Hot line phone +7 495 660 75 89


Massive Forecasting: The Industrial Sector Implements Predictive Maintenance System

28 April 2018
Massive Forecasting: The Industrial Sector Implements Predictive Maintenance System

Optimizing Mistakes

The new digital service Predictive Maintenance (PdM) becomes more and more popular in production companies. The service is implemented in order to collect ans analyze the information on equipment condition for further predicting the terms of maintenance and preventing any production failures.

As distinct from the traditional preventive maintenance, the prediction is based on the data array and digital models rather than on average statistics. Collection and processing of current information, early detection of malfunctions and errors and optimization of resources are among the main PdM components.

According to Markets and Markets, the global market of Predictive Maintenance will reach US $ 1.9 billion  in 2020 (against 582 million in 2015). Among the global key players – General Electric, Siemens, ABB, Emerson, IBM, SAS, Schneider Electric and other.


Successful examples of PdM implementation will be showcased in the framework of theme track “INNOPROM.METALPROCESSING” at INNOPROM - 2018.  The exposition will present  metal-working equipment and the companies may offer their products to a big number of business visitors. Every year, more than 50 thousand visitors from 95 countries of the world visit the exhibition.

More than 20 thousand decision makers from mechanical engineering companies, which are interested in purchasing metal-working equipment and robotic automation of their facilities, from industrially developed Russian regions and CIS countries visit the Industrial Forum of the Specialized Exhibition every year.


Forecast for Industry

Predictive analytics in Russia is still on an initial stage: according to expert estimates, the market in 2017 amounted to US $100 million, in one year this figure may grow to 200 million. Currently, the industrial sector increasingly applies PdM.

According to Konstantin Gorbach, director for intelligent applications at Cifra (part of Renova Group), the objectives of predicative maintenance are relevant to the clients using sophisticated and expensive equipment. These are the industries where a failure of one element leads to substantial losses and endangers the safety: fuel and energy complex, metallurgy, petrochemicals and transport. Early detection of problems enables preventing possible accidents and reduction of costs.


"Remote monitoring and predicting system makes the maintenance of equipment more transparent for the management,"

explains Maxim Lipatov, technical director of PRANA equipment condition prediction system at ROTEC JSC.



The cost of PdM implementation in different companies is different: the pilot projects may be worth several million rubles and their commercialization – a few hundred million.


"Over the past couple of years, the interest towards predictive analytics in Russia has grown. The main players on our market are ROSTEC with PRANA system, Datadvance, Clover Group and integrators. In our practice, we had both the pilot projects at the price of 10 million rubles and the industrial introductions at 1 million US dollars",

tells Sergey Morozov, general director of Datadvance.


Protection for a Gigawatt

In 2017, power generating companies have joined the PdM club. For example, T-Plus Holding has signed a contract with ROTEC for implementation of PRANA predictive system on 16 power generating units. ROTEC is connecting turbines, waste-heat boilers and gas booster compressors of the generating company to its own situation center. According to the management, this solution will help the energy engineers to reduce repair costs and downtime periods and find design defects of equipment well in advance.

"Ten power plants of the company will be protected from any technological risks. This is a very important step towards large-scale digitization of power engineering: the total power of the equipment connected to PRANA system exceeds 3 GW,"

– clarified Mikhail Lifshitz, Chairman of the Board, ROTEC.



Certain metallurgical companies also begin applying the technology. Gold-mining company Nordgold has established a repair system using PdM. However, some practices require investment and maintenance costs in the future, says Alexander Brezhnev, mining machinery maintenance manager of the company.

The service began to be applied by PJSC Severstal — at the Cherepovets steel mill. PdM has been put in operation in order to reduce the downtime on the hot rolling mill 2000.


"Predictive model identifies the likelihood of pinion stand bearings overheating — one of the most frequent and costly causes of the unit shutdown. It is the first model in the area of predictive repairs, that has been implemented at the production site of the Cherepovets Metallurgical Works in the framework of the company digital strategy," pointed out the Severstal press service.

The company experts have developed a digital model for obtaining the data from temperature sensors and prediction making. In case the indicators obtained from the mill deviate from the norm, an operator receives an alert. It makes possible to prevent an unscheduled shutdown.

"With the predictive model computations, we expect to reduce the downtime by 80%. We also plan to introduce the similar models for other types of shutdowns on the mill-2000, as well as on other units,"

commented Sergey Dobrodey, director for repairs of Severstal Russian Steel division.


Unreliability of a Logbook

According to experts, the PdM service is only appearing in the RF regions and the customers have not yet realized its potential.







"The economic effect of predictive maintenance implementation may reach hundreds million rubles, if we take into account Gazprom and RZhD,"

estimated Alexey Shovkun, director for consulting of Datalytica.


The developers mention several barriers that hinder a wide overreach of new technologies today. Thus, the major part of production machinery is not equipped with data transmitters, and production plants have no data collection and online monitoring systems. Moreover, the defects and repair logbooks are often kept inaccurately. Unpreparedness of personnel to IT solutions and rejection of new servicing concept constrain the implementation of PdM systems at Russian industrial companies.

Predictive technologies are one of the elements in the production digitization. At the specialized exhibition area "Industrial Automation and IoT" during the days of INNOPROM-2018, the visitors will be able to get acquainted with new industrial digital projects including software and hardware systems of production infrastructure and resource management automation, industrial robotics, embedded systems and system integration technologies. We invite you to take part in the show.

All news
  • Sponsors
  • Partners
  • Media partners
  • Business program partners
  • Organizers

  • Operator
  • Recommended builder
  • Service Partner