Skip to content
// the virtual factory

How factorIQ works with your systems.

ERP, MES and documents remain the foundation. AI modules take on individual tasks.

View inbox →Arrange a call →
// layers

The layers of the factory

People
Interfaces
AI modules
Checks and approval
Your systems and data

Tasks

Purchase orders

Work instructions

KPIs

Workstations

Review screen

Shop-floor display

Control

Fixed rules check the details. Your people decide.

Integration

ERP

MES

Documents

Operation

At your site

At iosys

In your Azure

The status of each module is shown on the architecture page.
where it runs

Where does your application run?

stays with youDocuments and order data come to us, in Switzerland or the EU.Your plantYour Microsoft tenantEverything runs in your own Microsoft environment.Production data stays in the plantA local AI model is not yet proven in production.YOUR SYSTEMS REMAIN IN CHARGEERPMESMachinesif presentreads, writes backfactorIQOPERATION: IOSYS, SWITZERLAND OR EUOPERATION: IN THE CUSTOMER'S AZUREOPERATION: ON YOUR PREMISESOrderentryPDF to ERPrunningPlanningMaterial,due datesdevelopmentWork in-structionsPDF to MESrunningShop floorone screenper departmentrunningConditionMotor data,good / badrunningCauseanalysisexplainsdeviationsplannedUtility ductEvent log · Model gateway · Control points · MCP to the outsideWHERE PEOPLE WORKOwn interfaceChecks, dashboardsMCPTeams, Copilotconnected via MCPBack into the ERPand the MESTHE MODELCloud modelwith us, Switzerland or EUAzure OpenAIin your tenantLocal modelTarget, not yet proven in production
Only the blue line moves with the deployment option.
// technical view

The order flow in detail

factorIQ · Virtual factoryOrder chain, modules, systems and MCP integration.Model choice · Cloud, tenant or local · component · ConfigurationModel choiceCloud, tenant or localConfigurationSuppliers · Bought-in parts, dates · componentSuppliersBought-in parts, datesCustomer · orders and receives · componentCustomerorders and receivesOrder intake · read, check, create · factorIQ · runningOrder intakeread, check, createrunningLead time · Bought-in parts in the BOM · factorIQ · in developmentLead timeBought-in parts in the BOMin developmentShipping · Goods and documents · factorIQ · simulatedShippingGoods and documentssimulatedPlanning · Capacity and due dates · factorIQ · in developmentPlanningCapacity and due datesin developmentOrchestrator · classifies, forwards · factorIQ · in developmentOrchestratorclassifies, forwardsin developmentManufacturing · Machines and feedback · factorIQ · simulatedManufacturingMachines and feedbacksimulatedShop floor · KPIs per department · factorIQ · runningShop floorKPIs per departmentrunningERP and MES · Orders, BOMs · component · Customer systemERP and MESOrders, BOMsCustomer systemMCP integration · Teams, Copilot, Work IQ · factorIQ · InterfaceMCP integrationTeams, Copilot, Work IQInterfacePOOrderPlanGoodsDeliveryPromisesShortagetriagedFeedbackreads and writesToolsInferencefactorIQLegendFrontendBackendDatabaseCloudSecurityExternal

Order flow solid · Supporting work thin · System integration dashed. The colours show the type of module.

// modules

Modules and their current status

Available and in use

Order entry

available as SaaS

Purchase orders arrive in the mailbox as PDF. The model reads, fixed checks secure the result, a person approves. The result lands in the ERP in structured form.

Evidence: 20 → 2 hours per week at a CNC contract manufacturer

to the product →

Work instructions

running at customers

Unstructured instructions from PDF and DOC are extracted and created in the MES, with a confidence rating per field and a check by the operator.

Evidence: in production, export to the Hummingbird MES

Shop floor

running at customers

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

In development

Planning

in development

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

Lead time

in development

Tracks the bought-in parts from the bill of materials: promised dates, delays, stock. Reports bottlenecks before they show up in manufacturing.

Orchestrator

in development

Receives, classifies, forwards. Decides nothing itself: that stays with fixed rules and with people.

Knowledge

in development

Manuals, work instructions and file stores can be queried. Every answer names its source.

Planned

Root-cause analysis

planned

Explains deviations from measurements, machine data and tool changes. A person chooses the action.

Service

planned

Queries after delivery: order, drawing and inspection report pulled together, a draft reply for approval.

// the framework

What connects the modules

Event log

Every module records what it did, with a reason and a source. Interfaces read only events and need no knowledge of the modules.

Model choice

No module calls a model directly. Whether computation happens in the cloud, in the customer's tenant or locally is a matter of configuration.

Control points

Every module has at least one point where a person decides. The model interprets what is unclear, fixed tools secure what can be calculated.

External integration

The modules expose their capabilities via MCP so that the customer's AI environment can use them: Teams, Copilot, Work IQ.

deployment options

Three deployment options, the same software

Which deployment option is available depends on the module.

Operated by iosys

iosys runs the application in Switzerland or the EU. You work in your own review screen.

running today

In your Azure

The application runs in your Microsoft environment. The integration with Teams and Copilot is partly proven.

partly proven

In your plant

Production data stays in the plant. A complete solution with a local AI model is not yet proven in production.

partly proven
// the proof
20h2h

One module is already a product: inbox

At a CNC contract manufacturer, the time spent on order entry fell from 20 to 2 hours per week. 90 percent of entry now runs automatically.

published case study

View inbox →
get started

Where does your factory start?

Tell us which repetitive work you would like to simplify.

Arrange a call →