AWS IoT SiteWise collects data from industrial equipment and organises it into asset models and hierarchies: a site contains lines, lines contain machines, machines have measurements. Manufacturers, energy companies and building operators use it to bring plant data into AWS without building their own pipelines. Once the data is there, most people still need a SiteWise Monitor portal, a dashboard someone else built or an engineer who knows the property aliases.
With Agentic CMMS, Claude, ChatGPT and other AI agents read your SiteWise assets directly. A maintenance planner asks about a machine by name and gets its latest values, its alarm state and how it behaved this month, in one answer. Nobody has to open the AWS console.
What your AI agent can do with AWS IoT SiteWise
- Browse the asset hierarchy. "Which machines sit under the Packaging line at the Leeds site?" The agent walks the hierarchy and lists each asset with its model.
- Read the latest values. "What are the current motor current and spindle speed on CNC-14?"
- Pull history as aggregates. "Give me the hourly average temperature of Oven 3 for the last two days." The agent asks for aggregates instead of raw points, so the answer stays short and readable.
- Triage alarms. "Which assets have an active alarm right now, and how long has each one been active?"
- Explain a model. "What measurements and metrics does the Chiller asset model define?" Useful when a new engineer joins the team.
- Compare identical machines. "Rank our 12 extruders by their downtime metric this week."
- Check data freshness. "Which assets have not reported a value in the last hour?" The agent compares the latest time stamps and lists the quiet ones.
AWS IoT SiteWise with your other systems
- Turn a SiteWise alarm on a chiller into a corrective work order in HxGN EAM (Infor EAM), with the alarm time and the last values in the notes.
- Ask which machines with rising temperature trends also have overdue preventive maintenance in MaintainX.
- Match the downtime metric for a line with the production orders in SAP S/4HANA to see which orders slipped because of it.
- Put live SiteWise readings next to the long-term history your data team keeps in Snowflake, and ask whether this summer looks like the last one.
For teams who built their plant model in AWS
If your engineers spent months on SiteWise asset models, the agent puts that work to use. It speaks in the names of your sites, lines and machines, not in property IDs, and it follows the hierarchy your team designed. Questions that once needed a new dashboard now need one sentence.
Your AWS credentials stay with Agentic CMMS and never reach the agent. Each person approves their own agent, owners and admins choose which roles may use each SiteWise tool, and every request is written to the activity log.