New Maintenance Management Topics

Abstract

This chapter deals with new topics in maintenance management. The need for maintenance as a result of changes in the substantial environment of organisations increases. Based on current maintenance problems in organisations and social and environmental needs of society, key management trends can be deduced through the system analysis. It follows a large area of quite changing legislation. The field of Artificial Intelligence and the Internet of Things and so on also come into play in maintenance. The chapter is also based on the practice of authors in this field. It therefore affects the area of production equipment, human resources, software support, costs and the material base. Maintenance management risks are also significant. Without clear trends, organisations cannot direct their strategy and thereby effectively manage their own maintenance. This chapter is intended to help organisations strengthen their overall competitiveness through maintenance management.

1. Introduction

The economic maturity and the current situation of a given country determine the state of maintenance management and quality management in organisations. The dynamics of the development of maintenance management can be documented in the example of the Czech Republic. As a result of changes in the essential environment of organisations, the need for complex maintenance (asset management) is increasing. At first sight, the economic field does not present this trend. However, higher maintenance costs are evident. In particular, the increase in taxes in the Czech Republic and two higher value-added tax rates are essential. The price of the highway mark is higher. Changes characterised by increases in social and health insurance for employees and self-employed persons, such as maintenance employees, affect not only the labour costs of maintenance prices. To this must be added higher travel allowances. All this means an increase in the cost of logistics services and labour, including maintenance [1]. In this context, publication [2] calls for the inclusion of crisis management in organisations and the use of a process perspective. Emphasis is to be placed, among other things, on avoiding waste of resources and energy.

The legislative area is also an essential environment, as in the article [3]. Legislative changes in the Czech Republic mainly concern the strengthening of fire safety and the environment towards a circular economy in relation to industrial activities. The assessment of compliance with general requirements for the competence of providers and the technical requirements for certain materials such as paints, sealants, leather and textiles are also essential. The aim is to reduce the production of microplastics and microparticles in general. An open environment that will be yet legislated to a greater extent is modern information technology (including Industry 4.0 and so on).

This chapter provides a basic literature review of current research in the field of maintenance management. It presents a list of methods for maintenance management and lists its problem areas. It proposes a set of characters for the maintenance process, describes the current state of this issue in organisational practice and sets trends for the possible development of maintenance management in particular for the need of organisation’s strategy.

2. Maintenance management in publications

The thematic direction of maintenance management must be conceived from the external context of organisational practice and from the theoretical results of maintenance research. Current publications on the field discuss the following areas.

There are corrective, preventive, and predictive maintenance as it is common in current organisational practice, the publication [4] wrote not only. When predictive maintenance is applied, a 46% saving in maintenance costs is based on this publication. However, it is necessary to increase the proportion of preventive maintenance compared to post-failure maintenance to reflect the cost reduction, as added in publication [5]. The maintenance of a multi-component system under uncertain operating conditions is discussed in the text [6]. Quite often, decisions are made regarding extending the lifetime of components that should already be replaced due to their wear and tear, as added by the text [5]. The publication [7] presents a custom metric oriented towards decision-making and forecasting in relation to predictive maintenance. The useful life of degrading components is determined.

The article [4] describes maintenance planning using the CBM (condition-based maintenance) method, here using aircraft maintenance as an example. It updates the list of maintenance tasks on a daily basis and tries to match them with the available hardware/software interface and resources. The scheduling is two-phase for equipment and for components, as for example in text [5]. At the same time, this paper concludes that a more complex system will fail faster due to the enormous number of components. A CBM methodology based on a combination of reinforcement learning and machine learning and remaining component lifetime is presented in the article [8]. Again, the goal is mainly to minimise the maintenance cost. Commonly, somewhat simplified component deterioration conditions are considered, which makes subsequent maintenance prediction difficult. For example, the impact of maintenance limitations on train availability, caused by various events of external context, is presented in the text [9]. It follows that it is clearly worthwhile to perform at least basic maintenance on objects. Not to omit maintenance altogether because of cost or other complications. The paper [10] investigates algorithms for the problem of facility maintenance frequency and moves to the next facility in similarities with the BGT method. The paper [11] addresses redundancy allocation for a series-parallel system. In doing so, each component can be characterised by two function states and further by its performance.

Maintenance scheduling is now supported using networks that also need to be topologically characterised, due to the reduction of maintenance overhead, which is addressed in the publication [12]. The text [13] describes data management issues for AI (Artificial Intelligence) and IOT (Internet of Things). It is necessary to ensure the reliability of data when the data are large. In addition, skilled personnel with different knowledge and skills than before are needed for maintenance in conjunction with AI and IOT. All of these lead to more accurate maintenance decisions. The study [14] describes the relationship of maintenance in the Industry 4.0 with respect to employee outcomes and develops a competency model of maintenance for the Industry 4.0, which in turn leads to the satisfaction of these employees. This is due to the change in maintenance employees’ activities as well as material and information flows as a result of the introduction of the Industry 4.0 as a technological innovation and as a result of socio-economic changes. This model is linked to a hierarchical structure. The optimization strategy of predictive maintenance is presented in Ref. [15]. Digital twins of each object (i.e., an electronic copy of the object including the associated functions and properties) are created, as in the text [16], and their behaviour over time is monitored based on historical maintenance data. The goal here is to gain predictability of maintenance to avoid incurring unpredictable costs. The combination of spare parts management (management, circulation, performance, etc.) for maintenance and the Industry 4.0 is discussed in the article [17]. The connection between TPM (total productivity maintenance) and the Industry 4.0 is described in the research [18]. It establishes characters (parameters) of system sustainability also here on a hierarchical structure. It sees knowledge of TPM issues, employee involvement and support from the organisation’s leadership as key success factors. The application of AI and IOT is now perceived by domestic society as favourable. However, the author of [19] points out that the power of words from an emotional perspective is not yet included in AI. Facility management in conjunction with IOT is described in the article [20]. It points out the lack of educated staff in this context, as does the text [16]. The software applied has a fundamental influence on the quality of the system implementation. In maintenance, the shortcomings of IOT in the form of not taking out the rubbish, not cleaning the corners and not cleaning the work area are evident. On the other hand, there is an effort to conceptualise modern facility management in a responsible way, with emphasis on social aspects (application of less employable people) and ecological aspects (application of eco-friendly means, recycling, and carbon footprint monitoring) and the corresponding application of diagnostics. A hot new feature is the inclusion of non-financial reporting in maintenance management, which means tracking characters that mainly describe social, environmental and safety areas. The text [16] mentions the problem of expanding the digitalisation of organisations in the persistent shortage of chips and therefore of object controllers. On the other hand, it sees a great benefit in more accurate corrections in production and maintenance and in the automatic adjustment of objects. The paper [21] then presents possibilities of data collection and value monitoring. The speed of data processing and the production of specific alerts about the state of the components in the system is high and certainly pleasing for users in maintenance. The open connectivity, interface options and open-source applications are advantageous. However, it does not recommend an anonymous cloud environment or linkage to a single brand to store and process data.

Deep learning (a type of machine learning algorithm development for artificial neural networks with extensive features) in predictive-maintained critical safety systems is described in the text [22]. The study [23] discusses the maturity levels of a machine learning system (basic device learning from experience without intentional programming by another person) for predictive maintenance. The characters of the system are object reliability, performance and intelligence. However, it encounters imprecise machine learning terminology and poor access to the necessary maintenance data in the organisation. A major problem is the cost of maintenance and the difficulty of integrating the systems overall, both technically and organisationally. Sustainability in this area is also under scrutiny. Kinds of turbine failures and maintenance with a focus on sustainability are presented in the text [24]. The key objects in this type of system are the bearings and their tribology. Hence, the maintenance costs of the system are important to know. It is also necessary to monitor the conditions under which nonconformities of sub-objects occur.

Donor funding and its effects on system maintenance, using water supply as an example, are explored in the text [25]. It identifies a number of problems in this context, such as expensive spare parts, delayed delivery of parts, imposition of foreign technology and disregard for local maintenance staff, which leads to complications in maintenance and its resulting cost to the end customer. The text [26] provides a discussion of moisture management in the building with respect to maintenance, which is becoming a major problem. The reference [27] focuses on avoiding employee exposure to ionising radiation during maintenance of a nuclear facility in this example. It assesses the risks of components. Again, it targets sustainability and system safety.

The issue of maintenance auditing and its limitations is addressed in the article [28]. The internal climate of the organisation, the influence of management power on maintenance productivity, psychosocial factors and so on are not included in audits. Auditors do not observe everything and often do not perceive serious problems and do not specify these findings in reports. The organisation then has no substantial basis for planning and management, not just maintenance.

A summary of survey results from several mentioned publications in relation to the issue of maintenance management is given in Table 1.

Resource	Method	Research	Main result	Key problem
[4]	CBM	Aircraft	Planning,; cost reduction	Quality maintenance data
[5]	Markov processes	Component systems	Planning model for component system	Component life extension decision-making algorithm
[6]	Weibull distribution, stochastic programming	Component systems	Decision model in events	Problem analysis in different scenarios
[12]	Deep Q-Learning, Net topology	Network systems	Reduced overhead for network	Different failure scenarios
[8]	CBM, learning, Kaplan Meier Product Limit	System maintenance	Optimization to the CBM	The function of the component deteriorates unevenly
[9]	KPI	Trains	Simulation of impact of constraints	The model of the failing device
[14]	Publication review, interview	Publications and organisations	Competence model for Industry 4.0	Implementation of model into the organisation
[13]	Publication review, predictive analyses	Industry engineering + AI + IOT	Data management and AI + IOT system	The relationship of spare parts and Industry 4.0
[23]	Publication review	Application of machine learning	Maturity levels for learning	Object failure scenarios and its safety
[22]	Markov processes, machine learning	Industry organisations	Deep learning in systems and safety	Infrequent object failure scenarios and its safety
[28]	Publication review, audits	Audit documentation, organisation	Audits do not consider climate	Specification of risks of object during the audit
[24]	Discussion, Publication review	Wind power plants	Failures, sustainability	Premature object failure and costs
[18]	Publication review, expert interview	Industry organisations	TPM, Industry 4.0, sustainability	Impact of TPM on the sustainability
[27]	HAZOP, LOPA, Aaalysis of scenarios	Nuclear equipment	Risk assessment and safety	Failure scenario and risk analysis

Table 1.

The comparison table of survey results in several mentioned publications in relation to the issue of maintenance management (own of authors by [4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16, 17, 18, 19, 20, 21, 22, 23, 24, 25, 26, 27, 28, 29, 30, 31, 32, 33, 34, 35, 36, 37]).

3. Methods for the maintenance

In relation to publications cited here and to the findings of the authors of this paper, it can be noted that methods used for planning and managing maintenance are varied. The combination of the time to failure of an object and its maintenance costs is frequent and other mathematical and statistical calculations of the reliability of objects or the determination of maintenance KPI (key productivity indicator), which follow more the economic side of the matter, as well as OEE (overall equipment effectiveness) are also applied.

More sophisticated methods include the Monte Carlo method, Markov processes, the Kaplan Meier Product Limit, the stochastic programming to the Deep Q-learning, which may encounter a lack of quality maintenance data. Due to the complexity of the maintenance system and its essential environment, in order to consider multiple factors and features, the multicriteria decision-making and the full consistently method are applied, and from the point of view of reliability and human safety, HAZOP (hazard and operability study) and LOPA (layer and protection analysis) methods are applied.

However, research often encounters imperfections in maintenance models and therefore many research works are based on a literature search and guided interviews with rank-and-file employees or experts in the maintenance field. Somewhat unique, but essential for facility maintenance planning, is the so-called the bamboo method (BGT).

Methods for the maintenance are more categorised in Figure 1. Methods are divided into basic and expanding type, quantitative and qualitative type, inductive and deductive type and according to the difficulty and the area of targeting. For example: Sophisticated maintenance management methods require valuable input data, are computationally demanding, are more expensive to apply and are therefore less used by organisations. Deductive methods help to explain the consequences of failure states.

Figure 1.

Methods for the maintenance (own of authors by [30, 31, 36, 37]). Blue—Quantitative methods; Green—Qualitative methods; Bold—Basic reliability analysis methods; Italics—Inductive methods (bottom-up state analysis); No italics—Predominantly deductive method; MTBF—Mean time between failure; MTTR—Mean time to repair; TEEP—Total equipment effectiveness performance; RBD—Reliability block diagram; FMECA—Failure modes, effects and criticality analysis; FTA—Fault tree analysis; ETA—Event tree analysis.

Note: Two-colour marked methods can be of either type depending on the given data.

4. Problems in the maintenance management

The following problem areas arise from the above.

Essential for the prediction of maintenance is its appropriate planning and optimisation, for example, according to the text [4]. However, this is preceded by the collection and analysis of good-quality maintenance data. The component function deteriorates unevenly, for example, according to the reference [8].

The aim of surveys not only in these publications is usually to determine and reduce maintenance costs, as stated in the text [4]. Prediction methods from models need to be compared and evaluated. It is also necessary to analyse the problems in different maintenance scenarios according to the text [6], and to consider different failure scenarios, as added by the paper [12]. The effect of the inventory model on the maintenance system and its performance metrics needs to be investigated. Furthermore, a decision algorithm needs to be developed for extending the lifetime of components that should have already been replaced in the system, as reported, for example, in the article [5]. However, the safety aspect is not given enough priority in this direction.

Better computational procedures need to be developed for complex maintenance scenarios and a more heuristic approach needs to be incorporated into the scalability problem. More efficient approaches for evaluating maintenance planning that is intertwined with the use of networks are also needed. There is also a need to develop a model of a failing device, for example, a car, as most models are based on an ideal fully functional system, as the text [9] adds. There are also frequent oversights and existing difficulties in implementing maintenance models in organisations, as described, for example, in the text [14]. Even in simpler cases of BGT problems, not all variables (such as approximation ratios for similar objects) are known, which makes maintenance planning inaccurate, for example, according to the text [10]. Efficient data management within AI and IOT and appropriately skilled maintenance staff and a new business model for the organisation should lead to an efficient system, for example, text [13] believes. Furthermore, there is still little knowledge about the relationship between spare parts management and Industry 4.0. The research [18] also recommends investigating the impact of TPM on sustainability in different sectors of the economy. The publication [27] recommends improving the safety and sustainability of equipment through failure scenario analyses and risk analyses.

Research should focus on predictive maintenance with respect to safety-critical problems and low-frequency facility failure scenarios, which are often forgotten so far, as confirmed by publications [22, 23]. Also, the specification of facility risks should be addressed when the auditing of the maintenance process is performed as concluded by the article [28]. The question is also how to involve donation and aid policy in the maintenance, as pointed out in the text [25].

5. Determination of maintenance trends

Overall, it can be concluded that there is not yet a set of trends that should guide future research on the maintenance management. Therefore, this matter is further addressed. Based on the previous research [4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16, 17, 18, 19, 20, 21, 22, 23, 24, 25, 26, 27, 28, 29] and according to the authors’ own knowledge of maintenance management, 6 categories of problems in this area were determined using the Affinity diagram (in the book [36]). These are:

1. Component life extension (mentioned six times)

2. Methods in maintenance (prognostics, scenarios, audits and heuristics) (mentioned four times)

3. Maintenance personnel and their competencies (mentioned three times)

4. Spare parts (mentioned two times)

5. Sustainability (mentioned two times)

6. Artificial Intelligence (mentioned one time)

Categories (e) and (f) are still developing. Category (e) is linked to the environmental management system (according to the standard ISO 14001 [33]). The system sustainability attributes from the publication [18] are qualitative but harder to understand and to compare. Therefore, they need to be redefined. From the authors’ experience, characters such as: system downtime (minutes), data flow rate (Mbit/s), number of conflicts and number of security incidents, can be recommended. Category (f) is still a kind of loose discipline, but it carries several risks and will have to be guarded in the future. The terminology is not professionally finalised according to standards ISO 9001 [32], EN 13306 [34] and ISO/IEC 27000 [35] and so on. The connectivity is not stable, although technically and software-wise the problem is being solved.

Category (a) is currently the most mentioned. It reflects the current situation with shortages of raw materials and spare parts and poor customer-supplier relations. The safety criterion is essential and only then should the cost item be addressed. Uncertainty is associated with category (b). Complex scenarios need to be dealt with in a predictive and auditable manner to obtain more realistic pictures of the system. This places increased demands on maintenance personnel, and especially on their observational skills. So far, the use of simplified component state conditions has dominated in maintenance models as the article [8] states. It refers neither to their gradual deterioration and uneven wear nor to unstable ambient conditions.

Spare parts inventories are a problem constant in maintenance management. Their relationship to Industry 4.0 under category (d) will be key. There are problems with the supply of spare parts. They are expensive, lead times are long, and, in some cases, there is a tendency to apply cheaper substitutes. There is also the implementation of fitting the equipment with several sensors for each key component, which hits the cost growth. The maintenance employees in category (c) must change in terms of competencies and qualifications under the influence of Industry 4.0 and Artificial Intelligence, which are more demanding requirements for maintenance employees compared to, for example, the findings in the book [30]. These are different algorithms and patterns of behaviour and thinking that previous generations of maintenance employees are not used to. In addition, maintenance is facing an ageing workforce.

6. Characters in maintenance management

Commonly used in maintenance management are characters (parameters) such as time to failure, repair time, maintenance cost and cost of spare part. Critical characteristics in quality management, as the basis of an integrated management system, including maintenance, are related to the safety of people, equipment, the environment and data. So far, they have generally been underestimated by society. Safety measures for maintenance employees are essential. This is a very risky job. They often work while the machine is running, under voltage or with media in a critical condition (e.g., hot water, hot oil, pressurised steam, volatile substances, etc.).

Critical characteristics of the equipment, as a product, should also be identified in the technical documentation from the designer or technologist. They often correspond to limits of operation of products (objects) and their action in the surrounding environment (e.g., minimum or maximum temperature). They are based on properties of materials used and manufacturing technologies applied. Predictive maintenance should be aware of these characters and focus on them. From these, it should design other maintenance control activities. Production and maintenance planning and control models often contain sets of coefficients that may not correspond to the real system. Reconfiguring them often causes difficulties.

It can destabilise a functioning system or put it outside the legal limits of its functionality (e.g., exhaust emissions). Models usually address ideal objects and obvious scenarios. They reject uncertainties, for example, in the maintenance system, as unrealistic. Of course, it is the unlikely scenarios that start to appear more frequently in turbulent times, but the system is not prepared for them. The use of end-of-life and maintainability components are related. Ignoring these safety limits at the first moment leads to savings in maintenance costs. However, it can usually lead to a safety incident later, which will make the issue considerably more expensive even if no harm to the people in the vicinity (maintainers or other stakeholders) is caused. There are known cases of serious accidents in the history of maintenance where the original cause was the failure to replace or treat a minor component. At present, this phenomenon must be prevented.

In retrospect, these characters and safety/accident rules should by definition be automatically checked when auditing the maintenance management system. Failure to do so is a failure on the part of auditors, as this area is within their remit. Alternatively, auditors should have an expert on that special equipment with them during the audit.

7. Current status of maintenance issues in organisational practice

Observations of maintenance in organisations and analyses of the maintenance management system and the quality management system have revealed many discrepancies. Current maintenance issues in organisations include:

• Incorrectly selected type of maintenance task (confusion between breakdown and preventive maintenance).

• A fault is assigned to a different piece of equipment than the real one (the error may be caused by the fact that it is not clear at the beginning what caused the malfunction).

• The maintenance record cannot be assigned to the device because the device is found to be missing from the register.

• A device is found to be in the register even after it has been removed or replaced by another device.

• Maintenance schedules are not consistent across different information systems or their modules.

• The fault codebook is not used correctly.

• Inconsistency in the number and type of stock items between the information system and the actual situation.

• Selection procedures do not adequately address the quality and reliability of equipment and spare parts.

• The budget of the maintenance section does not consider management’s requirement for high availability of equipment functions.

The frequency is not relevant to the above nonconformities. Most of the failures are systemic, not random. It is also not possible to determine the severity because the same problem will have different severity at different facilities and in different organisations. Each device is differently important and therefore the severity of the nonconformities associated with them will be different. Based on the 6 M method, it can be concluded that all these types of nonconformities in maintenance management were originally caused by a human factor. Either they are caused by a maintenance person, a warehouse employee, a maintenance application programmer or some manager. It is also noticeable that the maintenance supply of spare parts is often compromised. Building a functioning information system with up-to-date data is also impaired. All of this can then overlap with the integration of maintenance with Industry 4.0. Above and beyond this, the AI acquires data on the shortcomings of the maintenance system, which then does not always guide its reasoning in an appropriate way. Rather, nonconformities require the attention of employees, their systematic and consistent work with the system and the rules set. Sophisticated maintenance methods are somewhat less relevant in this sense. They may be too robust and costly for operational problem-solving in small and medium-sized organisations. Therefore, it pays them to use proven simpler tools and to pay attention to the diligence of the staff. In addition, the findings of the survey in Ref. [29] should be noted. Preventive maintenance is rather on average to poor condition in organisations in the Czech Republic. Even medium-sized organisations have an unnecessarily organisational and administratively complicated maintenance management system. Positives are the linkage of maintenance to strategy; the care of spare parts and digitalization is developed through different types of sensors connected to the equipment and the associated monitoring of various reliability parameters of the object. On the other hand, the support of extensive administration around maintenance and the form of work orders for maintenance in conjunction with its directive management in a hierarchical organisational structure can be seen as negatives. This in turn leads to an increase in maintenance overhead, long information flows and a decrease in the motivation of maintenance staff. The visualisation side of the maintenance software now dominates over the hardware support (interfaces, interconnects and so on). However, the lack of inclusion of maintenance safety, environmental and information safety and social responsibility aspects in maintenance management systems in organisations is fundamental. The road to predictive maintenance will therefore be a long one.

8. Discussion

Prior historical data on the system are needed for any decision-making and prediction. However, in organisations in the Czech Republic, historical data not only from equipment maintenance is often missing, as in the text [23]. This makes subsequent prediction and planning almost impossible. Also, training AI on this basis is quite difficult.

For example, the CBM method is not used so much in the Czech Republic. However, several other methods mentioned here, such as mathematical-statistical calculations of object reliability, determination of maintenance KPI, the reliability centred maintenance (RCM) method, the multicriteria decision-making, or the HAZOP method are used. This is an indisputable advantage for solving maintenance problems of complex systems that often surround people nowadays.

The fact that a complex system will fail faster due to a large number of components is true from the system theory perspective (as in the book [36]). It includes the individual components as elements, but also their links due to interconnections or other interactions such as vibrations. The individual quality levels of the individual objects build on each other, which contributes to a faster destabilisation of the resulting system.

Binding of the maintenance management model in the study [14] to a hierarchical structure is possible. However, it creates disharmony with respect to organisations that apply the process approach in their management system. Thus, they should have some kind of a heterarchical structure. While the incorporation of various methods into maintenance management is a normal continuation of the development of these system areas and the link to Industry 4.0 is to be expected.

However, linking maintenance management with facility sustainability is a new phenomenon. Part of this is the perceived pressure from the European Union to change the ratio from landfill to end-of-life recycling facilities. On the other hand, there is a deterioration in the degradability of some layered materials into base materials, such as metal-plastic. In the context of facility sustainability, it should be specified that it can be understood as careful maintenance to ensure the long uptime of the facility, or as the production and management of the facility regarding both environmentally sound operation and subsequent recycling.

The determining of causes of nonconformities is not so common in conventional maintenance, outside critical infrastructure. However, it is becoming increasingly important for predictive maintenance purposes. It forms a kind of basis for it. It is only on this basis that the appropriate form of maintenance planning and management can be chosen so that the subsequent predictions are consistent with the reality of the system in its essential environment. This then leads to a more realistic model of the facility maintenance system.

The knowledge and skills of the maintenance staff are becoming a combination of maintenance and computer expert due to the involvement of information technology. This consequently changes the competence model of maintenance employees. In addition, different levels of hierarchy have different scope of authority and responsibility. Furthermore, the skills required are not clear, and there is no clear terminology around Industry 4.0 and AI for a part of the maintenance workforce. In addition, these modern technologies are running up against their limits in real-life organisations, as well as their financial capabilities.

9. Establishing basic trends in maintenance development

The original six trends were further analysed considering findings above and characteristics of the maintenance process. This is based on the quality management system according to the text [32]. Factors are maintenance employee, equipment, spare parts, documentation (the maintenance passport), tools, methods, digitalisation, characters, maintenance costs, legislation and essential environment.

Extending the life of components is a necessity in organisations rather than a need. This is due to the stock disparity of spare parts in many organisations in terms of availability and deteriorated customer-supplier relationships. It should also be noted that extending the life of components should be consistent with their maintainability and sustainability, but at the same time must not compromise the functionality or safety of the equipment and people around. The limit is exceeding the critical lifetime, which leads to the destruction of the system. However, the maintenance cannot be done without spare parts.

The area of maintenance personnel and their competencies should focus on consistency of records, parts drawdown and their timely de-stocking in written documentation and computerised information systems as the production planning and control system (PPC) is. The AI draws from the internet and the company’s own information system database. So-called the best maintenance practices are not adequately covered even within the Internet. It will allude to the emotional component of the actions of maintenance personnel. It should help to monitor the consistency of maintenance tasks carried out within the computerised information system or Industry 4.0 and not allow discrepancies to arise. It should also monitor the maintenance person’s state of mind and his or her immediate competence (i.e. perception, work speed, error rate, risk-taking and so on).

Therefore, the essential maintenance trends for organisations are modified as:

• Competence of maintenance employee

• Spare parts

• Digitalisation – IOT, PPC, AI and so on.

• Environmental system – but society has not fully come to accept it.

• Methods – often simple rules, procedures and boundaries are sufficient, as well as an obvious definition of maintenance features. Common methods of maintenance management and strategic planning and control are described in detail in books [30, 31], so they are not listed here.

10. Conclusion

As a result of extensive changes in the physical environment, there is an increasing need for equipment maintenance as a form of equipment renewal. At the same time, there are risks that can jeopardise it. Most of these are personnel and systemic. The aim is to extend the life of a whole facility and all its components, usually at low cost. This puts pressure on the planning and management of maintenance and the availability of spare parts. Organisations need to steer their maintenance strategy in the appropriate direction to strengthen their target competitiveness.

Therefore, based on the analyses, key maintenance trends for organisations have been identified. The most key area is the competence of maintenance personnel. This is followed by spare parts and their appropriate management. Digitalisation affects the internal and external context of the organisation. It is the most dynamic factor. It provides several benefits, such as the collection and processing of previously unsuspected amounts of production and maintenance data. However, possible risks lie in its current lack of exploration and imperfection. The environmental system is the answer to the requirement for sustainability in maintenance, as it will cover both degrading equipment and spare parts in terms of their recycling and packaging and the environmental impact of operating substances. The final area of trends is methods suitable for maintenance. Here, it will depend on the requirements and capabilities of the organisation, what they are inclined to do and what they have financial resources to do.

Changes in the essential environment of organisations and their maintenance are very dynamic and therefore trend priorities could shift. However, this list has also been established with a view of its longer-term stability. Furthermore, the planning and management and maintenance methods could not be described in detail within the defined scope.

Acknowledgments

This work was supported with institutional support for long-term strategic development of the Ministry of Education, Youth and Sports of the Czech Republic.