Starting point
Maintenance & service
Detect and dispatch machine faults automatically
Every reported machine fault is automatically turned into a service order and assigned to a technician. Dispatch only decides on exceptional cases, instead of reading each alert and assigning work orders by hand.
Process architecture
From a machine alert to a service order.
The chain takes over the machine's alert, classifies the fault and priority and creates the order in the system of record. Standard cases go straight to the technician, exceptional cases to dispatch, and return to the same flow after approval.
01Input
Fault & context
The fault is picked up and enriched with machine master data and history.
2 Steps
02Process
Classification & approval
Fault and priority are determined, the rules are checked and exceptional cases are prepared for dispatch.
4 Steps
03Output
Order & record
The service order is created, handed to the technician and the case is documented.
3 Steps
Source01Pick up the fault
The machine reports a condition value or fault code that detects the breakdown and starts the case with the machine and the time.Details
Pick up the fault
The machine reports a condition value or fault code that detects the breakdown and starts the case with the machine and the time.
Interfaces
- Microsoft Graph API
- SharePoint API
Next steps
- Load machine and master data
Data02Load machine and master data
Machine type, bill of materials, service history and location are attached to the case through the prepared view.Details
Load machine and master data
Machine type, bill of materials, service history and location are attached to the case through the prepared view.
Interfaces
- Azure AI Document Intelligence
- Mistral OCR
- Dataverse API
Next steps
- Determine the fault and priority
Action03Determine the fault and priority
Fault code and machine type produce the fault pattern with a priority, complemented by recognisable consequences for production.Details
Determine the fault and priority
Fault code and machine type produce the fault pattern with a priority, complemented by recognisable consequences for production.
Interfaces
- Power Automate
- Azure Functions
- Microsoft Graph API
Next steps
- Check the rules
Decision04Check the rules
Fault code, spare parts and the criticality of the machine are checked against the agreed rules instead of a decision by eye.Details
Check the rules
Fault code, spare parts and the criticality of the machine are checked against the agreed rules instead of a decision by eye.
Interfaces
- Azure OpenAI GPT
- Anthropic Claude
- AI Builder
Next steps
- Create the service orderStandard case
- Approve the exceptional caseExceptional case
- Record the caseUnknown fault code
Approval05Approve the exceptional case
If the fault code is unknown or the machine is essential, dispatch decides on the basis of the case.Details
Approve the exceptional case
If the fault code is unknown or the machine is essential, dispatch decides on the basis of the case.
Interfaces
- Teams Adaptive Cards
- Power Apps
Next steps
- Create the service orderAfter approval
System06Create the service order
The approved case is created in the system of record as a service order with machine, priority and spare parts.Details
Create the service order
The approved case is created in the system of record as a service order with machine, priority and spare parts.
Interfaces
- Dynamics 365 API
- Azure Service Bus
Next steps
- Hand the order to the technician
Result07Hand the order to the technician
The service order is available to the responsible technician with fault pattern and priority on the tablet.Details
Hand the order to the technician
The service order is available to the responsible technician with fault pattern and priority on the tablet.
Interfaces
- Power BI API
- Teams Webhook
Next steps
- Inform dispatch
- Record the case
Result08Inform dispatch
Dispatch receives the status with machine, priority and planned assignment.Details
Inform dispatch
Dispatch receives the status with machine, priority and planned assignment.
Interfaces
- Power BI API
- Teams Webhook
Result09Record the case
Alert, classification, decision and completion are archived and stay linked to the machine.Details
Record the case
Alert, classification, decision and completion are archived and stay linked to the machine.
Interfaces
- Power BI API
- Teams Webhook
System landscape
Interfaces and apps working together.
- Microsoft
Microsoft Graph API
Möglicher Baustein für die Microsoft-Systemvariante.
- Microsoft
SharePoint API
Möglicher Baustein für die Microsoft-Systemvariante.
- Microsoft
Azure AI Document Intelligence
Möglicher Baustein für die Microsoft-Systemvariante.
- Microsoft
Mistral OCR
Möglicher Baustein für die Microsoft-Systemvariante.
- Microsoft
Dataverse API
Möglicher Baustein für die Microsoft-Systemvariante.
- Microsoft
Power Automate
Möglicher Baustein für die Microsoft-Systemvariante.
- Microsoft
Azure Functions
Möglicher Baustein für die Microsoft-Systemvariante.
- Microsoft
Azure OpenAI GPT
Möglicher Baustein für die Microsoft-Systemvariante.
- Microsoft
Anthropic Claude
Möglicher Baustein für die Microsoft-Systemvariante.
- Microsoft
AI Builder
Möglicher Baustein für die Microsoft-Systemvariante.
- Microsoft
Teams Adaptive Cards
Möglicher Baustein für die Microsoft-Systemvariante.
- Microsoft
Power Apps
Möglicher Baustein für die Microsoft-Systemvariante.
- Microsoft
Dynamics 365 API
Möglicher Baustein für die Microsoft-Systemvariante.
- Microsoft
Azure Service Bus
Möglicher Baustein für die Microsoft-Systemvariante.
- Microsoft
Power BI API
Möglicher Baustein für die Microsoft-Systemvariante.
- Microsoft
Teams Webhook
Möglicher Baustein für die Microsoft-Systemvariante.
- Interfaces in use
MQTT
Reports condition values and faults of the machines over the plant network to the processing chain.
- Interfaces in use
Odoo External API
Creates the service order in the system of record and keeps it up to date.
- Interfaces in use
PostgreSQL
Read-only access to the prepared view of the machine master data.
- Interfaces in use
n8n
Drives the flow from the alert to the work order, visible and adaptable by the company.
- Interfaces in use
RabbitMQ
Queue between intake and processing, absorbing load spikes and maintenance windows.
- Possible apps & platforms
Odoo
System of record for faults, service orders and machine history.
- Possible apps & platforms
Grafana
Overview of machine conditions, kept in view by dispatch.
- Possible apps & platforms
Odoo Mobile
Feedback and completion of the order directly at the machine.
- Possible apps & platforms
Mail
Notification about new service orders.
Implementation in context
Robust in everyday use.
The process replaces no system, it uses the existing ones: the machines keep reporting over the plant network, the maintenance system remains the system of record and master data stays where it is maintained. New is only the path in between.
Decisions are rule-based where the rules are unambiguous and human where they are not. The company sets the limits itself; both paths are documented on the case.
Machine data does not leave the plant network. If a model classifies the fault pattern, it runs in the company's own data centre; only read access to the prepared view reaches the master data.
If the processing fails or the maintenance system is down for an update, a queue holds the alerts. Every case carries an identifier so the same fault never produces two orders.
Target picture
Every reported fault exists as a service order with priority and ownership as soon as the machine reports it. Dispatch only decides where the rules are not clear-cut, and each decision is documented together with its basis.
Expected benefits
- Every reported fault leads to a service order without a manual trigger.
- Dispatch works on exceptional cases instead of individual alerts.
- Duplicate orders for the same fault are ruled out.
- Every decision carries its basis and its approver.
- Faults, reaction times and completions can be evaluated at any time.
Prerequisites to start
- 01A documented access to the maintenance system with an interface and a technical user
- 02A machine connection that delivers condition values and fault codes over the plant network
- 03Maintained machine master data with unique identifiers, bills of materials and owners
- 04One named person in dispatch to approve exceptional cases
- 05A decision on criticality and priority rules per machine type
More use cases
Matching solutions from your industry
Field service & operations
Capture data directly at the work site
Teams document inspections, damage, time and materials on mobile devices; photos, signatures and location data feed the next workflow.
Current, complete field data without paper or duplicate capture, even with unstable connectivity.
- ERP- oder Field-Service-API
- Kamera, GPS und Gerätespeicher
- Identity- und Sync-Service
- Individuelle PWA oder Mobile App
Complaints & returns
Automate Complaint and Returns Handling
Every complaint from the mailbox and the web form is linked to its order and checked against photos and the reported defect. Only cases outside the agreed deadlines or approval limits reach quality assurance or sales management; everything else gets its decision the same day.
Every complaint is linked to an order on the day it arrives and checked against the same rules. Quality assurance and sales management only decide where the deadline, the defect, or the value falls outside the set limits, and each of those decisions is documented together with its basis.
- REST API
- Webhook
- Ollama
- DocuWare REST API
Accounting & invoicing
Create and send compliant e-invoices automatically
Every approved outgoing invoice is turned into a structured e-invoice in XRechnung or ZUGFeRD format, checked for mandatory data and sent over the agreed channel. Only invoices with incomplete or conflicting details reach the accounting team first; everything else leaves the company in the same run.
Every approved invoice is created, checked and sent as a compliant e-invoice over the agreed channel in the same run. Accounting only corrects incomplete or conflicting details, and every sent version is archived audit-safe together with its proof of sending.
- REST API
- ZUGFeRD und XRechnung
- Microsoft Graph API
- Gmail API
Finance & documents
Validate and post incoming invoices
Invoices are captured, validated and moved through approval, accounting and archiving with explicit exception controls.
Fewer capture errors, clear validation steps and an auditable path from document to accounting entry.
- ERP REST / OData / IDoc
- E-Mail, Peppol oder Upload
- DMS- und Archiv-API
- Azure AI Document Intelligence
Knowledge & assistance
Make internal knowledge securely searchable
An assistant answers questions from approved company sources, cites evidence and respects existing access rights.
Faster access to reliable knowledge with traceable sources and controlled permissions.
- Microsoft Graph / SharePoint
- Dateiserver-Connector
- Entra ID / SSO
- Azure AI Search
People & onboarding
Handle applications and prepare the first day
Every application is captured on the day it arrives, matched against the job profile and handed to the hiring manager ready for a decision. Once an offer is accepted, contract, accounts, equipment and the induction plan follow from the same case instead of a chain of reminders.
Every application is captured on the day it arrives, checked for completeness and handed over ready for a decision. After the offer is accepted, contract, accounts, equipment and induction plan run from the same case, and every decision is traceable with its reason and its deadline.
- Microsoft Graph API
- REST API
- Azure AI Document Intelligence
- Azure OpenAI Service
Your next step
Which process costs your team unnecessary time every day?
Give us a short outline. We will tell you honestly where automation is useful—and where it is not.
