The practice of implementing the EAM – an interview with the head of the direction Smart.EAM
Kirill Kostanetsky, head of IT-Enterprise Smart.EAM, gave an interview to APPAU about the latest implementations at Interpipe enterprises. The interview shows the key trends of industrial development, and also speaks about the need to change the culture of enterprises and new requirements to the level of competence of employees. Here it is also clearly stated about the importance of automated data collection and docking of the EAM-MES levels with the process control system.
– Cyril, why did the EAM project start in INTERPIPE?
The main initiator of the project was the Director for Finance and Economics. Before him was the task of achieving transparency in understanding the structure of working capital and optimizing their size. Summing up, I will say that this is a typical problem of the procurement process for any company – namely, how to optimize applications for material resources to ensure the smooth operation of production. Such as the maintenance of critical components in the warehouse (for example, bearings), and consumables (like oil). The procurement and finance services always try to optimize purchases in terms of finances. We are told some curious cases about the potential for development here. I remember one gos. an enterprise where the procurement department went to an experiment – they took and deleted everything from the procurement plan, and they did not buy anything for a month. Surprisingly, the plant worked, as if nothing had happened. This example shows what “urgent necessity” of goods means, which can come from the production workers.
Returning to the project in Interpipe, I will say that similar problems arose here, and the management of the company asked how to manage this process. “Here we come to the warehouse, we look – there are bearings. We ask exploitation: “And why two more bearings, if already 3 lies?” – “Well …” There is no clear answer. “And for what equipment this part or part?”, “And what is the life of this car or the parts?”, “And when was the last time changed?”, “And what is the normative …” – the answers to all such questions were clearly not convincing. Then they finally had the idea of a management system.
– Have you already implemented systems at that time?
Yes. The management system of OPF (note the main production funds) has been working in IT-Enterprise since 2002 and has been implemented already in dozens of enterprises. By this time, the direction of EAM has become progressive and we have already completed the first modern project at the Poltava GOK. The new solution itself has perfectly demonstrated in practical implementation. Interpipe NIKOTYUB was the first factory of the company, which began to introduce our EAM. Then there was INTERPIPE NTZ. In Nikopol, they started from one shop, then – they were replicated to the whole enterprise.
– And then – INTERPIPE STEEL?
Yes. On INTERPIPE STEEL, we saw completely different management possibilities. There were already developed automated process control systems and MES. In other words, there was already a lot of information about how the equipment works and worked before. This information is used to control the production. For us it is an opportunity to manage the equipment maintenance processes much better. We were able to contact the Q-MET database (MES from Danieli) and get access to information from 23 thousand sensors. This allows us to evaluate more than 1 million events on equipment for 1 day. For example, the system automatically records the operating hours of more than 100 pumps of energy services.
– And what was the state of normative-reference information (NSI) before the beginning of the project?
If we talk about INTERPIPE STEEL, then, of course, at the enterprise part of the NSI was implemented. You know that it was a very big project and the whole team that implemented it did a good job. But something was, something was not, something was translated, something was not. It was obvious to us that the documentation necessary for subsequent maintenance was secondary to the priorities of the project. After all, the main thing was to start the plant.
Therefore, our task on the NSI was just to describe the many different types of equipment from suppliers. More precisely, it is more than 6 thousand technical places and almost 1,5 thousand objects of repair. For example, a bucket of this kind consists of eleven elements. Some of them need to be checked once a half a year, others – every month, etc. Something needs to be changed periodically. Thus, the descriptions turn into rules in the system. The system will then itself remind, prompt, guide. For example, this week – “you need to change the gasket in the node X of the unit Y”.
– Did you make these descriptions or the customer yourself?
We are deeply convinced that the data and regulations must always be entered by the customer. If this does not happen, the system will not work – the customer will be too dependent on the artist and / or will not learn how to work. In addition, the customer better understands his own equipment and knows him at times better performer.
In addition to integrating with data from MES- what else did you implement?
That the information could be used in business processes of service, it should be prepared. We usually introduce 3 levels of changes:
- Collection of information. We do not need information from the sensor every second. It is much more important to understand the long-term trend (changes) in the behavior of a particular piece of equipment. And these accumulated volumes of information are constantly growing.
- Preparation of information for storage and analysis. We use a distributed architecture, where information is stored on different servers. The information should be stored in a special way to take up less space so that it can be analyzed more quickly – there are dozens, and in the near future there are hundreds of terabytes of data. This information must be carefully processed. For example, when you start – stop the equipment, pass – large fluctuations of certain physical parameters. The trend of these parameters (temperature, current, etc.) can lead to false conclusions and false alarms. All these are pre-treatment tasks. They are quite nontrivial given the large amount of equipment in the plant as a whole and various possible states. All such situations the system should clearly distinguish and not allow false conclusions.
- Data processing – the construction of trends, the use of mat. statistics, issuing recommendations and visualization.
– How are things today with preventive and predictive diagnostics?
Let’s start with the preventive one. 80% of problems can be closed by introducing the so-called “yellow zone”. If the indicator in it approaches the border of the “red zone”, it means that it’s time to intervene, the system should warn about the upcoming breakdown. This is how preventive diagnostics works.
There is a second part of the problem, where the boundary between the yellow and red zone floats – this is about 18% of the situations. “Floats” for various reasons. It can be different types of dependence – linear or exponential. If it is exponential, the problem may be tomorrow. If the linear one – in a week. It depends on how the indicator has changed in the past. To understand this, we use mate. the statistical apparatus – trends show when the equipment enters the red zone. This is already a forecast with the help of a mat. apparatus and statistics.
There are 2% more, where we can not use the mat system of statistics. We need methods of machine learning. This is in cases where we can not describe these links. The system itself trains and builds these ties. The human mind can not embrace it. We can not yet talk about 100% mastery of these competencies, but we are rapidly moving in this direction.
– How did you describe these situations in the regulations with the red and yellow zones?
Zone parameters are set by the employees of the customer company. This technology, first of all. But there is also documentation from the manufacturers. Parameters of normal operation and anomalies, as a rule, are indicated there.
– How are the EAM systems developing in the world?
EAM is a category of software for the management of a manufacturing enterprise. All EAM systems operate according to the same rules. It can be said on the logic of common sense. The main way of planning has always been the planning plans of the PPR. Under which you need to pull up resources and materials. That’s why at first the plans were considered PPR, and the systems were called CMMS.
What’s new now? First and foremost, there is an emphasis on diagnostics – any decisions to stop equipment, maintenance should be based on specific measurable performance indicators. Therefore there is a rapid development of diagnostic devices (sounds, vibrations, temperature, electro-indicators and even smells). Diagnostics can be collected by staff or automatically … The starting point of the maintenance process is an automated system for collecting information. Previously, many parameters were collected manually. Now everything is automated. This excludes the notorious “human factor” and a host of errors related to this. We underestimated the accuracy of planning and the conditions under which the mass of information automatically enters the system. Today we see that this radically changes the accuracy of planning, allows you to track micro-trends, free up many people, etc. This is a fundamental difference. That is, the starting point for these processes in the EAM today is information from the process control system.
If we talk more about changes, then with the advent of preventive and predictive diagnostics, the associated functional changes.
– What indicators or results do you consider to be the main ones in this project?
First, let’s look at the economic indicators. In Interpipe, this is a reduction in spare parts stock for current and capital repairs and a reduction in the share of low-turnover spare parts. We have reached a 10% reduction, and this is millions of dollars. But the main effect of the project will come in the future, when all systems will be set up and statistics will be accumulated.
If we talk about trends in the world, then everyone goes into operational statistics. Heads of enterprises began to consider money very carefully – it is impossible to remain cost-effective and allow yourself to have classical PPPs in full.
And how do we conduct a PPR?
At us on the majority of the enterprises the human factor while dominates. That is, Uncle Vanya comes and, using his many years of experience, says: “It will last another six months, it’s 3 months, it’s already necessary to change,” etc. It is interesting that such an approach is no longer a classic PPR. Classic – it’s like a car – after 10-15 thousand miles, we change the oil and filters. Such classical PPRs did not remain at many enterprises, while the majority of others were dominated by the “Uncle Vanya effect”. But you understand the shortcomings of this approach – because everything now depends on the qualification, and maybe even the purely subjective factors peculiar to the specific “uncle Vanya”.
– It is clear with economic indicators. But the middle manager often asks “How will this change in my work and my services. For example, should I now cut them down by, say, 20%? “
This is a very subtle question. In practice, the leaders of the level of the chief engineer want to retain, rather than reduce their people.
Usually we give the following answer: “Experts should stay. But the profile of their work is changing – they have to do more analytical work. ” The modern EAM significantly increases the requirements for specialists! Earlier – it was a mechanic who was the first to run after the incident and knew how to make quality welding. Now – this is a mechanic who knows modern materials, understands lubricant, causes pressure drop, understands instrumentation and that can analyze and diagnose cause-effect relationships with the help of a computer. Therefore, people should be retrained.
– Nevertheless, do you see in the practice of the EAM a bunch with the requirements of the leadership? For example, we heard from metallurgists about 2 growing trends – the requirements to reduce people, as well as the massive independent layoffs of specialists – they leave the country. Both are related to the pace of implementation of such systems as the EAM.
Trends, which we see by enterprises at their level, are more likely from the leadership. In some places there is pressure to reduce people. But, in my opinion, this can lead to the fact that there will be no one to service the equipment. After all, whatever you say – the rate of obsolescence of productive assets in our country as a whole is higher than the pace of modernization. Therefore, it seems to me that the solution should consist in the parallel implementation of such systems. But people should be trained and trained. There should not be a situation where the reasons for stopping the machine are diagnosed within 6-8 hours – this should take minutes. Of course, at the same time such specialists should be paid with dignity. And such changes are taking place in a number of enterprises.
But in some enterprises, in my opinion, there are no clear policies of communications in relation to such systems. I repeatedly heard “Here we will implement this system and I will be fired. Why do I need it?”. On the other hand, and if you look at it objectively – the information system allows you to see the real load of people, their employment and efficiency. By experience, we see a completely different situation in the servicing at different enterprises of one industry – this is different at times. For example, at one plant the equipment operates without stopping, on the other – it stops every month.
– You mentioned that a lot of statistical data is being processed. Are there any patterns already visible that allow you to optimize already – to improve work in MRO?
Examples of improvements are many. For example, on INTERPIPE NTZ and Interpipe NIKO TYUB previously had only the maintenance schedule and PPR, all went along it. Now they make measurements and see the situation in real time and this is already a preventive diagnosis. People really enjoy it. Earlier – it was impossible to prove that there is or is a problem with the equipment if it is not obvious. There were external contractors who came with their expensive equipment, measured something, made conclusions and assessed the state. Now all this is available to the specialists of the enterprise, and they can anticipate negative scenarios.
The second example of a change in culture, and it is very revealing, in my opinion. There is always a period of time between the occurrence of a service request in the system (generated automatically) and a real reaction to this application. The system allows you to clearly see this reaction time, and this statistics is available to management. So, – according to the results of the first such analyzes, many managers had a shock. They saw that the service personnel reacted to applications quite differently, as is supposed by the regulations. For example, there is an urgent application for the repair of critical equipment, but nothing happens for 3 days.
– That is, – services do not respond to signals and system recommendations?
They react, but in their own order and in their understanding of convenience and priority. It happens not only because of negligence. People often have other priorities, and generally there may be significant differences between them. Analysis of these deviations, understanding of priorities, as management sees it, and how – the service of the plant – leads to optimization of work and better service efficiency.
– But how do these priorities fit in practice?
In general, this is solved by organizational and software methods. What is “organizational methods” – these are the rules and procedures, these are the rules of service. Software – this is when we set priorities for different types of equipment in process algorithms. Everyone can have different causes of breakdowns – from cosmetic, and to the safety of people. So priorities are allocated and according to them you need to act.
– Does the system have dashboards for KPI and what level?
One of the main indicators in the world is OEE (the total efficiency of installed equipment). OEE is often measured in manual mode. Somewhere do not measure at all. Only a small part of the enterprises we use automated means. I will explain the importance of this with the example of downtime – one of the 3 main components of the OEE. Usually, the sensors do not say why the equipment costs. They just say it’s worth it. Therefore, the cause is diagnosed by a person. For this we have a separate business process. For example, on INTERPIPE NIKO TYUB there is a process that allows you to identify the reasons for downtime. Often this is not obvious. For example, repair personnel say “you are using equipment improperly”. And the operational one – “yes no, you did it badly for the last time.” Conflicts arise. Therefore, rules and processes are needed that unambiguously interpret these or other situations. On Interpipe NIKO TYUB at the end of the month, the chief engineer sorts out such controversial situations and uses all the information from the system.
– It’s great … Is there an automatic measurement of OEE in Interpipe?
Yes, the company has moved to a single OEE measurement system, with uniform rules for calculating OEE elements, including downtime. As a result, we know that OEE has improved by 10-15%. The premium part of specialists is tied to the results of the OEE.
– You analyze the reliability of equipment using the RCM technique. What is RCM analysis?
RCM analysis is a process (technique) aimed at establishing the root causes of why the equipment is failing. Any simple more regulatory time generates a series of questions. The system automatically identifies to the group of experts on reliability clarifying questions such as “what is the reason for idle time”, “did simple things affect the safety of people, the environment, production processes”, “how could they be prevented or quickly eliminated,” etc.? The questions are constructed as simply and unambiguously as “yes” “no” and the result is a recommendation. For example, – “it is necessary to increase the frequency of checking the oil pressure”. The system enters this into plans, and then the regulations change. Thus, there is a constant improvement of the service system.
– How have your own views on the EAM and the development of such systems in Ukraine changed after this introduction?
This is a modern approach – without this enterprise can no longer manage. It is impossible to collect statistics on equipment without automating the maintenance business processes, ideally with automatic input of information from the process control system and MES.
Of course, we also completely withdraw from the standard PPR and go to the preventive one, and then more and more – predictive diagnostics, where data processing plays an increasingly important role. Another trend is the qualitative visualization of the data. We already use a lot of chips, so that the information is reflected in the best way.
Behind these trends, the future and we are actively working on it, and we are already implementing it.
– Is it now easier to convince the customer with economic arguments such as “here look, how did the cost of maintenance at the plant X decrease”? That is, how much easier is it than say 5 years ago? And can you guarantee the improvement of such indicators?
In systems of this complexity, the result can only be guaranteed if the project activity is properly organized. At the stage of review of the project, we always discuss with the customer what key indicators will influence the results of implementation. Usually we give our and world statistics and specify the conditions under which such improvements can be obtained. Therefore, for example, improving the OEI indicator is important yesterday, today and tomorrow will be important, but the technologies are moving forward, and tomorrow we will present changes to what is advanced today. Earlier we talked about automation of business processes of PPP, today we say that everything in the world goes to predictive diagnostics, but this requires information and processing in the data warehouse.
– Thanks for the interview! I wish you new successes and good projects!