Today
Waste wood arrives as construction and demolition timber, pallets, furniture or bulky waste, mixed and in containers, and every load has to go into a category; in Germany the Waste Wood Ordinance sets four, A I to A IV. That decides whether it becomes chips for particleboard or goes to a biomass plant. At the gate, the decision rests on a quick look from the weighbridge office and a delivery note that usually says A II. If impregnated fence posts, railway sleepers or old window frames slip through, treated wood ends up in the recycling stream and the customer rejects an entire batch of wood chips.
An example
A roll-off container with 6.8 t of construction and demolition wood arrives at Wertora Recycling, declared as A II under waste code 17 02 01. On the photos from the tipping point, the agent marks twelve greenish fence posts and two window frames, typical signs of wood preservatives and therefore A IV. It also shows that the same supplier had A IV wood in A II loads four times in the past three months. The yard supervisor has the posts and frames sorted out and booked as A IV under 17 02 04*, and the rest goes to processing for the particleboard plant as A II. Sales receives a draft for the sorting fee and the feedback to the supplier.
Your data, wherever it lives
Today what is known about a load of waste wood sits in the weighbridge software with its camera images, in suppliers' delivery notes and declarations on paper or as PDFs, in phone photos from the tipping point and in the ERP with contract and waste code. demi reads, understands and links it per delivery and supplier, with your access rights. The AI proposes the category in seconds and marks contaminants on the photo, and the yard supervisor decides.
Where the data lives today
- Photos, video & audio
- Scans & paper
- PDFs
- E-mails
- ERP
- Databases
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 weighbridge with its camera images, the ERP with supplier, contract and waste code, accompanying papers as PDF or scan, and photos from the tipping point. demi replaces nothing and links, per delivery, weight, number plate, declared category, waste code, photos and later findings from sorting and lab. Every load then comes with the history of its supplier.
- 2 · PermitPermission Layer
Intake data and photos are visible to the weighbridge, the yard supervisor, plant management and quality; prices and contracts only to sales and dispatch. Only authorized staff may change a category, reject a load or charge a sorting fee; every decision is recorded with name, time and photo.
- 3 · ActIntelligence Layer
The agent reads delivery note, weighbridge ticket and the supplier's declaration, handwriting included, and evaluates the photos from weighbridge and tipping point. It marks typical signs of treated wood, such as greenish discolouration, tar oil, sleepers, poles or window frames, and foreign matter such as metal, plastic or stones. Category and route follow fixed rules your plant stores; the agent proposes, the yard supervisor decides.
How it runs
- 1
The truck drives onto the weighbridge; camera image, weighbridge ticket and papers come in with the declared category.
- 2
The agent compares papers and photos with the contract and the supplier's history.
- 3
After tipping, it evaluates the photos of the load, marks contaminants and proposes the category.
- 4
It recommends the route, whether processing for particleboard, sorting or energy recovery, and shows when a sample should go to the lab.
- 5
The yard supervisor confirms or changes the category; the ERP booking and the feedback to the supplier follow.
What changes
Every delivery classified and documented with a photo
Treated wood shows up before processing, not at the customer
Suppliers with repeated mis-sorted loads become visible
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
+288 hfreed per year
Assumptions and calculation
demi evaluates photos and papers of each delivery and proposes the category; you decide.
8 min → 5.6 min per case
How we calculate
600 waste wood deliveries × 8 min × 30% × 1.0 (manual) × 12 months = 17,280 min = 288 h
Value per year, before demi costs
Freed per year+288 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
+$18,720
gained per year
288 h ÷ 1,760 h per role = 0.2 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
Can a photo show whether wood was treated with preservatives?
Not with certainty. A photo shows signs: greenish or gray-green discolouration, tar oil, typical items such as sleepers, poles, palisades or window frames. The agent marks them and says how strong each sign is. Where it matters, it suggests a lab sample or a quick test; the yard supervisor makes the decision.
What happens when papers and load do not match?
The agent shows the contradiction, such as A II on the delivery note and sleepers on the photo, and proposes the right category and waste code. If hazardous waste wood is involved, it points to the records that are required. Whether the load is sorted, reclassified or rejected is up to the yard supervisor.
Where does demi get the rules for category and route?
Your plant stores them: the waste-wood categories that apply to you with their standard assignments, the acceptance criteria of your customers and your contracts with suppliers. When a rule changes, the new version applies from then on, and earlier classifications stay traceable under the old one.
Do we need new cameras to classify the wood?
Usually not. demi uses the images your weighbridge and your people already take, such as the camera above the weighbridge or phone photos at the tipping point. Better images help, so the agent shows for which deliveries the photos were not good enough. Whether an extra camera at the tipping point pays off, you will see after the first weeks.
