15th of May 2018 in Moscow at the Microsoft Technology Center “Consyst Business Group” held a presentation about the new service of predictive analytics Consyst Smart Service. The service is intended to predict the repair of equipment and uses the technologies of Machine learning and IoT.

Experts of company “Consyst Business Group” presented their approach to solving the problem of equipment repair planning. At the event a new smart service, Consyst Smart Service was demonstrated, which is built on the Azure cloud platform. Using data about the real state of the equipment, obtained from the sensors installed on the equipment, the Consyst Smart Service allows you to configure and train the prediction model of the equipment failure date, and then transfer the received data to ERP system.

The solution is integrated with the Microsoft IoT Hub service and uses it to aggregate data from sensors installed on equipment. After an online analysis of the received information, the filtered data enters the Azure Machine Learning forecasting service, which allows you to make breakdown forecasts. Due to the flexibility of the service setting, predictive analytics can be performed for different equipment and using different forecast models.

About the new possibilities of the solution tells Efim Fish, deputy director of Microsoft division on business development “Consyst Business Group”: “The Consyst Smart Service solution provides enterprises with real help in digital transformation. It allows to increase the reliability of the equipment, which is especially important in a highly competitive environment, in which all companies now operate.  Such an effect is achieved through the use of modern methods for predicting equipment failures using data that we get using IoT, the Internet of Things. Consyst Smart Service allows to minimize operating costs by avoiding unnecessary repairs and unnecessary maintenance, and ultimately increases the enterprise’s willingness to reject equipment due to timely forecasting and preparation of resources”.

The demonstration of the solution took place within the framework of the conference “Digital Transformation in action” organized by “Consyst Business Group” in cooperation with Microsoft Corporation. The event was dedicated to the questions about the implementation of the digital transformation process in enterprises of the real economy and gathered at Microsoft Technology Center top-managers, IT directors and development managers of industrial and operational companies. The speakers of the conference shared their practical experience of digitalization of production processes with participants and showed how solutions using machine learning, the Internet of Things, cognitive services and other modern technologies help enterprises to increase the added value of products and successfully compete in the market. Special guests, Tatyana Delyagina, director of marketing for the Microsoft cloud platform, and Oleg Belousov, representative of company Wonderware Russia, also made speeches at the conference.

Vladimir Egorov, vice president for strategic development of “Consyst Business Group”, described the latest trends in the digital transformation of enterprises in the real sector this way: Trend number one is, of course, an increase in the intelligence of equipment and devices, that are involved in the industry. The cheaper cost of processors allows them to be embedded in more and more devices, and they begin not only to form data flows, but also to process these data.

The second main trend of digital transformation is the increasingly widespread distribution of the decentralized model of computing. If so far all information systems were built on the principles of client-server, today, increasingly, data processing tasks are distributed between services that can be placed both at the customer and in public clouds.

Thus, now, gathering more and more data and using distributed computing, real sector companies are able to process more data in less time and for less money, radically transforming existing business processes”.