Anatomy of a Smart Industry: AIoT and Autonomization

2019-12-24 | 6 min Logistics and Manufacturing

The revolution in industry and logistics did not end with the Internet. New technologies such as digital twins and artificial intelligence are changing production processes and principles, which in turn leads to a transformation to dynamic and autonomous management of production and supply processes. Distributed production, intelligent automation and dynamic scalability are key features of manufacturing enterprises in the coming decade.

Decentralized Enterprise Infrastructure

Decentralized systems are what differentiates the era of the Third Industrial Revolution (brought about by the advent of microprocessors) from the current Fourth Industrial Revolution. Centralized systems with a single central database were limited by the capacity of the central server, which posed an increased risk of outages. Therefore, in complex digital ecosystems, such as manufacturing companies, it is essential to use decentralized solutions.

Just as the physical layout of the enterprise is decentralized, so is the responsibility and need to make operational decisions at individual levels. It is best to make such a decision as close as possible to the place where it is made, because the necessary information, connections and quick feedback are available at this place. Therefore, this hierarchy should be copied by Smart Industry control systems.

The concept of the fourth industrial revolution therefore envisages the creation of an intelligent network of diverse distributed entities along the entire value chain, i.e. across production, economic, commercial, logistical and other sections and departments.

This distributed infrastructure enables communication between individual components (people, machines, objects, data and systems) of the network and from the position of equal partners. On the other hand, components connected to the industrial Internet of Things network generate a large amount of heterogeneous data, which Smart Industry systems must constantly process and transfer in real time.

Data is a commodity whose importance will also grow radically in the industrial environment. Last but not least, the benefit of distributed infrastructure and decentralized Smart Industry systems is the extensibility of the network, or rather the scalability of the platform, as well as its increased resistance to failures of the network itself, individual connected systems or their components.

Benefits of smart enterprise initiatives: 12% increase in labor productivity, 11% increase in enterprise capacity utilization, 10% increase in total production output (source: Deloitte)

Collective Intelligence of Equipment and Processes

Innovative production and logistics management, such as that provided by Smart Industry systems, cannot be done without artificial intelligence (AI) tools and technologies. The combination of the Internet of Things (IoT) and artificial intelligence (AI) is coming to the fore not only in the common consumer environment, but also in industry. Smart Industry systems already use intelligent algorithms to monitor, control, manage, and plan complex processes and operations throughout the production process and in supply chains.

The use of cognitive technologies in industry will gradually increase and contribute to increasing the autonomy of individual components involved in the enterprise Internet of Things (IoT). Thanks to this, each connected element will gradually become intelligent ("smart"). This means that it will work relatively autonomously and communicate with other network members as needed.

The interconnection and interoperability of "smart" components of the decentralized ecosystem of the Industrial Internet of Things (IIoT) are a prerequisite for the emergence of collective intelligence systems modeled after colonies of organisms in nature (so-called swarm intelligence).

The use of the principle of collective intelligence leads to the emergence of solutions based on the self-organization of individual components (devices, materials, data, systems) participating in internal company processes, including production and logistics operations. It is the ability to self-organize that paves the way for the emergence of autonomous industrial control systems, which are not as futuristic a technology as it might seem at first glance.

Artificial Intelligence (AI) Implementation by Function: 29% Maintenance, 27% Quality, 20% Production, 16% Product Development (R&D), 8% Supply Chain Management (Source: Capgemini)

Industry to Order or Dynamic Pull Production

The development of Smart Industry systems is proceeding towards maximizing their modularity and the ability to autonomously reconfigure based on automatic situation recognition. The self-organization of production processes therefore also includes reconfigurability characterized as independent adaptability to internal and external stimuli. Its manifestation is the self-optimization of processes, which represents the next evolutionary stage of intelligence, also with regard to the artificial intelligence tools used for its implementation.

A prerequisite for achieving this capability is the collection and transfer of diverse data across the industrial Internet of Things (IIoT) enterprise ecosystem and its accelerated analysis at all management levels. The implementation of technologies such as artificial neural networks with machine learning (ML) elements then shifts the functionality of auto-optimization from a reactive position to a proactive and predictive position.

In recent times, the demand for greater process flexibility has come to the fore. This is due, on the one hand, to the constant development of external circumstances (economic, technological, or political), and, on the other hand, to the growing demand for increased variability and personalization of products. The currently established model of mass production is becoming a burden and manufacturers must adapt to individual customer wishes, which may exceed the capacities of existing possibilities for varying a given product.

No longer fashion waves or time-limited trends, but custom production is an increasingly common mode in the mass production environment. Current Smart Industry systems allow companies to respond more flexibly to internal or external stimuli and transform business processes into dynamic ones.

An example is the digital transformation of a warehouse of a trading and distribution company, where ANASOFT implemented the function of auto-configuration of the warehousing process. When receiving each new package of goods, the Smart Industry system EMANS (based on an analysis of the turnover of goods) adjusts the location of a specific package in the warehouse so that the most requested goods are as close as possible to the shipment. This minimizes the time subsequently needed to pick and send the order to the customer. As soon as demand patterns change, the system automatically adjusts the algorithm used without complex device settings or training of warehouse workers.

Dynamizing processes through auto-reconfiguration has revolutionary potential, which, for example, will lead to the emergence of fluid (non-linear) production workplaces in production. This production mode will copy the structure of the decentralized model, products will not move along the production line according to a fixed and unchanging schedule, but will autonomously access constantly regrouping workstations according to the current availability of resources and the requirements for personalization of selected elements.

Smart Factory Survey: 86% of manufacturers surveyed believe that smart factory initiatives will be a major catalyst for manufacturing competitiveness in 5 years, 83% of manufacturers surveyed believe that smart factory initiatives will change the way products are made in 5 years, 30% of global budget allocated to smart factory initiatives, 58% of manufacturers surveyed expect their budgets for smart factory initiatives to increase in 2020 (source: Deloitte)

From automation to intelligent autonomy

Interoperability, decentralization, intelligence and reconfigurability are features that currently bring additional benefits to businesses, for example in the areas of safety and environmental issues. The Industrial Internet of Things (IIoT) and the enormous amount of data that this network generates create greater transparency in business processes, thus reducing the risk of human or technological errors.

Industrial Internet of Things (IIoT) technologies and Smart Industry systems also have an impact on increasing the safety, reliability and sustainability of the production environment, and ultimately on preventing negative impacts on product quality.

Greater transparency also contributes to reducing waste, reducing the share of rejects or identifying their origin of poor quality, thereby minimizing the environmental footprint of the manufacturing company. Naturally, the integration of three-dimensional printing technology ("additive manufacturing") will also bring about a further reduction in production waste as well as a reduction in the necessary logistics processes.

The above-mentioned principles of smart production and logistics are also the basic driving force for maximizing the added value of individual processes throughout the company. Industrial automation primarily characterizes the third industrial revolution. The current one, essentially the Internet of Things (IoT) and artificial intelligence (AI) revolution, no longer aims at automation, but at the autonomization of industrial and supply processes.