Preventive maintenance of equipment. The Advanced Preventive Maintenance (APM) project implemented at the OJSC Philip Morris Kuban factory. Reliability service

Each of the following five equipment maintenance strategies has its own distinctive features and optimal applications.

1. Preventive maintenance (PM)... With this strategy, equipment maintenance activities are carried out even before the breakdown occurs, so there is no equipment downtime and the number of manufactured items does not drop. It is almost always cheaper to perform scheduled maintenance than it is to wait for a breakdown.
2. Reactive maintenance (RM)... In this case, the maintenance of the equipment is a reaction to its failure. Oddly enough, sometimes this strategy can turn out to be the best, which will be discussed later.
3. Predictive maintenance (PdM)... Service is based on specific hardware information, which is a reliable precursor to imminent failure. Vibration or thermal analysis can be cited as examples.
4. Condiction based manitenance (CBM)... This term is sometimes used synonymously with scheduled maintenance. What is the difference? With scheduled maintenance, the frequency of maintenance procedures is set in advance, in a scheduled mode. In the case of state-of-the-art maintenance, the work is performed depending on what information about the equipment is provided by the data collection systems in real time - from sensors and other sensors that measure certain parameters. The data collection system correlates it with data specific to the emergency condition so that maintenance can be performed before the equipment fails.
5. Reliability Centered Maintenance (RCM)... Covers a wide range of activities and often includes other strategies. Reliability service refers to the process of determining the minimum safe service level. RCM starts by answering seven hardware questions:
   1. What features and associated performance standards exist in your existing production?
   2. How does a situation arise when the equipment ceases to perform its functions?
   3. What are the reasons for each type of functional failure?
   4. What happens when each type of failure occurs?
   5. What is the significance of each type of failure?
   6. What can be done to prevent each type of failure?
   7. What should you do if you cannot find the right preventive maintenance procedure?

RCM is an integrated engineering approach, the goal of which is to carry out all the work necessary to ensure the highest level of equipment reliability, while minimizing maintenance costs.

How to choose the right strategy

Preventive (planned) maintenance

In order to keep your equipment running, you need to design detailed plans works that will stipulate all the actions for the maintenance of equipment and the time intervals for each such action. You will also need a system for collecting information about the condition of the equipment, which will accumulate during inspections. This is usually done by issuing scheduled maintenance tasks, and data is collected for further analysis by a computerized maintenance management system (MMS).

Emergency service

Is the correct strategy only if there is no easy and inexpensive way to repair the equipment. If it is cheaper to replace a part than to repair it, then you need to use an emergency maintenance strategy and maintain a constant stock of replacement parts. An example of such a part is an electric motor with less than 5 hp.

Predictive maintenance

To carry out such a service, it is necessary to spend money either on the purchase of analytical equipment and user training, or on the involvement of contractors to perform the analysis. Right choice will depend on your specific situation.

Equipment Condition Maintenance

Overall, this is a good strategy (depending on your hardware, of course) and is also a good way to save money on your investment. CBM typically requires investment in remote data collection and user training.

Reliability service

Overall an excellent service strategy and is considered by many experts to be the most cost effective. As stated above, RCM can cover all other strategies depending on the hardware components and the resources available.

Photo taken here: http://xcompstore.pl/uslugi/.

Who might be interested in this

March 2013

Bill Lydon

Maintenance strategy is becoming an increasingly important topic with the increasing numbers of aging automated systems management production processes(APCS) in developed countries, and a lack of qualified personnel in other parts of the world. The goal of the maintenance strategy is to maximize the availability of process control systems in production, without compromising safety and unnecessary costs. Accessibility in this context is defined as a state in which production system can be used and functions correctly. When availability is less than 100%, revenue is lost. What is the best way to achieve the highest availability?

Service focused on reliability

The concept of Reliability Centered Maintenance (RCM) provides some useful insights and stimulates thought. SPE implies the implementation or improvement of a maintenance program by using a systematic structured approach based on an assessment of the consequences of failures, the functional importance of the system components, and their history of failures / maintenance. The concept goes back to the early 60s. last century, when wide-body aircraft first appeared on commercial airlines. The main concern at the time was the expectation that the then preventive, scheduled maintenance programs would have a negative impact on the cost-effectiveness of the larger, more complex aircraft... But the experience of airlines with the UN showed that service costs remained generally constant, while the availability and reliability of aircraft improved. The UN is now standard practice in most airlines in the world.

SAE Technical Standard JA1011 (www.SAE.org), Evaluation Criteria for RCM Processes, starts with the following 7 questions:

1. What is the role of this component and according to what standards should it function?

2. What are possible cases refusal?

3. What events can lead to such a failure?

4. What are the consequences of such refusals?

5. How important is the failure event?

6. What systematic actions can be taken proactively to prevent or reduce the consequences of failure to an acceptable level?

7. What if a suitable preventive action cannot be identified?

Severity levels are assigned to the consequences of failures. Some functions are not critical, while others must be preserved at all costs. Maintenance tasks target the root causes of failure. The ONN process is directly focused on failures that can be prevented by proper maintenance. Potential failures caused by unlikely events such as natural disasters, etc. usually do not get preventive action if the risk is not too high, or at least tolerable. If the risk of failure is very high, UN practice recommends making changes that will reduce the risk to a tolerable level.

The goal of a maintenance program is to focus limited economic resources on those components or systems that will have the most severe consequences if they fail. The UN pays special attention to the use of predictive maintenance techniques alongside traditional preventive measures.

Highly reliable devices

Mean time between failures (MTBF) and mean time to repair (MTTR) of devices are very important in terms of assessing device availability. Of course, not all industries can afford the cost of backing up critical systems. However, using products with higher MTBF and lower MTTR can clearly improve overall system availability. It makes sense to create a list of unreliable devices that have failed over a period of time, and this list will form the basis for the process of determining which devices need to be replaced with more reliable ones. In general, taking MTBF and MTTR into account when making purchasing decisions is much more useful than focusing on cost alone. Ask these questions to manufacturers or integrators.

Predictive maintenance

Predictive maintenance is based on the actual health and performance of the component. Service is not carried out on a rigid schedule, but in case of changes in the characteristics of the device. An example of a predictive approach is the use of corrosion or vibration sensors. In addition, analytical software products are now becoming widespread, which make it possible to predict failures based on information coming from automation systems in real time.

Improvement existing systems

The reliability and maintainability of existing systems can be improved to improve availability under limited time and budget conditions. This usually means involving technical support staff in the development of new systems or modernization of existing ones. A simple walk through the facility with your maintenance colleagues will allow you to learn more about existing or potential problems in the process control system and, therefore, take the right action to improve the situation.

Improving preventive maintenance

Of course, you can let go of preventive maintenance, especially when budgets are tight, but this is unwise. Monitoring preventive maintenance costs provides valuable information for making equipment replacement decisions and can also be of great help in justifying upgrade costs.

Remote consultant services

Many vendors are starting to offer remote services, in which production personnel can use the help of experts in difficult situations thanks to modern IT facilities. This could be a regular web conference hosted by a tablet computer to discuss issues with a remote expert. Of course, your staff should have basic information on the design of automation systems, however, it is difficult to know everything, and an expert physically located on the other side of the world can sometimes be very helpful.

Outsourcing

Most vendors of industrial automation systems have started offering outsourcing services to their customers in the last two years, at least in the largest markets. This service can be another important component of a service strategy. However, you need to realize that the customer is still responsible for the production efficiency. Of course, there can be benefits to outsourcing maintenance, but these need to be carefully considered. The main factor to consider is the relationship between the likely downtime and the savings from outsourcing. Manufacturing personnel will still respond more quickly with appropriate training. Deciding which systems should be supported by local staff and which can be outsourced is important in risk analysis.

Pilot projects on predictive maintenance are launched in the industrial sector of Russia... The first models are already operating at metallurgical plants and power units of generating companies; by the end of 2017, the market volume is estimated at USD 100 million.

With PdM, industrialists intend to reduce equipment downtime and increase the efficiency of their facilities. Predictive technologies are one of the elements of the digitalization of production. Which companies are developing predictive technologies? What barriers prevent the market from developing rapidly?

Optimizing errors

A new digital service called predictive maintenance (PdM) is gaining popularity in the industry. The service is being implemented to collect and analyze information about the condition of the equipment, to predict the timing renovation works and preventing production disruptions.

Unlike traditional preventive maintenance, forecasting is based on data sets, digital models, and not on average statistics. Among the main components of PdM are collection and processing of current information, early detection of faults and errors, and resource optimization.

According to Markets and Markets, the global predictive service market will reach $ 1.9 billion in 2020 (versus $ 582 million in 2015). Key players in the world - General Electric, Siemens, ABB, Emerson, IBM, SAS, Schneider Electric and others.

Industry forecast

In Russia, predictive analytics is still at the starting positions: by expert assessments, the market at the end of 2017 amounted to 100 million US dollars, in a year the figure may grow to 200 million. Recently, the industrial sector is increasingly using PdM.

According to Konstantin Gorbach, Director of Intelligent Applications at Tsifra (part of Renova Group), predictive maintenance tasks are relevant for customers using complex and expensive equipment. These are industries where the failure of an element entails significant losses and threatens safety: fuel and energy complex, metallurgy, petrochemistry, transport. Early detection of problems can help prevent accidents and reduce costs.

« The remote monitoring and forecasting system increases the transparency of the operation of equipment for management", - specified Technical Director systems for predicting the state of equipment "PRANA" JSC "ROTEK" Maxim Lipatov.

The cost of implementing PdM at enterprises is formed in different ways: the cost of pilot projects can amount to several million rubles, industrial implementation - from several hundred million.

« Over the past two years, interest in predictive analytics has grown in Russia. The main players in our market are ROSTEK with the PRANA system, Datadvance, Clover Group and integrator companies. In our practice, there were pilot projects for 10 million rubles, and industrial implementations for 1 million US dollars.", - said the general director of Datadvance Sergey Morozov.

Protection for gigawatt

In 2017, energy companies joined PdM. For example, the T Plus holding signed a contract with ROTEC to implement the PRANA forecasting system at 16 power units. ROTEC connects turbines, waste heat boilers and booster compressors of the generating company to its own situational center. Due to such a solution, the energy management plans to reduce repair costs and equipment downtime, to find design defects in equipment in advance.

« Ten power plants of the company will be protected from technological risks. This is the most important step towards large-scale digitalization of the energy sector: the total capacity of equipment connected to the PRANA system will exceed 3 GW", - explained the chairman of the board of directors of" ROTEK "Mikhail Lifshits.

The technology is also starting to be used by metallurgical groups. The gold mining company Nordgold uses PdM to establish a repair system. However, some practices require investments and expenses for their maintenance in the future, says Alexander Brezhnev, manager for the organization of maintenance of mining equipment at the company.

The service began to be used at the enterprise of PJSC "Severstal" - Cherepovets metallurgical plant... PdM was launched in production to reduce downtime at the 2000 hot rolling mill.

« The predictive model reveals the likelihood of overheating of the bearing of gear stands - one of the most frequent and resource-intensive reasons for a unit shutdown. This is the first predictive repair model implemented at CherMK as part of the company's digital strategy.", - noted in the press service of Severstal.

The company's specialists have developed a digital model to receive data from temperature sensors and form a forecast. In case of deviation of the indicators obtained from the mill, the operator receives a corresponding notification from the norm. This prevents an unplanned shutdown of the mill.

« We expect the predictive model to reduce downtime by 80%. We plan to introduce similar models for other types of failures at Mill-2000, as well as at other units.", - comments Sergey Dobrodey, repair director of the Severstal Russian Steel division.

Economical effect

According to experts, the PdM service is now only at the stage of formation in the regions of the Russian Federation, customers have not yet realized the potential of the direction.

« The economic effect from the introduction of predictive services can amount to hundreds of millions of rubles, if we take into account Gazprom and Russian Railways", - said Alexey Shovkun, director of consulting at Datalytica.

At the moment, the developers note, several barriers prevent technology from spreading widely. So, most of the equipment in production is not equipped with sensors for transmitting information; enterprises do not have data collection and online monitoring systems. In addition, factories often keep inaccurate logs about defects and repairs. The staff's unpreparedness for IT solutions and mistrust of the new service concept hinder the implementation of PdM systems at industrial enterprises in Russia.

Manager of the Industrial Training Department and lean manufacturing, OJSC "Philip Morris Kuban"

Improving the efficiency of using equipment that is already performing almost flawlessly is incredibly difficult. As in sports, where tenths of a second separate a world champion from just a master of sports, in production a few percent, which brings efficiency closer to the cherished 100%, is achieved through painstaking work. What separates flawlessness from just good results? Often, subtle but qualitative improvements make the difference.

At the time of implementation project "Progressive Maintenance - APM ( Advanced preventive Maintenance) " in 2012, our factory has had a maintenance system (MOT) for more than 10 years, based on the principles of World Class Maintenance (hereinafter WCM). In particular, strategies and maintenance plans were developed depending on the type of equipment, minimum stocks of necessary spare parts were created and regularly reviewed, scheduled maintenance work was planned and monitored, RCFA (Root Cause Failure Analysis) of major breakdowns was carried out, agreed and regularly the main indicators of the efficiency and cost of maintenance were monitored. In addition, at that stage, elements of the lean manufacturing concept were already successfully applied in our organization, for example, 5S, A3, SMED, etc. Thus, at the start of the project, the necessary base was created and all the prerequisites existed for further improvement of the equipment operation processes.

It's no secret that preventive maintenance and risk identification of mechanical damage industrial equipment allow you to guarantee the safety and reliability of the operation of the machines. It also saves fabric time and costly downtime. In this situation, the repair of mechanical components of the equipment becomes an extreme measure. In turn, engineers, who are responsible for the perfect functioning of the machines, as well as the maintenance, always strive to avoid downtime and prevent possible malfunctions.

In principle, traditional equipment maintenance strategies are similar. It is possible to adhere to the described strategy in the case of a damage-oriented (reactive) approach to maintenance, when the system components work until they are completely worn out and are replaced with new ones in the event of a malfunction. The only way to ensure the operability of the machine in such a scenario is to have a large warehouse of spare parts, highly qualified personnel and the ability to promptly replace the equipment with a new one. The most popular maintenance strategy is to carry out preventive maintenance at scheduled or fixed intervals. This approach has the distinct advantage of making downtime more predictable. However, it can lead to significant financial risks.

Minimizing downtime as much as possible and drafting strategic plan services that cannot be avoided - these are the goals of every responsible engineer, who looks not only at the purchase price, but also at the total cost of costs throughout life cycle cars. The only way to significantly reduce your owning costs is to maximize machine uptime while reducing maintenance costs. This, in turn, requires rethinking traditional service strategies.

However, there is a perception that the practice of preventive maintenance can really only lead to problems with the equipment, creating conditions for future breakdowns and downtime. Knowing all this, our engineers have created their own approach to maintenance.

First of all, it is necessary to tell about the reasons that made us think about the need to improve the process of equipment maintenance. Works performed during scheduled preventive maintenance (PM) (Fig. 1) are divided into work on request, performed in PM (PM2), and regular planned work performed in PPR (PM3). All this partly resembles the process of servicing a car. Since work on requests, performed in the PM (PM2), is critical for the equipment operability, during the PM, more attention was paid to them. Regular scheduled work carried out in the PPR (PM3) was carried out partially, malfunctions were not detected in time, the necessary spare parts were not ordered and, as a result, we received emergency downtime and the need for additional adjustments. As a solution to the problem, it was proposed to divide the activities for carrying out the PM. We deliberately decided to increase the total PM time in order to increase the efficiency of equipment use by reducing emergency downtime, reducing stops during technical adjustments and, as a result, more stable work machines.

Figure # 1:

The concept of progressive maintenance focuses on reliability and implies the implementation or improvement of a maintenance program by using a systematic, structured approach based on assessing the consequences of equipment downtime, functional features system components, as well as their service history. Of course, you can refuse preventive maintenance, especially when it seems that there is no temporary and material resource... However, practice shows that this is not constructive. It is preventive maintenance and early diagnosis that allow you to stay within budget and even save your company's financial resources. Monitoring the costs of preventive maintenance is a kind of valuable information for making timely and correct decisions about replacing equipment, as well as maintaining statistical records. It can also serve as a good justification for upgrading costs. It is no secret that when carrying out preventive maintenance, the need for overhaul of equipment is also determined.

Progressive maintenance consists of two strategically important areas:

• increasing production efficiency indicators;
• improving the quality characteristics of the product.

The first area consists of several key stages. We started by revising the frequency of the SPR, prepared a checklist (a sheet with a list of checked and monitored parameters on the equipment) with a list of control of basic parameters (adjustments, measurement of backlash, check of cycle diagrams using an indicator - control of the cycle moments).

Next, we analyzed all the work carried out as part of regular scheduled inspections performed in the PM3 (PM3), corrected and almost completely changed the checklists. The next stage is to change the composition of the PPR team. It originally included an operator, an expert technician, three mechanics and an industrial electronics specialist. Now the list of participants has been expanded. Another mechanic joined the team to perform urgent work that may arise at the time of inspection for each type of equipment (cigarette and packaging).

We have also introduced the practice of using replaceable assemblies to reduce the time for servicing or repairing equipment. For example, maintenance of an elevator belt takes 15 hours, and a nodal replacement takes 5 hours. At the moment, more than 130 replacement units are stored in the warehouse, which can be used at any time with minimal time for troubleshooting. Of course, we pay attention to what kind of replaceable units should be stored and in what quantity. Monitoring is carried out on an ongoing basis, thus ensuring effective approach to storage in the required quantity of spare parts without unnecessary freezing financial resources in warehouse stocks.

Figure # 2:

The result of applying this strategy can be seen on the graph (Fig. No. 2). As an example, two periods were selected - the beginning of the implementation of the described approach and the current moment. The graph illustrates a notable increase in equipment utilization efficiency, from 79.06% to 82.88%. This result was achieved due to a more stable operation of machines, a decrease in adjustments and emergency stops.

What has been done to improve the quality of the products? One of the parameters characterizing the quality of cigarettes is the indicator of the stability (quality) of ventilation. Cigarette ventilation is the percentage of air that enters through the filter to the air that passes through the burned holes in the filter. The laser carries out the process of burning holes that serve to provide the necessary ventilation of cigarettes (Fig. No. 3).

Figure # 3:

Getting from the resonator into the divider, the laser beam is focused and burns holes on the scrollable cigarette (to ensure uniform distribution of holes around the circumference of the cigarette), located on the axis between the two drums (Fig. No. 4).

Figure # 4:

After analyzing the results of this quality parameter, which was at the limit of the permissible zone, the the following actions to stabilize the ventilation quality indicator (Fig. No. 5):

• caliber made;
• prepared five replaceable assemblies involved in piercing holes;
• two replaceable units were modernized;
• the procedure for starting up equipment after annual laser repairs has been changed.

Figure # 5:

Applying in practice all of the above, we have achieved an improvement in the ventilation rate, the result of which can be seen in the diagram (Fig. 6).

Figure 6:

The next steps in the development of APM are aimed at applying a progressive maintenance approach to the entire fleet. We plan to project the accumulated experience on other key indicators efficiency, develop a training methodology for all groups technical staff and to standardize approaches for the further transfer of skills to future generations.

PdM (predictive maintenance) is a type of equipment maintenance based on diagnostics and monitoring of its condition.

The modern approach to equipment maintenance is based on a methodology, according to which the purpose of maintenance is not to maintain each piece of equipment in perfect condition (which requires unreasonably high costs), but to ensure the reliability of production and technological processes... In this case, for non-critical equipment, traditional species service:

• Reactive - MTBF without maintenance; used when equipment can be easily replaced or repaired without affecting the operation of the enterprise
• Preventive - similar to the system of scheduled preventive maintenance (PPR); used for equipment with low downtime costs and fast repairs

Predictive (predictive) maintenance It is used in cases where the degree of use of equipment in the production chain is assessed as high, and its failure or prolonged downtime leads to significant financial losses.

Predictive maintenance, as opposed to preventive maintenance, allows you to make repairs not according to a predetermined plan, but when the need arises. Thanks to this, on the one hand, it is possible not to waste money and time on scheduled maintenance of equipment, which can work normally for several more months without repair, on the other hand, the likelihood of unplanned downtime caused by an unexpected breakdown is reduced.

This is achieved through:

• collection of data on the technical condition of equipment and their preliminary processing
• early fault detection
• predicting the time of failure
• service planning
• optimization of resources allocated for equipment maintenance

With the development of means (IIoT), in particular, thanks to the equipment of equipment with various sensors, the collection of data on its technical condition can be performed not periodically, but continuously, without interrupting the operation of the equipment. Timely detection of even small deviations in operating parameters will allow you to quickly take measures to ensure the normal operation of the equipment. Technologies (Big Data) will make it possible to predict the time of failure with high accuracy.

The main advantages of the PdM system:

• service planning efficiency
• preventing unexpected failures

With information on what equipment needs maintenance, the work can be scheduled for the period when it is most cost-effective. Thus, unplanned long downtime is converted into shorter planned downtime and equipment availability is increased.

Other potential benefits of PdM:

• increased equipment service life
• improving production safety
• reducing the number of accidents with negative impact on the environment
• formation of an optimal set of spare parts and materials

In an IT-Enterprise system, predictive maintenance is provided by the product "". It ensures a high level of reliability for every piece of equipment.