Today
Chilled goods pass through cold room, loading, reefer truck and often a cross-dock, and at every station a different system records the temperature: the cold room monitoring, the reefer unit in the carrier's telematics portal, data loggers on the pallet whose reports arrive as PDFs by e-mail. Which batch sat on which pallet in which truck is in the warehouse system and the ERP. When a retailer reports a warm delivery or an auditor asks, quality spends a day piecing curves together. Real failures on the road get lost among harmless door openings and often surface only at the customer.
An example
A retail distribution center in Augsburg tells Tarvina at 7:20 on a Monday that one pallet of SC 250 g herb sauce measured 7.8 °C at receipt and holds back the whole delivery of 18 pallets. At 7:35 the agent presents the history of both batches: 26 hours in cold room 2 at 1.9 to 2.8 °C, loading at 3:40, 2.0 to 3.5 °C on the road according to reefer unit and data logger. The curve rises only at the distribution center's ramp: 52 minutes with the door open and the unit switched off before receipt began. The quality manager sends the report with every source, and the distribution center releases the delivery after its own core temperature check. The same morning the agent reports a tour to Kempten whose return air has been rising for 25 minutes. Dispatch reaches the driver, the unit was stuck in defrost, and after a restart the return air is back below 4 °C before the HACCP plan's limit is reached.
Your data, wherever it lives
Today temperature data sits in the cold room monitoring, in the carriers' telematics portals and in data logger reports as PDFs in the mailbox; which batch was on which pallet and tour is in the warehouse system and the ERP, and deviations are in the QMS. demi reads, understands and links it per batch and shipment, with your access rights. AI reports a deviation while the tour is still running and presents the history in minutes; your quality team decides on release or block.
Where the data lives today
- Telematics
- Machines & sensors
- PDFs
- E-mails
- ERP
- QM system
How demi solves it
Three layers in this order: connect the data first, then the rights, then the AI.
- 1 · ConnectData Layer
Data sources in this case: the cold room monitoring, the reefer trucks' telematics in the carriers' portals, data logger reports as PDFs, the warehouse system and ERP with batches, pallets and tours, and the QMS with HACCP plan and deviations. demi replaces nothing and links every temperature curve, by time and place, with the batches, pallets, tours and vehicles that were there at that moment. Every batch then has its temperature history from filling to delivery.
- 2 · PermitPermission Layer
Temperature histories are visible to quality, shipping and dispatch; carriers see only their own tours, customers only their own deliveries and only after release. Only quality may block or release a batch, and every assessment of a deviation is recorded with name and reason.
- 3 · ActIntelligence Layer
The agent checks every curve against the limits of your HACCP plan, maximum temperature and permitted duration of an excursion, and tells from the reefer data whether a door was open, the unit was defrosting or it failed. For a real deviation it reports batch, place, duration and peak value to quality and dispatch at once. On request it assembles the full history of a batch with every source; the values come from the measurements, and the language model only summarizes.
How it runs
- 1
Temperature data from cold rooms, reefer trucks and data loggers reaches demi continuously or by e-mail.
- 2
demi assigns every curve, by time, pallet and tour, to the batches that were stored or on the road there.
- 3
The agent checks against the limits of the HACCP plan and reports a real deviation to quality and dispatch at once.
- 4
Quality assesses the affected batches and decides on release or block; the assessment stays with the history.
- 5
When a customer or auditor asks, the agent presents the batch's full history with every source as a report for release.
What changes
Every batch's temperature history from filling to customer in one view
Failures show up while the tour is still running, not at the complaint
An answer to a complaint in minutes instead of a day of searching
What does it bring you?
Check it with your own numbers
Every starting value is an assumption you can change. The calculation sits right below, step by step.
Your numbers
Result
+216 hfreed per year
Assumptions and calculation
demi links the temperature data per batch and presents the history; your quality team assesses.
90 min → 54 min per case
How we calculate
30 deviations and inquiries × 90 min × 40% × 1.0 (manual) × 12 months = 12,960 min = 216 h
Value per year, before demi costs
Freed per year+216 h
Assumption: labor cost of $48.62 per hour in US manufacturing in June 2026 (BLS), plus workplace, IT and overhead.
= Value per year, before demi costs
+$14,040
gained per year
216 h ÷ 1,760 h per role = 0.1 full-time roles
An estimate, not a promise. The starting values are assumptions for an industrial company with 150 to 500 people.
Combine with more cases and get the packageQuestions about this use case
Which temperature data does demi link?
The cold room monitoring, the reefer units of the trucks via your carriers' telematics, data loggers on pallets or in cartons whose reports come as PDFs or files, and manual readings at goods receipt. If a carrier only provides printouts or scans, demi reads those too. How your carriers' data gets to demi, we clarify in conversation.
Is every short excursion a deviation?
No. Your HACCP plan sets what counts, for example a maximum temperature and how long it may be exceeded. The agent recognizes door openings during loading and defrost cycles from the reefer data and treats them by your rules. A deviation is always assessed by your quality team.
How does demi assign a temperature curve to the right batch?
By time and place: the warehouse system knows which pallet stood in which cold room when and which tour it was loaded onto, and the tour knows the vehicle. If a link is missing, such as a pallet without a loading booking, the agent shows the gap in the history instead of bridging it.
Can customers see the temperature history of their deliveries?
Yes, if you release it. Each customer then sees only the histories of their own deliveries. For a complaint the agent drafts the report with curves, measuring points and sources, and your quality team sends it.
