With factorIQ, AI agents connect orders, knowledge, and processes across your existing systems and independently move work forward.
From iosys projects
Today: finding out where an order stands means asking around the shop.
DUAP · Ten screens in around 30 hours of engineering effort.
Production order numbers, parts, and KPIs blurred
Today: work instructions are retyped by hand from PDFs into the MES.
GEWO · In the example shown: 32 of 36 fields with high confidence.
Sample document, no customer data
Today: when a supplier moves a date, planning recalculates by hand.
In development, not yet in use at customers
invented sample data
Purchase orders, drawings and ERP tables describe the same orders, parts and operations, each in its own way. Agents propose mappings; your accountable experts confirm them.
From raw data to shared meaning
Orders, parts, operations: one modelPDF purchase order in the mailbox
Details sit as text in a PDF, not as fields in the ERP.
becomesCustomerOrderPart
Work instruction as PDF or DOC
Unstructured; every extracted field carries a confidence rating.
becomesOperationPart
Scanned drawing
No text layer, 8 of 11 sheets rotated, 7 title-block layouts.
becomesPart
ERP database
Proprietary and undocumented: about 5,000 objects narrowed to six core tables.
becomesOrderDepartment
Handwritten check marks and notes
A supplier's inspection remarks, not drawing content.
becomesSupplierPart
Promised delivery date
When a supplier moves a date, planning recalculates by hand today.
becomesSupplierPartOrder
Today this model is built per workflow and it grows with every one. Company-wide data cleanup is not required for the first workflow.
Choose a module. Each one names its status and, where available, its evidence.
The connected factory, module by module
Color shows each module's current status.
Order office
Planning and control
Production hall
Ontology: orders, parts, operations, suppliers
Shop-floor quality checks and customer complaints from Service feed root-cause analysis independently. Your team approves changes to knowledge and work instructions. The next order tests their effect.
The model reads PDF purchase orders from the mailbox; deterministic checks validate the result, and a person approves. The approved result enters the ERP in structured form.
Evidence: 20 → 2 hours per week at a CNC contract manufacturer
to the product →Unstructured instructions from PDF and DOC are extracted and entered into the MES, with a confidence rating per field and a check by the operator.
Evidence: in production, export to the Hummingbird MES
KPIs from ERP and MES where the work happens: one screen per department, orders, due dates, and deviations as soon as they occur.
Evidence: ten screens in around 30 hours of engineering effort
Stock, bills of materials, and capacity over six months. Changes remain a preview until the planner confirms them.
Evidence: demonstrable version with preview, confirmation, and reset
Tracks the bought-in parts from the bill of materials: promised dates, delays, stock. Reports bottlenecks before they show up in manufacturing.
The module receives, classifies, and forwards. Fixed rules and people decide.
Manuals, work instructions, and file stores can be queried. Every answer names its source.
Explains deviations from measurements, machine data, and tool changes. A person chooses the action.
Queries after delivery: order, drawing, and inspection report pulled together, with a draft reply ready for approval.
Under the factory: ontology, data and existing systems
Lines show where a module takes its data from. Choose a module above to highlight its path.
Operations
Ontology
Data sources
Existing systems
Learning loop
Orange: modules whose approved improvements flow into the next order and quality check
Orders, material, capacity, and quality share context with ERP and MES. Agents connect the work and learn from approved corrections.
Your data stays in your plant.
We run it in Switzerland or the EU.
The application runs in your Microsoft environment.
The blue square shows the intended processing location.
Availability depends on the module. Use of the AI model provider is agreed separately.
factorIQ is a product of iosys in Menzingen, Switzerland. AI agents connect orders, knowledge, and processes across your existing systems and independently move work forward. You set the goals and boundaries.
factorIQ works on your existing systems. ERP and MES stay customer systems that factorIQ reads and writes. Write permissions are approved separately for each workflow.
Your team. Agents work within agreed permissions. Your team decides trade-offs, your people approve deviations, and uncertain values are flagged.
We settle that with you. The application can run on iosys servers in Switzerland or the EU, in your Microsoft environment, or, depending on the module, in your plant. Which deployment option is available depends on the module. Complete production operation with a local AI model is not proven, and the location of the application is not automatically where a model provider processes requests.
At DUAP, ten screens show orders, due dates, and deviations, one per department, in around 30 hours of engineering effort. In the GEWO example shown, 32 of 36 fields were read with high confidence. maxon runs MIND from our maxon project in production in its own Azure environment today. maxon case study
We do not publish a universal price for setting up or running factorIQ. These are scoped separately: interfaces and data access, workflow and checks, operating location, usage and support, and later changes. The order-processing case study is a single project, not a price list.
Pick an available workflow with usable data and an accountable owner. Before a bounded pilot, we agree the baseline, review workload, and release criteria, then expand after observed performance. Write to us or book a call.