*2.1. Changing of Business Priorities of Factories*

The changing demands of customers and emerging progressive technologies are revolutionising not only the existing manufacturing environment but, at the same time, bringing about a change in the main paradigm of business. Business priorities are dynamically changing. The technical level of production factories and systems for the planning and control of the activities of these factories fundamentally affect the productivity and efficiency of each factory and hence its competitiveness.

In the past, when world markets were not saturated, it was a priority for businesses to achieve high production capacity utilisation. Capacities accounted for capital invested, and businesses tried to assess it. The aim was to produce simple products in high production volumes (mass), which guaranteed the benefits from the economy of the quantity. Businesses were looking for low-wage territories, which triggered a mass shift to countries with cheap labour, known as offshoring. After the market was saturated, the customer's requirements and preferences gradually changed, resulting in increasing product variants and the combined growth of production complexity. Gradually, the priority of high capacity utilisation and low wages has been replaced by the priority of high flexibility and management of complexity in the production.

The flexibility of production is a prerequisite for the production of a wide range of different products. Thus, the priority of the high capacity utilisation was replaced by the requirement of the high flexibility of production and the ability to cope with the complexity of such production. Only this approach guarantees sustained productivity growth. The relationship between flexibility and productivity is shown in Figure 1.

**Figure 1.** Flexibility versus productivity relationship [9].

Since the late 20th century, the latest information and communication technologies have begun to be mass-deployed in manufacturing companies, resulting in cheaper means of automation and industrial robots. The cost of automated production was lower than labour costs, and it brought a new phenomenon, now known as reshoring. Businesses began to move their production capacity from countries with cheap labour to their parent countries [10]. A good example is large corporations from the US (Intel, General Motors, General Electric).

The company is, therefore, transformed from production to services. New ways and forms of value creation are created. This brings significant changes to today's factories. The classic factories and their manufacturing systems are gradually transformed, as they end up with classic products we have perceived for centuries. Customers today do not need to possess physical product models when they look for a service that fills their requirements. The young generation no longer needs things to possess. They want to share them, thus stimulating the emergence of a so-called sharing economy. The importance of leasing has grown. Competition is shifting from products to business models. Today, businesses are already competing with their business models [9]. The price of the product is still the determining criterion by which most customers choose from the offerings. If several manufacturers offer similar products, the selling price is the only criterion in their decision-making.

The objective is the simple integrability of product variants in customer-oriented manufacturing systems and, consequently, the efficient implementation of mass production, so that the production of an affordable product is ensured through the economies of scale. It is known that the reduction of unit costs is due to the increasing volume of products produced. There are usually several global producers in the markets that offer similar products with a similar level of quality at relatively low prices. Small firms can compete in such an environment only by finding a more efficient production method, similar to that used by large producers. If the manufacturer wants to sell more expensive products, it must bring new value to the market, e.g., new products with different characteristics that customers will appreciate and buy. The new customer requirements are thus linked to the growth of product variants. The highest form of satisfaction of customer requirements is the personalisation of demand. This means that each product is tailor-made to the customer. The strategy of mass customisation may be appropriate for the economies of scope, and its effective implementation is not possible without advanced manufacturing systems capable of responding rapidly to changes [11]. These changes affect many of the long-established patterns of behaviour, known as paradigms.

#### *2.2. Paradigm Changes*

We live in a time when paradigm changes are underway, and their main drivers are new, emerging technologies. Advanced technologies affect the life of the whole society. Their most significant impact is reflected in the production sphere. Robotics is one of the areas that have, for several decades, undergone a technological revolution. Industrial robots have already become commonplace in production. The development and deployment of mobile robots in production and logistics are also on a similar path. Over the past five years, research labs have found their way into manufacturing workshops and cooperative robots (cobot). These represent an intervertebral stage in the transformation of manufacturing systems by integrating the activity of man and robot cooperatively.

At this stage, the man remains part of the production processes. Another development step is humanoid robots, which are gradually becoming a priority in research and quickly penetrating manufacturing practice and services. One of the main priorities for research is collective robotics, namely, the control and coordination of the target behaviour of a group of heterogeneous robots. Manufacturing systems and overall production are in the permanent transformation phase.

The current driving forces of the development of future enterprises by [9] can be classified into two independent groups, as shown in Figure 2.

**Figure 2.** Illustration of PULL and PUSH changes [12].

Emerging Technologies (EmTe)—these are technologies that are often from the outset overlooked and, later, can transform entire industries or services sectors. These drivers are also called PUSH-changes because the changes are due to a technological push.

Market changes (changes in customer requirements)—these changes can often be tracked in long-term trends. Customers demand products (services) that better and more precisely satisfy their requirements, which manufacturers implement through the customisation and personalisation of products. As it is the customers who make this move, we call these changes "PULL changes".

In conventional manufacturing systems as a transfer line, lean manufacturing and flexible manufacturing systems may be advantageous to mass production.

In the mass production of standard products, the primary source of the competitive advantage is transfer lines for the production of one product type using fixed jigs and tools. The aim is to produce one product type in large production volumes and the required production quality, which means low production costs.

The concept of lean manufacturing was created at Toyota as an extension of the functions of mass manufacturing. It is also known as the Toyota Production System, and its primary purpose was to reduce the production lead time, linked with quality growth, cost reduction and loss-making and wastage.

Flexible manufacturing systems (FMSs) have been designed as solutions to ensure sufficient production of the family of similar products. FMSs use NC-controlled machines, automatic handling and transport systems and integrated production management. The deployment of information technologies with the possibility of rapid reprogramming has allowed the product range to be easily changed.

Conventional manufacturing systems used to control the production of classic PUSH approaches and, later, PULL-control. This manufacturing system was, at the time of origin, sufficient. However, now, some areas of markets in which conventional manufacturing systems are used have started to require a variety of products, with little time of product placement in the market. Hence, this growing demand of customers has created a complexity of production that has prompted the increasing experimentation of development of reconfigurable production lines or so-called competence islands.

The paradigm change that is currently occurring is characterised by the use of agents and the principles of multi-agent control in manufacturing. In practice, this means that the classic control systems will be progressively replaced by multi-agent control. Multi-agent control brings the manufacturing of emergence, which means that the characteristics of manufacturing systems are also changing as they become emergent.

The traditional manufacturing systems were complicated. The multi-agent control application in manufacturing represents a transformation of complicated systems into complex systems. For complex systems, the complexity of interrelationships between the various elements of the system is already so significant that it often tends to be very demanding, if not impossible, to use mathematical modelling for their studies. The dynamic behaviour of such systems can only be studied using the theory of complexity. Future manufacturing systems will operate as adaptive, dynamic manufacturing networks. New simulation systems must also be adapted to this new requirement. They must possess the ability to simulate agent systems and modelling large networks. Modelling and simulation will, therefore, be an integral part of the planning and control of the processes of factories of the future. In manufacturing, in addition to real objects, there will also be their virtual representatives, which we now refer to as digital twins. Such a dual representation of production is also known as virtual manufacturing. For the visualisation of future manufacturing systems, we can see a similarity to living organisms. Holonic production with multi-agent control will resemble more the emergence of the functioning of living organisms rather than a mechanical automaton.

A new trend in manufacturing systems development is reconfigurable manufacturing systems. Nature teaches us that when changing the environment, the living organism strives to adapt to changed conditions. It uses the change of internal structures and the number of elements and their composition. At the molecular level, it "stretches and recycles" unnecessary structures and reconfigures them into new, necessary structures. Recycling is a process of decomposition. Reconfiguration then represents the new use of existing structures [13].

Most of the activities in the industry of the future will be performed by intelligent robots. In order for the robots to be able to carry out their tasks, often in an unfamiliar environment, they have to possess autonomous capabilities, hence, the ability to adapt to their surroundings and the changing conditions of the surrounding area, collect and evaluate information about their internal state and environment (perception), predict future situations, make the necessary decisions and, of course, learn from the situations. Such tasks can now be tackled by the individual, advanced robotic systems.

The growing interest in mobile robotics applications has not only made changes to the part of users of robotic solutions but also to the part of their suppliers, i.e., manufacturers of mobile robots. Users increasingly prefer more complex mobile robotics solutions, with autonomous intelligent control, localisation and navigation.

The behaviour of future collective robots must resemble the behaviour of living organisms. From that point of view, we have to distinguish the concepts of robotics swarm and collective robotics.

Swarm robotics include a set of relatively simple, homogeneous robots. The behaviour of such robots imitates the behaviour of simple living organisms (we refer to them as swarms or flocks) such as bees, ants, or flying birds. For the collective behaviour of such a swarm, relatively simple rules apply. Each member of the swarm has a specified range of activities, which it carries out in favour of the whole swarm [14].

Collective robotics usually involves many, often very heterogeneous robots. Heterogeneous robots may include a whole set of autonomous robots, not requiring a human operation, from mobile robots, road robots, through to flying (drones) and floating robots. These robots possess strong autonomous functions, intelligence, and mobility capabilities. Such robots work with intelligent sensory networks and computer systems organised into cloud-based solutions. In the complex management of collective robots, it is no longer possible to use classical, centralised management. The results of the research in progress have shown that the management of collective robots will require a "proprietary" operating system [15].

The cooperation of collective robots differs significantly from the cooperation of simple swarm robots. In performing complex tasks, in a challenging and unfamiliar environment, collective robots must use distributed control mechanisms that can combine the behaviour of individual, autonomous robots into the complex behaviour of the entire group of robots. We refer to this behaviour as "holonic". For the control of the holonic systems, it is typical to use agent access and multi-agent systems (MAS). The process of cooperation of the group of individual and autonomous robots creates a higher level of collective intelligence, which we call emergence.

In the human body, we can change all the organs except the brain. Its change (disintegration structures and remastered) is blocked. Likewise, the company. Most structures change when reconfigured, but the central control system remains unchanged. It is possible, like the brain, only to expand its function (augmentation) through external expansion. Its architecture must be designed to reflect future changes. The custom control architecture remains to be maintained when reconfigured. A reconfigurable enterprise tries to behave like a living organism [16]. New manufacturing concepts are developed as a response to this paradigm.

### *2.3. New Manufacturing Concepts Designed for Factories of the Future*

All new manufacturing concepts seek to meet one of the main objectives and, thus, adaptability, the ability to react immediately to rapid changes in the environment, is also referred to as turbulence. Adaptive manufacturing systems are, at present, a peak of scientists' efforts to formulate the contours of the future production environment. In order to meet the requirement of adaptability, it is possible to approach this in several ways, so scientists have developed and tested a whole group of new manufacturing concepts such as:


The manufactured product will behave in new manufacturing concepts as a smart entity, able to communicate with its surroundings and able to organise its processing entirely autonomously. Such a product will itself determine the sequence of its processing, allocate the required capacity in the relevant competence islands and sump a mobile robot to ensure its transport in production. To enable such a system of organisation to work safely and reliably and to fulfil the required tasks, it will require new ways of manufacturing planning and control. Next, the seemingly "chaotic" world of production will no longer operate current push control systems. With a vast number of smart elements (entities) in the manufacturing system, there will be complicated relationships and situations that are no longer able to deal effectively with today's hierarchical management. Complex relationships between individual entities cause a status called emergence, that is, the state in which it will no longer be challenging to predict the future behaviour of such complex systems. Therefore, researchers are experimenting with new management approaches based on the relative autonomy of the individual elements of the manufacturing system and their behaviour, which will resemble the behaviour of intelligent, living organisms. In production, in addition to real objects, there will also be their virtual representatives, which we now refer to as digital twins. Such a dual representation of production is also known as virtual manufacturing.

### 2.3.1. Reconfigurable Manufacturing Systems

The Reconfigurable Manufacturing System (RMS) is a production system, the structure of which is merely adjustable, with the possibility of scaling capacity and flexibility bounded by the selected product family [11]. Figure 3 illustrates the vision of reconfigurable manufacturing systems.

**Figure 3.** Comparison of the static reconfigurable manufacturing system and the dynamic reconfigurable manufacturing system [12].

Reconfigurable manufacturing systems represent the evolutionary phase of the development of manufacturing systems. Their application requires a new approach in which they play a dominant role in reconfigurable machines, jigs, tools, logistics, and reconfigurable control systems [11].

RMS is built to allow for easy and rapid conversion (reconfiguration). This feature pushes reconfigurable manufacturing systems into the adaptive systems area. Reconstructions enable the production system to be adapted to new product types (functionality) and new production quantities (capacity) [17]. Reconfigurability has thus become a new technology that can better meet market fluctuations and turbulence through the gradual rebuilding of the manufacturing system. Reconfigurability represents the operational ability of the manufacturing system to adapt its functions and capacities to a particular product family.

It results in the desired flexibility of the manufacturing system. As opposed to reconfigurability in the manufacturing system, flexibility is firmly defined. Reconfigurability and elasticity make the adaptive ability of the manufacturing system, which is achieved through a change in its structure. Such a structural change makes it possible to adapt the functions and capacity of the manufacturing system to new requirements. The condition for effective reconfigurability is the requirement to minimise the effort undertaken and maximise the reduction in the time required for the implementation of the changes [11].
