Analyzing data

Use Maximo® Monitor to calculate essential key performance indicators for your organization from your IoT and non-IoT data.

Input data is ingested or loaded into the data lake from various sources, such as through the IoT tool, from .csv files, from external IoT systems and data connections, and from database tables.

Note: Starting from Maximo Monitor 9.1, streaming data metrics are no longer available. The documentation provided for streaming data metrics applies only toMaximo Monitor 9.0 and earlier versions. Try using batch data metrics. For more information, see Creating batch or streaming data metrics
Maximo Monitor performs calculations on the data in the data lake every 5 minutes for batch data metrics, by default. The output data from these calculations is added to the data lake where it is available for analysis or as input to other calculations. Calculations can be performed within a minute for streaming data metrics, which are available in Maximo Monitor 8.9 and later.

Several built-in functions are available in a function catalog for you to apply to your device types and hierarchy nodes. The functions are written to apply across multiple calculations. You configure the input and output parameters to meet the requirements of your device type or your environment before you use a function in a calculation.

If the built-in functions do not meet all of your requirements, write your own custom functions. You can add expressions and simple functions by pasting the code into a function in the user interface. If your function is too large to paste into Maximo Monitor or if your function has multiple outputs, store your function in a GitHub repository and register it with Maximo Monitor.

In Maximo Monitor 8.10 and later, streaming data metrics support custom functions only for ONNX models.

In Maximo Monitor 8.9, streaming data metrics do not support custom functions.

Functions are run automatically as part of a function pipeline. The output of some functions feeds into the other calculations. Maximo Monitor determines the order in which to process calculations.

Sources of data

One source of input data is metric data from IoT devices that was ingested by the IoT tool and stored in the data lake. Other sources of input data are IoT and non-IoT data that you load for devices into the data lake, for example by using data connectors.

Modeling data

Devices are the digital representation of items in your organization that either produce data or that you would like to track by using data. For example, devices can be vehicles, industrial equipment, homes, or workers. Devices in the IoT tool are modeled as devices in Maximo Monitor.

Adding calculations

Explore the function catalog to see the range of built-in functions that are available to you. Create some sample data and explore the calculations that are associated with each template. For more information, see Built-in functions.

A set of built-in, sample, and base functions are available in a companion package, IoT Functions, in GitHub Enterprise. You can derive your custom functions from the functions in this package.

Apply functions from the catalog to a device types and hierarchy nodes. Configure the inputs and outputs of each calculation to adapt the calculation to suit the device type and your environment. You can aggregate your view of the metric data by using time-based granularities, such as weekly, daily, monthly. Alternatively, you might use a dimension, such as manufacturer or location, as a grain.

Maximo Monitor builds a pipeline of functions. For example, you might configure two functions, ƒ1 and ƒ2. If ƒ1 is used in the calculation of ƒ2, ƒƒ is run first.

For more information, see Working with calculations.

Adding alerts

You can create alerts and configure them to trigger when your data reaches or exceeds a specific threshold. For more information, see Alerts.

Detecting anomalies

You can use some of the built-in anomaly detection functions to find patterns in your time series data that does not confirm to expected behavior. For example, you can use the NoDataAnomalyScore function to alert you when your sensors are no longer sending data. You can add alert information to your line graph cards to easily find the data points that are anomalous. For more information, see Detecting anomalies.

Anomaly simulator functions are provided in the function catalog for simulating anomalies. You can use the simulator to become familiar with unsupervised anomaly detectors. For more information, see Simulating anomalies.

Adding dashboards

Visualize your input, metrics, and alerts for your devices in monitoring dashboards. Configure dashboards for hierarchy nodes for business users to monitor their line-of-business KPIs. Configure device dashboards to monitor metrics and alerts for individual devices. For more information, see Monitoring data.