How do you visualize a massive amount of sensor-collected data quickly, easily, and effectively? There are many ways, but Power BI is one of the easiest and fastest. Wirepas and Microsoft teamed up to create a proof of concept that would show the power of Wirepas supported by Power BI Embedded.

The solution simplifies displaying massive amounts of sensor data collected for a parcel tracking service. Which parcel is at what location and when? What is the parcel’s status?

All data is collected with the help of Wirepas IoT technology.

Key technologies used

Core team

Customer profile

Wirepas

Wirepas focuses on providing the most reliable, optimized, and scalable device connectivity to its customers. With Wirepas, customers can digitalize their current business processes and innovate for new disruptive models. There is no need for traditional repeaters with Wirepas Connectivity because every wireless device is a smart router of the network. The connected devices are the network—easy as that.

Wirepas has its headquarters in Tampere, Finland, and offices in France, Germany, South Korea, and the United States. Established in 2010 in Tampere, Wirepas has its roots in Tampere University of Technology where a 10-year research project was conducted around the radio frequency mesh technology.

Problem statement

Wirepas technology collects a wide variety of data through its connectivity service. Every wireless device built on Wirepas software technology can collect and send a huge amount of data. This data is collected in several ways in databases and files of the customers. Visualizing this data is key to getting an overview of the current state of “things” tracked by the technology.

The end users of the developed solution are the Wirepas customers. They need an “interpretation” of the collected IoT data to get an overview and to make easier decisions.

Solution, steps, and delivery

  1. Acquire the data.

    The data collected by the Wirepas technology is collected in customer “storage.” This can be files, databases, and more. A supporting customer gave us access to their data. This data has to be imported and cleaned up. For the proof of concept (POC), we used Power BI directly.

    Also for the POC, we imported the data from a CSV file, which was exported from a customer’s database. In future versions we will connect directly to the database with DirectQuery; in the meantime, we will import.

  2. Clean the data.

    The data has a domain-specific model. This has to be cleaned up and converted to a human-readable format. We did this with Query Editor. We gave meaning to plain IDs to enable readable values to the end consumer, such as “1000” –> “Region South”.

    Power BI Query Clean up and Conversion


  3. Design the report using Power BI Desktop.

    After importing and cleaning up the data from 1 million sensors, we used Power BI to design dashboards that enable the customer to get overviews of all data and enable them to drill down to one single parcel and the parcel history.

    Power BI Design


    Some impressions of the dashboard.

    Power BI Design


    Power BI Design


    Power BI Design


  4. Create a Power BI workspace.

    To create the Power BI workspace in Azure, we used the powerbi-cli. Currently there is no UI available to create workspaces for Power BI in Azure. Therefore, we used the Power BI command line tool for managing Power BI Embedded workspace collections.

    Power BI CLI can be easily installed via NPM:

     npm install powerbi-cli -g
    

    Creating a workspace is done with the following statement:

     powerbi create-workspace -c wirepasreports -k <key_to_powerbiembedded_in_azure>
    


  5. Upload PBIX-File.

    To upload a Power BI report to this workspace, you only need an additional command of powerbi-cli.

     powerbi import -c wirepasreports -k <key_to_powerbiembedded_in_azure> -w <the_name_of_workspace> -f <the_filepath_to_the_thereport> -n Overview -o
    


  6. Embed Power BI in a web app.

    Power BI Embedded enables developers to embed reports in almost every kind of app. The easiest way is to embed the report into a website. To get a website, we used the Web Apps feature of Azure App Service to host a simple ASP.NET MVC app based on a standard template for Power BI Embedded (A Power BI Embedded sample that shows you how to integrate a Power BI report into your own web app).

Architecture diagram

Example below:

Power BI Embedded Architecture Diagram


Next steps

The running solution is Wirepas’ first Azure project using Power BI Embedded. Wirepas is fascinated with Power BI Embedded on Azure and the ease and speed of implementation. We are already working on the next steps—pushing the sensor data to Azure IoT Hub and, with the help of stream jobs, to SQL Database and also for custom analyzing tools to Blob storage as well. Blob storage is triggering Azure Functions to send messages to customers such as “Your parcel has passed station xyz.”

For Wirepas, this was proof of the power of Microsoft technologies, delivering IoT data fast and easy to consumers. Wirepas made the first public presentation of this solution at Bosch Connected World.

Conclusion

This was a smart way for Wirepas to bring their IP to the cloud in an easy and fast implementation. The whole project needed only 10 hours of calls, consulting, and implementation.

For Bosch Connected World, this was an easy demonstration of complex data based on Azure and Power BI Embedded.

Since the workshop, Wirepas has won several new customers who are using its products and dashboards powered by Power BI Embedded.

Learnings

  • It is very easy to bring complex data to Power BI and convert it to easy data for faster decisions on the web.
  • One learning by Microsoft and Wirepas was how to handle the amount and type of data that we structured and converted. Query Editor is an awesome tool for ETL processes.
  • Wirepas was surprised by the speed and friction-free tooling and uploading of dashboards.

Opportunities going forward

  • After demonstrating the fast and easy process of creating and deploying dashboards to the Azure cloud, the next steps are to enable real-life data by connecting customers’ databases directly to Power BI Embedded instead of manually importing data.
  • The evaluation of the Azure IoT Hub is also on the roadmap, which could be used in different customer scenarios.

Customer quote

“We have used Power BI to reformat millions of lines of sensor data for use in Azure. Handling Power BI was extremely easy, and results are visible on the fly. Microsoft Azure immediately took this data for visualization and we were able to analyze our data on abnormalities right away.”

— Rudolf von Stokar, Country Manager, Wirepas

Additional resources

Documentation

GitHub repos used