The development and expansion of production requires constant monitoring of the huge amounts of data that come from sensors installed on industrial equipment. Today it is very difficult to supervise production points or divisions without the introduction of IoT technologies (Internet of Things). Transition to such technologies can be carried out on the basis of the Microsoft Azure cloud analytical platform, with the help of which a system for managing structured and unstructured data flows (data stream) is developed and implemented in 24/7 mode.
One of the leading companies in Ukraine producing water vending machines was able to completely reformat the work of the factory complex by introducing a system for monitoring technological indicators and the depleted resource of industrial equipment based on data collected from devices.
The main problem of the company was the lack of understanding of how to simultaneously track the performance of various devices and organize a clear data tracking system, as well as prevent unexpected equipment failure. In addition, the management of the company was faced with the task of ensuring sustainable growth in sales and increasing the number of regular customers, that is, the creation and development of a trading network that brings maximum and predictable income. To monitor the equipment and control the level of sales, IoT sensors were connected to the devices, which made it possible to automate the collection of data in real time.
Metrics from sensors are processed by the Azure Event Hubs data streaming platform — part of the Microsoft Azure cloud solution, which allows you to receive BIG DATA streams and display them on dashboards (Power BI) in real time (Hot Path, real-time analytics). In addition, the company decided to create a cloud DWH to collect, store and then analyze the collected indicators (Cold Path, historical and trend analytics). For this, automatic loading of incoming data into Azure Blob Storage and Azure Data Lake Storage was organized, which in turn made it possible to consolidate information and identify patterns that lead to failures or breakdowns of equipment, as a result – to prevent equipment failure. But most importantly, the information received from devices in live mode helped to install vending machines in places where they bring maximum profit.
Thanks to this decision, it became possible to analyze the feasibility of placing them in various retail outlets, compare the level of profit and understand the reasons that lead to certain commercial results. Thus, the company’s management got the opportunity to directly influence the sales volume by adjusting the number of vending machines in shopping centers and, if necessary, changing the location and quantity.
The company plans to implement the Microsoft Azure Data Bricks, namely, the development of machine learning models (ML-models), which will be able to give recommendations on the optimal timing for a maintenance of machines and determine the most profitable locations for their placement.
Thus, the Microsoft Azure solution made it possible to automate equipment diagnostics, analyze the causes of high (low) loads on the industrial complex and understand the reasons for the downtime of production facilities, as well as create a steadily growing trading network with a high level of income.
- Now it is possible to track critical equipment indicators in real time;
- The number of manufactured equipment increased by 35%;
- Sales increased by more than 50%;
- Production costs decreased by 40%.
Microsoft Azure Synapse Analytics