About this Scenario
The Munich IoT Center scenario is a live design demonstration scenario on the use of operational, performance, and anomaly monitor dashboards using Maximo Asset Monitor. The scenario outlines the monitoring and reporting tasks over a typical day at the Munich IoT Center for the fictitious operator and facility manager Bryan.
IBM is investing over USD 3 billion into Internet of Things – including USD 200 million to the Munich IoT Center. In Munich, the Internet of Things comes of age with advanced Watson cognitive computing technologies and the world’s first state-of-the-art client ‘collaboratories’. Take a virtual tour of the Watson IoT Center in Munich.
The floor plan of the Munich Twin Tower building, hosting the IBM IoT Center, is sectioned into East- and West-wing workspaces and meeting rooms. The wings are separated by the central elevator section, conference rooms, hallways, and utility spaces.
The IBM IoT Center in Munich has been instrumented with devices and sensors from many IBM IoT business partners. In this design scenario, we are using the 800+ devices from Yanzi Networks deployed to most of the IoT Center floors of the building.
The Munich IoT Center scenario has played a major role in the design and development of the Watson IoT Platform and Maximo Asset Monitor solutions. It demonstrates the everyday experiences and challenges for Operators and Administrators;
- The scale of deployment and management
- The scale of data ingest and storage
- Variability in sensor types, abstraction of sensor data, and normalization of measured units
- Configuration of operational and performance metrics
- Configuration of data analytics like data anomaly models and alerts
- Configuration of dashboards
- Use of dashboards for monitoring w/ search, filter, navigate and compare building data
This article overviews and demonstrates the design outcome of Maximo Asset Monitor in eyes of Bryan, the Facility Operations Manager.
Mastering Dashboards in Maximo Asset Monitor
Hands on Lab for a practical introduction to the use of Maximo Asset Monitor.
Monitor Scenario Personas
Bryan is a fictitious Operator and Facility Manager at the Munich IoT building. He relies on the Maximo Asset Monitor solution for his operational and engineering roles. He depends on building KPIs like utilization, compliance, and comfort to ensure tenant satisfaction. He depends on operational metrics like CO2, temperature, noise, light, and humidity levels to monitor comfort. He depends on AI anomaly models to notify of abnormalities in the building sensor data, operational metrics, and KPIs. He depends on dashboards to present summary and drill-in to operational metrics and KPIs, allowing him to prioritize the alerts, validate the root causes, take action and assign service requests to his facility engineers.
Munich IoT Monitor Scenario Storyline
Starting the day
Bryan starts his day at the office by getting an overview of the facility. He logs into the Maximo Asset Monitor tenant and selects the ‘View Dashboards’ option to pick the Munich IoT Center summary dashboard. He opens the summary and alert dashboards from the Dashboards page to get an update on the state and the building, the floors, and the zones. He looks for critical alerts he needs to prioritize. He will then start validating the alert, find the root cause and take action.
Getting an overview
Brian focuses, in order of urgency, on the following tasks supported by Maximo Asset Monitor dashboards.
Building overview – He explores the Munich IoT Center Summary dashboard to confirm the overall state of the building and identify any abnormalities. He looks at the alerts table for a quick overview of the building’s health. He then applies filters to drill into floors and zones that look suspicious.
What’s broken – He looks for ‘assets in trouble’ and alerts and correlates these to the floors and zones in the building using the Alerts Summary dashboard. He applies filters to only see alerts on floors and zones that look suspicious.
What are the trends – He views the performance and compliance indicators for the last day(s) and looks for any abnormalities in the trends using the Performance dashboard. He applies filters to only see KPIs on floors and zones that look suspicious.
Using the monitor dashboards in Maximo Asset Monitor, Bryan identifies three issues he needs to follow up on and investigate root-cause.
Looking for root-causes
Carbon Dioxide (CO2) alerts at Floor 27, Zone 3 –
Bryan first prioritizes the CO2 alerts in the Floor 27 Zone 3 meeting room. He follows the link from the critical CO2 alert to the workstation dashboard displaying the room conditions and trends. He validates that the CO2 levels are above the compliance level of 800 PPM and that the abnormal levels seem to be occurring in the afternoons. He suspects this to be due to telephone conferences in the room with several participants. He suspects overutilization of the room and that air quality is impacted. He needs to ensure that the root cause is not related to a failing air conditioning unit. During the day he plans to follow up with the project members on Floor 27 west wing, to validate his suspicions and look for options. He will also follow up with the facility engineer regarding the HVAC unit.
Temperature alerts in Zone 1 on multiple floors after 3 pm local time – Bryan inspects the temperature alerts in Zone 1 on multiple floors. Zone 1 is located in the west wing of the building and Bryan suspects that afternoon sunlight might cause this high-temperature abnormality. He proceeds and validates any correlation between workstation temperature alerts, operations of the automatic building blinds, and the sunlight radiation level from weather data.
Utilization of the meeting rooms – Bryan notices a drop in the utilization of the meeting rooms on some of the floors. He is concerned if this is related to tenant satisfaction, or changes in the project staffing. He follows up with the projects on the suspicious floor and looks for the best options to improve the performance and efficiency of the space.
After prioritizing the alerts, and analyzing the root causes using the data provided by the Maximo Asset Monitor dashboards, he can start to take actions to resolve the issues.
Scenario Configuration
The Munich IoT Center scenario is running on a live monitor configuration. This section discusses some of the steps performed by an Application Administrator to configure the solution.
Registering the IoT Devices
The IBM IoT Center in Munich has been instrumented with devices from Yanzi Networks deployed to most of the IoT Center floors of the building. See deployment on the floor plan below. Three types of devices have been deployed.
- Yanzi Motion devices (blue) for monitoring motion and temperature.
- Yanzi Motion+ devices (red) for monitoring motion as well as temperature, humidity, ambient light, and sampled ambient noise.
- Yanzi Comfort devices (green) for monitoring air quality by measuring levels of carbon dioxide (CO2) and volatile organic compounds, as well as temperature, humidity, barometric pressure, and ambient noise
The devices contain sensors that are individually registered as Device Types and connect with Maximo Asset Monitor over the internet.
The IoT Platform Service in Maximo Asset Monitor provides views to the device types, devices, the ingested device events, the event data, and the transformation of data performed prior to storing data in the Maximo Asset Monitor data lake. Device types and Devices are presented in tabular form.
Devices may be assigned metadata in Watson IoT Platform service, for example, a Region, Building, Floor, Zone, and Workstation. This metadata information is used to logically associate the device with its location and use it later for analytics and reporting purposes. As an example, as shown in the illustration below, the workstations in zone 1 are instrumented with Motion and Motion+ type devices with temperature sensors. Column filters help to identify devices of a specific sensor type and their location. Details on devices are presented on the device details page, for example, the connection state, recent data events, and the latest sensor state values.
Processing Munich IoT Center Sensor Data in IoT Platform
Events are the mechanism by which devices publish data at a regular heartbeat to Maximo Asset Monitor. The frequency is set by the device, for example, every second, every minute, every hour, or once a day. The 800+ devices instrumenting the Munich IoT Center are sending sensor data reading every minute. In total, 500.000 data events are to be stored every month in the Maximo Asset Monitor data lake.
The device sends the sensor data as a payload in the event. Data from devices of the same device type share the same payload data structure defined by an Event Type schema. By default, the IoT Platform expects events in JSON format specified by a JSON schema. The device detail page presents the connection state, the data events, and the event payloads.
Maximo Asset Monitor provides capabilities to create shared abstractions of device data so that devices of various brands all confirm a common schema definition of data, independent of the device type. The device data abstractor is specified using a Logical Interface. The logical interface may also normalize and transform data using JSONata mapping expressions. In the Munich IoT Center demo configuration, interfaces have been configured for all sensor types. These mappings transform data units from Kelvin temperatures to Fahrenheit, and noise levels to dB. Mappings also compute KPIs like Comfort Levels and Regulatory Compliance levels for temperature, CO2, noise, light, humidity, and air pressure. Mapping on the motion sensor data computes occupancy and utilization KPIs using.
Classifying, Aggregating, and Analyzing Munich IoT Center Data in Maximo Asset Monitor
Maximo Asset Monitor adds a concept of Dimensions. A dimension is a name/value pair used for asset classification and metadata. Dimension values are static or slow-changing. In the Munich IoT Center configuration, dimensions are used to classify the devices using their location expressed as REGION, BUILDING, FLOOR, ZONE, WORKSTATION, and DEVICE dimensions. Dimensions are automatically set by metadata values set in the devices.
The values are simple mappings to the Munich building instrumentation; Building: Munich, Floor:27, Zone:3, Workstation:1-2, Device:809646_EUI64-0080E103000226A9. Using aggregation functions and classification dimensions, Maximo Asset Monitor can compute aggregations across the building, a floor, a zone, and down to a workstation. For example, filtering all sensors on a floor the average across the isOccupied metric across sensors over an hour can be computed resulting in a metric of floor Utilization expressed as a 0.0 – 1.0 percentage value. This aggregation design is used across all of the KPIs in the demo design.
Designing the Munich IoT Center Operational Dashboards
Maximo Asset Monitor provides two types of dashboards; Summary Dashboards and Entity Dashboards.
A Summary Dashboard is a dashboard that presents aggregated KPIs and provides drill-in to the metrics using hierarchical filters. Let’s decompose that statement a bit. Bryan, the Facility Operations Manager, the summary dashboards provide an overview of KPIs across all regions and buildings, including the Munich IoT Center building. And it allows filtering to a selected building and its floors, zones, workstations, and environmental sensors. Summary dashboards present aggregations by the hour, by day, by week, by month. This allows the KPIs to be tracked and compared over shorter and longer time ranges. An example of a Workstation Summary dashboard is presented below.
In the Munich IoT Center demo, three main summary dashboards were designed for operational metrics and KPIs; Building overview, Performance, and Alerts. All dashboards use Region, Building, Floor, Zone, Workstation dimensional filters.
An Entity Dashboard is a dashboard that presents the operational metrics of a single entity. In this scenario a single workstation. Metrics from a single entity can be presented without any aggregation and in near real-time. Workstation sensor values and anomaly alerts are presented on value cards, tables, and graphs on the dashboard. The Workstation Entity Dashboard for the meeting rum at Floor 27 Zone 3 is shown below.
The dashboards use cards and layout designs:
- Value cards of SMALL size to present current metric values.
- Time-series graphs cards of MEDIUM size to present a thumbnail metric view.
- Full-size graphs cards of LARGEWIDE size to present Mean, Max, Min aggregated data for metric trends.
- Alert card in LARGE and LARGEWIDE size to present any alerts on the filtered level.
- Image card with hot-spot overlays presenting data metrics, and alert thresholds.
- Cards are presented in an 8 column layout.
The definition of a dashboard is encoded in a JSON file that may be uploaded and downloaded using the Maximo Asset Monitor dashboard editor.
Get Hands-On with the Munich IoT Center Demo and Maximo Asset Monitor
Explore all the steps outlined above to configure the devices, sensors, data processing, data storage, analytics, and dashboard configurations in the hands-on lab ‘Maximo Asset Monitor 101’.
In the lab, you will start by exploring more in detail the device, data, analytics, and dashboard configurations in the Munich IoT Center scenario. You will use the monitor dashboards to explore the live state of the Munich building. View the hands-on lab material for ‘Maximo Asset Monitor 101’ for more details on the scenario configuration.
Mastering Dashboards in Maximo Asset Monitor
Hands on Lab for a practical introduction to the use of Maximo Asset Monitor.