Workflows
Build powerful AI automations
Turn production events into action with visual AI workflows. A stop on a line, a finished shift, an incoming document: each can trigger a chain of steps that gathers data, applies AI where it helps and delivers a result.
From event to action, automatically
Most operational work is repetitive: a stop happens, a report is needed, a task gets created, someone gets notified. Today those patterns run on habit and memory, which means they run late, or not at all, on a busy shift. Workflows let you wire the pattern once and let it run every time: pull the data from the OptiPeople Data Platform, let AI summarize or classify it, and push the result to the people and systems that need it. Your team keeps the judgment calls; the workflow keeps the routine.
Building blocks for any process
Visual, no-code canvas
Drag in nodes for agents, conditions, loops, web search, guardrails and custom code. Connect them on a canvas and you have an automation, readable by the people who own the process, not just developers.
Trigger on real events
Start a workflow from a downtime event, a schedule, a new document or an inbound request. The trigger carries its context (machine or workstation, order, shift), so every step downstream knows what it's working on.
AI where it adds value
Summarize a night's stop events, classify a fault description, extract order numbers from a document, or draft a report. Guardrails keep AI output within the bounds you set.
Branch, loop and combine
Route long stops one way and short stops another. Loop over every machine or workstation on a line or every order in a batch. Combine data steps, AI steps and notifications in one flow.
People in the loop
Not everything should run unattended. Add approval steps so a person confirms before a task is created or a report is sent. The workflow does the legwork; you keep the sign-off.
Connected to your systems
Workflows read from your live platform data and reach the systems around it through integrations, so the result lands in ERP, email or the tools your team already checks.
How teams use it
The automatic shift report
At shift end, a workflow pulls OEE, output and stop causes, has AI write a short summary and sends it to the next shift and the team lead. Same structure, every shift, no one typing it up.
Downtime follow-up
A stop runs past its threshold. The workflow classifies the registered cause, creates a maintenance task with the machine's recent history attached and notifies the technician on duty.
Catching quiet waste
A scheduled workflow checks energy data against production: machines drawing power while idle get flagged, and the list lands in the right inbox before the next planning meeting.
Automate the work nobody wants to do twice
Bring one repetitive task from your operation to a demo. We'll map it as a workflow together.