory stops being a monthly snapshot and becomes a system that learns.
01RFID capture
Antennas read every tag in real time, no manual scanning.
02Model
The AI cross-references rotation, expiry, location and demand.
03Alert
The system warns before the cost happens.
04Action
Concrete orders: move, clear out, replenish.
05Learning
Every outcome retrains the model and sharpens the next alert.
Alert inbox · live
Demo environment
- Triggering signal
- Continuous reads of the refrigerated zone found 84 units past their expiry date.
- Model rule
- remaining shelf life = 0 → immediate picking and sales block.
- Recommended action
- Pull the batch, log controlled shrink and free the location. The stock stops being offered to stores.
- Triggering signal
- 9 days of shelf life left and the current sell-through won't clear the batch in time.
- Model rule
- remaining shelf life < 25% and projected sales < stock → FEFO priority.
- Recommended action
- Ship it early to the highest-turnover stores and trigger a markdown before it becomes shrink.
- Triggering signal
- Turnover dropped 40% versus last quarter and it still sits in a priority location.
- Model rule
- turnover < 30th percentile of its category for 4 weeks → reclassify.
- Recommended action
- Move to zone C, lower the reorder point and hand the golden slot to a high-demand SKU.
- Triggering signal
- 120 days without a single outbound movement recorded by the antennas.
- Model rule
- 0 outbounds in 90+ days → dead stock: capital and space tied up.
- Recommended action
- Propose liquidation, supplier return or donation; free the location and the working capital.
- Triggering signal
- Outbound velocity 3× its average; stockout risk in 5 days.
- Model rule
- projected demand > on-hand stock + in-transit orders → replenish.
- Recommended action
- Raise the reorder point, pull the purchase order forward and move it near the dock to shorten picking.
UHF RFID TECHNOLOGY
Full inventory counts without scanning a single label.
Barcodes require seeing and aiming at every label. UHF radio frequency doesn't: each unit carries a tag that answers on its own, and Warehaus captures it in two complementary ways.
A full aisle in a single sweep.
- An operator walks the aisle with a handheld RFID reader: UHF waves capture hundreds of tags per second, with no line of sight and without opening a single box.
- A full physical count goes from days of line-by-line scanning to minutes of walking.
- Ideal for cycle counts, spot audits and receiving verification.
0
line of sight
The tag responds even inside the box or behind the pallet.
100s
tags / second
Hundreds of simultaneous reads in a single antenna sweep.
>98%
accuracy
Inventory accuracy with WMS + RFID; 99.5% best-in-class.
24/7
by zone
Continuous coverage: receiving, rack, dock and sales floor.
Warehaus runs on EPC Gen2 / RAIN RFID, the global standard for UHF radio-frequency identification: passive, low-cost tags that are interoperable across reader and antenna makers.
FROM ALERT TO PHYSICAL ORDER
The alert doesn't end in an email. It ends in a moved pallet.
Warehaus classifies every SKU by turnover and proposes the slotting: high-demand products migrate to the golden zone, next to the dock, and each pick's travel gets shorter. Toggle it and watch the layout.
Class A · high turnover
Class B · medium turnover
Class C · low / no turnover
Average travel per pick line
24 m
with Warehaus: 19 m
Golden zone occupied by class A
31%
with Warehaus: 78%
FEFO compliance
72%
with Warehaus: 96%
Demo layout. The improvement shown stays within the band public benchmarks support: travel accounts for ≈50 % of picking time, poor slotting explains 30–40 % of that travel, and good slotting cuts 40–60 % of that waste.
THE BUSINESS CASE
Put in your numbers. The model uses public benchmarks, not our promises.
Your numbers
$100,000$20,000,000
$50,000$5,000,000
$10,000$2,000,000
Working capital freed
$300,000$700,000
15–35% of inventory value · one-time cash effect
Carrying-cost savings
$75,000$175,000
25% per year on the freed capital
Logistics savings
$37,500$75,000
up to 15% of annual logistics cost
Shrink avoided
$45,000$75,000
30–50% of current annual shrink
Estimated recurring annual savings
$157,500$325,000
Adds carrying cost, logistics and shrink: line items that repeat every year. Excludes the freed working capital, which is a one-time cash effect, not a recurring saving.
Assumptions: inventory reduction of 15–35% and logistics-cost reduction of up to 15% (McKinsey, supply-chain digital transformation; the conservative logistics band applies half the benchmark); carrying cost of 20–30% per year, modeled at 25% (NetSuite / Supply Chain Today); shrink avoided of 30–50% when moving from periodic counts to continuous visibility. Indicative estimate, not a quote.
RFID against shrink
Every tag is a witness. Shrink stops being invisible.
Shrink averages 1.6% of sales — USD 112.1 billion in the U.S. alone per NRF FY2022 — and comes, in this order, from expiry, damage, administrative error and theft. Warehaus attacks all four causes with the same infrastructure.
Cause 01
Expiry
Real FEFO: every batch enters with its date and the system sequences outbound by expiry, with per-batch and per-date alerts before the product is lost.
Cause 02
Damage
Less handling and less searching: every unit is found on the first try, without moving pallets to locate it. Fewer touches mean less damaged product.
Cause 03
Administrative error
The physical inventory and the system are one and the same: continuous radio-frequency reads, no manual entry. Keying mistakes disappear because no one types the inventory.
Cause 04
Theft
UHF antennas at exits and critical zones detect unauthorized movements in real time. A tag crossing where it shouldn't raises an alert instantly, not at the monthly count.
The result is measured in a single number: how closely what the system says matches what's on the floor.
>98%
accuracy with WMS
99.5%+
with best-in-class RFID
Security and trust
Your data stays yours.
An AI-driven WMS sees all of your inventory. That's why Warehaus's architecture starts from one principle: you keep control of your data, your rules and the decision to continue.
Isolated data
None of your inventory data is used to train third-party models. A dedicated instance per client, encryption in transit (TLS) and at rest (AES-256).
Every alert is auditable
Every recommendation shows the signal that triggered it, the threshold applied and the history behind it. No black box: you can trace every alert back to its source.
Works alongside your ERP
Warehaus doesn't replace your financial system: it connects to it with read access and hands back reliable inventory. Your ERP remains the accounting source of truth.
Success criteria in writing
The KPI and the threshold are defined before signing. If the pilot doesn't hit the agreed metric, you don't continue. The outcome risk is ours, not yours.
- Role-based access control
- Scheduled backups
- Event logging (audit log)
- Least privilege in integrations
Next step
Six weeks to your first alert in production.
WK 1–2
Discovery and data connection
SKU master, locations, batches and expiry dates. Integration with your ERP, without stopping operations.
WK 3–4
Model calibration
On your real history: turnover thresholds, expiry windows and the seasonality of your demand.
WK 5
Shadow pilot
Alerts run in parallel with your current operation. You compare against reality, with no risk.
WK 6
Production and re-slotting plan
Live alerts, the first wave of re-slotting and an executive dashboard with the agreed metric.
What the pilot needs from your side
- 01
One warehouse
The pilot lives in a single operation, not across the whole network.
- 02
Read access
To your ERP and inventory. We never write to your systems.
- 03
An operations lead
About 2 h/week on your side.
- 04
An agreed metric
KPI and threshold signed before we start.
We're not asking for a leap of faith. We're asking for one warehouse and six weeks.
What you'll get
30-minute demo
On sample data: real alerts, dashboard and workflow.
RFID fit assessment
Of your operation: volumes, materials and viable read points.
Pilot proposal
With a written KPI and success criteria signed before we start.
A Techpro product.