The Movement Set Statistics widget is a powerful and flexible tool, allowing users to monitor and analyze directional movements without requiring geo-registration of cameras or merging their views into one instance.
Replacing the older OD Matrix widget (available until version 1.18), it supports live and historical data processing, enables highly customizable data filtering and access through the FLOW API.
The widget is ideal for monitoring:
Intersections
Roundabouts
Other locations that require movement breakdown or traffic intensity analysis
Preparing the Analytics
Before configuring the Movement Set Statistics widget, the user should perform a structured analysis of the monitored site to ensure that the configuration will be correct. Following the steps below will make widget configuration easy, reliable and scalable.
Naming Guidelines
Consistent and systematic naming is crucial for the functionality of the widget. Mind the following tips for further configuration.
Element | Naming Suggestion | Example |
Site | Name of the monitored location or two main streets (unique for the city) | Shortman x Sport |
Analytics (Camera) | MainStreetA x MainStreetB (unique in the system) | Shortman x Sport 2 |
Site Arms | Street names (unique on the site) | Sport, Sport (Stadium), Shortman |
Gates | Named after the arm, add numbering if needed (unique in the analytic) | Sport1, Sport2, Sport (Stadium), Shortman1 |
Movements | [EntryArm] -> [ExitArm] (unique on the site) | Shortman -> Sport |
Events Widget Name | Inherits the movement name + suffix " - Events" | Shortman -> Sport - Events |
ℹ️ Tip: Usually, multiple cameras become data sources for a single site analysis. Add the suffix " - cartogram" to the first relevant analytics name to simplify later filtering.
Analyze the Location
Name the monitored site and define names of its arms (entry/exit points).
Analyze Valid Movements
Identify allowed and excluded movements. You can mark the list of valid movements either graphically or as plain text, listing the permitted origin–destination pairs. In this scenario, all three basic movements (left, right, straight) per arm are permitted.
Choose Suitable Analytics
Choose an appropriate camera for each monitored movement, based on the camera’s view.
Configure Gates and Movements
Set up entry and exit gates and define the movements in the selected analytics.
Define Source Data with Event Widgets
Place all movements (or gates, depending on the setup) onto the canvas in the FLOW analytics interface. For each movement or gate on the canvas:
Attach an Events widget.
In the Events widget Settings:
Set the Operator Attribute to Trajectory Events.
Check and confirm other configuration options.
Apply the settings.
Repeat this process for each monitored movement to make them valid data sources for the Movement Set Statistics widget.
Widget Configuration
After the preparation and data source definition are complete, you can proceed with the configuration of the Movement Set Statistics widget within the FLOW Dashboard.
The widget settings is configurable as follows:
Parameter | Description |
Widget Name | The name of the widget. |
Granularity | Specifies the data collection interval (e.g., 5 minutes, 1 hour). |
Arms | Define the intersection shape: number of arms, indexing, naming, and coloring of the arms (by entry). |
Statistic Movements | Define all monitored movements by selecting valid combinations of entry and exit arms, choosing the event type (Movement, A to B, A minus B), and selecting the correct data source (by the Events widget name). |
Visualization Type | Select one or more visualization modes: CartogramView, TableView, Time Series Visualization. |
Sequence Number Mode | Define which data set is used for the calculations. |
Tags | Tags can be applied if defined by the user. |
Choosing the Correct Data Source
Correct selection prevents data mismatches, especially in projects with multiple analytics using similar movement definitions.
Always use the search and filtering options in the selection dialog.
Carefully verify the Analytic ID where the Events widget resides and its Name.
Time Range Filtering Options
Comprehensive filtering options are available to only present the data of interest. The widget also displays the currently applied date/time range and granularity for transparency.
History
A toggle to switch between live (current time block) and enabling historical filtering options.
Overall Date-Time Range
Define the start and end date/time or a specific range. The final time range will respect the data collection granularity (any overlapping time blocks will be fully included). The maximum allowed interval is 10 days.
Allowed Time Formats are:
yyyy-mm-dd
yyyy-mm-dd hh:mm
yyyy-mm-dd hh:mm - yyyy-mm-dd hh:mm
Also, abbreviated form, e.g., yyyy-mm-dd hh:mm - hh:mm
For a full day/time range, the first date must not be later than the second
Daily Time Filter
Refines the selection from the Overall Date-Time Range with daily time conditions:
24 – default, full day.
AM – filter from 00:00 to 12:00.
PM – filter from 12:00 to 24:00.
Custom Filter – user-defined hourly and minute range.
Peak Hour Filter
Identifies the 60-minute interval with the highest traffic intensity within the selected date/time filters:
The peak interval does not need to align with full clock hours.
The calculation respects the configured granularity.
All widget elements will reflect only this peak hour data.
Visualization Types
Multiple presentation formats are available, reflecting different needs from quick visual checks to detailed numeric reporting.
Cartogram View
The Cartogram View provides a graphical representation of the directional flow at the monitored site:
Represents the shape of the intersection, arm names and numeric indexing.
Visualizes weights of specific movements and presents the total counts/subcounts per each.
Reflects user-defined colors for each arm.
Table View
The Table View displays a list of all defined movements, including:
Counts and subcounts for considered object categories.
Percentage shares of counts and subcounts relative to the overall detected traffic.
Reflects user-defined colors for each arm.
The table is split into two sections:
ENTRIES (left-hand side): Data sorted by entry gates.
EXITS (right-hand side): Data sorted by exit gates.
Time Series Visualization
The Time Series Visualization shows traffic intensity over time, presenting how movements fluctuate across the selected period.
Supports two modes:
By Arms: Displays the distribution among the arms of the intersection (incoming traffic), color-coded.
By Types: Displays the distribution between total counts and subcounts.
Export Options
The widget provides several export options for data extraction and reporting:
Export to Image: Captures a screenshot of the current widget view, including all visual elements and applied filters.
Export to CSV: Offers raw export of collected data, detailed per categories, by granularity, and with the option to aggregate time granularity to bigger time block overviews.
Export to PDF: Multiple Image exports per widget.
PDF Print Reports: Multiple widgets’ data into a consolidated report for a given time range.
What Next?
Learn more about widgets and sinks in FLOW, explore time processing, or check out Trajectory Events widget, or other widgets. Use the FLOW Public API to access data from widgets and sinks and integrate it with your external systems.
Make your traffic analytics smarter and more connected with FLOW.