Proprietary platform
RDM - Adeo's IoT platform
RDM (Remote Device Management) is the platform Adeo has used since 2012 to collect, process and present field data from vehicles, machinery, energy systems and sensors in industrial and research projects.
Summary
The engine behind Adeo's monitoring projects since 2012.
RDM collects data from field devices, processes it with configurable rules and calculations, and presents it through dashboards, alarms, reports and history. It is not a product written once: it is the foundation Adeo builds every monitoring project on, and it grows with them.
- Data acquisition from vehicles, machinery, energy systems, sensors and robots.
- Rules, alarms and indicators calculated on raw signals.
- Searchable history and a data foundation for predictive models.
Problem
Faults that develop slowly are spotted too late.
On-board and plant systems flag an anomaly only when a value crosses a threshold. A fault that develops slowly stays below the threshold until it shows up in the workshop or with the plant stopped, and costs more. On top of that, every project speaks different protocols and risks starting from scratch.
- Threshold alarms that miss slow drifts.
- Different devices and protocols from project to project.
- Scattered data, with no history to reason on.
Approach
One engine, configured for each project.
RDM separates acquisition from logic. Signals arrive from devices through collectors dedicated to each protocol; the logic lives in virtual devices, signals calculated with expressions over the real ones, which define alarms, conditions and indicators. A new project is configured, not rewritten.
- Separate collectors for each field protocol.
- Virtual devices: rules and calculations defined as configuration.
- A single data model for sites, resources and devices.
System
From the sensor to the decision, in the same platform.
The platform organises data in a hierarchy of sites, resources and devices, with templates that set up the devices of a new installation automatically. Alarms record every state change, the history keeps measurements with aggregations for analysis, and dashboards show real-time data, maps and trips.
- Resource templates and automatic device provisioning.
- Alarms with transition history and configurable delays.
- Real-time dashboards, GPS maps, trips and signal comparison.
- Indicators, scheduling and energy-management modules.
Results
From the TPER fleet to applied research.
In 2012 RDM was trialled on a few buses of TPER, the public transport company of Bologna: despite the small number of vehicles, it brought to light a series of malfunctions that, without that analysis, would have appeared later at a much higher cost. Since then the platform has been the data foundation of Adeo's projects in predictive maintenance, energy and structural monitoring.
- Public transport: monitoring of the TPER fleet in Bologna.
- Predictive maintenance on vehicles: SAMANTA funded project.
- Production equipment: SIGMAPro funded project.
- Energy and infrastructure: MOSES, TrackTango, EV4ER.
Technology
Several generations, the same principle.
The first generation of RDM stored data in SQL Server and received vehicle data through dedicated drivers. Since 2020 the platform has been rewritten on .NET and MongoDB, with separate collectors per protocol; the version in production today runs on .NET 9 with a Blazor web interface and real-time updates. The generation under development in 2026 moves to a time-series database and an isolated deployment for each customer.
- Protocols: MQTT, AMQP, SNMP, UDP, HTTPS, FTP, OBD-II, ROS.
- OIDC federated login.
- Compressed time series with 1-minute, 5-minute and 1-hour aggregations.
- Shared environment or dedicated installation per customer.
Reuse
Every project leaves the platform more capable.
The idea of virtual devices has been there since 2012 and has reached the latest generation intact: proof that separating acquisition from logic holds up over time. Each project adds a collector, a resource type or a module, and the next project starts from there.
- Separate acquisition from processing logic.
- Keep the history: it is the foundation for predictive models.
- Configure instead of rewriting.
CTA
Do you have devices or plants to monitor?
Tell us what you want to measure: we start from RDM and configure it for your case, instead of starting from scratch.
Talk to Adeo