Halton Group is a family-owned, global technology leader in indoor air solutions for demanding spaces. The company designs, manufactures, and delivers indoor environment solutions across industries from food service, healthcare, public spaces, marine, to offshore, and more. Halton’s mission is to enable people’s well-being in these environments. The company aims to enable safe, comfortable, and productive indoor environments that are energy-efficient and comply with sustainable principles.
“I have enjoyed working with Zure. The cooperation has been easy-going and informal.”
Jesse Jämsén, IoT Solutions Manager, Halton Group
Challenge
Halton wanted to collect all their time-series data and store it for long-term storage for research and development purposes, end-customer benefitting solutions, and potential other usages such as predictive maintenance. They had started developing an IoT platform for collecting data from their hundreds of thousands of censors worldwide but wanted additional expertise and resources for building a solution for storing and visualizing the data. Halton needed a reliable cloud-based solution to store their data, provide a graphical user interface to visualize the data, and an API to provide access to data via programmatically accessible interfaces.
Goal
The goal was to store all of Halton’s IoT sensors data in a time-series data set in Azure. The amount of data is vast as the system receives thousands of messages every minute and over 165 million messages daily.
During the project, Halton was able to clarify their goals for the data’s main purpose further, and recognized that the data is especially valuable for product development, predictive maintenance, customer reports, and an alerting system. Halton also wanted a graphical user interface to visualize the data coming in from their now over a million sensors on their devices worldwide. The system’s performance needed to be top-notch as only a break of few minutes causes a big accumulation of data.
Implementation
Halton had chosen Azure as their cloud platform. The company relies heavily on Microsoft’s stack so a strategic decision to build everything on Azure guarantees a seamless environment for the company’s operations. Azure also provides a reliable and secure place for Halton’s time-series data.
Halton had developed an IoT platform to collect data from their IoT devices but did not fully utilize the possibilities Azure offers. Zure developed a solution that ingests and processes the incoming data, ending up in an application, where it could be visualized and accessed programmatically. A console application was developed to read raw IoT data from the data broker and send it to Event Hubs from where it is further processed.
An API was developed which allows RND personnel to access all relevant information and data coming in from the monitored spaces. In the future, Halton can also offer their customers access to information and data regarding their own environments, such as a professional kitchen or an operating room. This work goes beyond just reporting and could potentially offer restaurant owners a highly informative view of their space, and where they could improve energy efficiency and safety.
In addition, a user interface was created to control data sensors and collect data. The interface is simple to use and offers a traffic light-inspired view of the sensors’ status. Halton Edge device reads sensor data from local automation devices, processes it, and sends it to Halton’s cloud services. Since not all sensors have network connectivity, Bluetooth and other similar technologies were used.
To complement and further utilize the collection of the IoT data, a solution was created to monitor data in real-time to enable sending out alerts when target conditions are met.
The solution was used to pick up alarm-related events from the data stream, but it can also be used to detect, for example, unexpected temperature variations and other situations. The alert trigger is sent from an application which can then be used to send out an alert, such as an SMS message, to relevant personnel.
Challenges during the Project
The project faced challenges when Microsoft announced that it would discontinue the chosen time series data platform solution. This meant that a new tool was needed for preserving and visualizing the time series data. Together with Halton, Zure did an assessment and research to figure out which could be the best replacement solution.
Due to this sudden turn of events, the initial architecture was renewed, and the initial crucial application was replaced with a new one. The new application does not offer a readymade data visualization tool, but it offers the possibility to make your own graphs.
This external issue was a hinder to the project but as a replacement was found and taken into use quickly, the project was soon back on track. During the change, over 70 billion rows of historical data were also migrated to the new system.
Achieved Benefits
As a result of the cooperation, Halton was able to store all their time-series data securely and reliably. The company has been able to use this data for product development, customer reports, and predictive maintenance. The solution also saves Halton on their travel costs now that the IoT package is included in most of the system deliveries.
Additionally, a graphical user interface was built to offer the possibility to show restaurant owners data related to their kitchen ventilation solutions. The alert engine sends out alerts based on certain thresholds, for example when an elevated fire risk is detected, to prevent dangerous situations and to improve the longevity of the devices.
Plans for the Future
Halton plans to continue working with Zure in the future. They are currently looking to continue to use Azure for their services and are open to exploring new ways to use their IoT platform to provide value to their customers. Constant development is continuously going on with the developed solution to guarantee optimal costs and processes.
