Today
Every morning planners juggle due dates, setups, material and absences in spreadsheets. The plan keeps changing. Nobody but the planner can see why one order runs before another.
An example
At Talwerk Antriebstechnik two rush orders for geared motors come in. The agent shows two variants: one keeps every due date but needs three extra setups. The other saves those setups and moves the non-critical order 4471 by one day.
Your data, wherever it lives
Today orders, routings, material and customer due dates sit in the ERP, machine status in the MES, setup sequences, shift plan and planning rules in the planner's spreadsheets, absences in the HR system, and rush orders arrive by e-mail. demi reads, understands and links them per order, machine and shift, with your rights. The planning system calculates the variants to your rules, the AI compares and explains them in minutes instead of an hour and a half of juggling spreadsheets, and the planner decides.
Where the data lives today
- ERP
- MES
- Office files
- HR system
- E-mails
How demi solves it
Three layers in this order: connect the data first, then the rights, then the AI.
- 1 · ConnectData Layer
The case needs data from the ERP, MES and spreadsheets. demi replaces nothing and links production orders, routings and setup families with material availability, the shift plan, customer due dates and the planning rules from the spreadsheets. demi writes back only after approval.
- 2 · PermitPermission Layer
Planners own the plan: the agent proposes, the planner releases. Every change to the plan is recorded with name and reason.
- 3 · ActIntelligence Layer
Sequences are computed by your existing planning system or a connected scheduling solver with fixed rules, not by the language model. The agent compares the variants on lateness and setup time and explains the trade-offs in plain words, every figure citing its order.
How it runs
- 1
New orders, rush orders or material delays come in.
- 2
The planning system computes several sequences by your rules.
- 3
The agent compares them on due dates and setup time.
- 4
It explains each variant in a few sentences.
- 5
The planner chooses, adjusts and releases.
What changes
Faster daily planning
Priorities everyone can follow
Fewer setups and late orders
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
+108 hfreed per year
Assumptions and calculation
demi proposes the sequence by due date, set-up order and material and explains it; you set it.
90 min → 63 min per case
How we calculate
20 planning runs × 90 min × 30% × 1.0 (manual) × 12 months = 6,480 min = 108 h
Value per year, before demi costs
Freed per year+108 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
+$7,020
gained per year
108 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
Does it replace our planning system?
No. demi uses your planning system or a connected scheduling solver for the computation. Results go back only after your approval. What is new is the explanation of why the plan looks the way it does.
Can planners override the proposal?
Always. Planners can move or pin orders or choose another variant. The reason is recorded, so you can adjust a rule that keeps getting overridden.
How does it handle rush orders?
A rush order triggers a recalculation. The agent shows which orders move and which due dates would slip. The planner decides whether and how the rush order goes in.
What if our rules exist only in the planners' heads?
Then demi writes them down together with the planners, from conversations and past plans. A planner confirms each rule before it applies. That way the knowledge stays in the company through holidays and staff changes.
