Time to take the next steps on the Industry 4.0 journey

One noticeable effect of the move towards digitisation is the impact it is having on the on sensors and devices employed across the plant. Obviously, device connectivity is improving – from 4-20mA current loops, digital I/O lines and proprietary serial protocols to standardised solutions like industrial Ethernet or the IO-Link protocol. In addition, today’s sensors are getting smarter as they need to be able to offer a lot more, in addition to measurement values. They are now also required to provide information relating to their identity and the condition of the application they are monitoring.

Data coming from these smart devices can now be deployed to control units and IT systems in real-time. Data is no longer locked in the sensor and can be used for condition monitoring or predictive maintenance tasks which can help reduce downtimes or improve OEE.

“This is all good news,” said Benedikt Rauscher, head of global IoT projects at Pepperl+Fuchs. “However, there are still challenges that face the control engineer if there is no common language that all components understand. A great deal of manual engineering work still required to integrate devices from different manufacturers into machines and plants. Because through-engineering across the entire value chain is one of the characteristics of Industry 4.0, a unified vocabulary and semantic has to be established.”

According to Rauscher the solution is to add an asset administration shell to the physical product to build an Industry 4.0 component. This shell will contain all data and information related to the asset. Its general structure is already defined and several initiatives are now working on sub models for specific domains as the next important step towards Industry 4.0. “An important next task for automation vendors is to provide asset administration shells for their products and to establish structures for linking the physical products to their shells,” continued Rauscher. “The components have to be marked with unique machine-readable identification labels or tags. During operation the identification data is also available through the component’s interface, so the asset administration shell is available always and everywhere. A good example of how the asset administration shell in the internet creates value is when it is accessed with mobile devices during maintenance and service tasks. Those can very often be carried out much more effective when device specific information is directly available for the worker.”

Eckard Eberle, chief executive officer of Siemens Process Automation unit, believes that the growing quantity of data, brought about by the digital transformation of industry, is opening up new productivity potential and new business models. He said: “The main drivers of this trend are future technologies such as AI; edge computing; autonomous handling systems; industrial communication with 5G; and IT security. A comprehensive communication infrastructure is a precondition for wide-area Industry 4.0 applications and to achieve this, Industry 4.0 needs not only greater bandwidth but also very short transmission times with maximum availability.”

Eberle went on to explain that Siemens is conducting its own industrial 5G research projects and is building a 5G Interoperability Test Center under real conditions. This includes evaluation and testing of the industry standards currently available, such as OPC UA, Profinet, and TSN. 

“Also, from an automation side, as manufacturers, Siemens has completely re-thought the actual technology of process control systems. The result is our fully web-based process control system Simatic PCS neo which provides direct and secure access to all the information required, regardless of location or time, and by using all standard devices, including mobile ones.”

The essence of Industry 4.0 is the digitalisation of assets and processes, which can be incorporated into the concept of digital twinning. Digital twins can be defined as a set of models that provide useful information to develop, validate and optimise an asset or a process in a digital environment. Digital twins are based on data collected by sensors installed on physical objects (things) to provide real-time insights about status, functioning, position and other parameters. 

“There are many advantages that can be brought about by digital twins,” said Giuseppe Surace, chief product & marketing officer at Eurotech. “Among them, the availability of a digital model of an asset or a product will significantly reduce prototyping and development phases, increasing product quality and speeding up interactions and feedbacks by final customers at the same time.” 

This concept becomes even more relevant because the digital data coming from a production plant allows the monitoring of every process – from raw material treatment to assembling to finalisation – monitoring can also be extended to logistics and power consumption which is extremely useful to optimise production and logistic processes to maximise return-on-investment (ROI) and efficiency. 

“Thanks to cloud computing all the data collected by sensors in the field can now be transferred to IT systems for analytics and integration with enterprise resource planning (ERP) and other applications. Eurotech offers an edge-to-cloud platform – Everyware IoT – designed to simplify and accelerate the development of end-to-end IoT applications from data collection in the field to data analytics in the cloud or on premises.

 

“Assets and products are therefore connected, thanks to IoT technologies, which enables new and innovative business models that could help enterprises to optimise their processes and maximise efficiencies while improving daily-life activities.  

“The continuous collection of data generates a ‘data lake’ that contains a great deal of information. Todays applications only use very small percentage of those data, for example to provide predictive maintenance or other Industry 4.0-related applications. The interesting part is that we do not know yet how to use the rest of the data,” continued Surace.

“The next step would be for data to be used by machines to perform tasks autonomously: we are entering into the era of Artificial Intelligence (AI), Machine learning (ML) and Deep Learning (DL). These three concepts are interrelated, but we could define ML and DL as branch applications of AI. 

AI can be defined as the ability of machines to perform autonomous tasks without human intervention: with Machine Learning, operators provide machines with data and algorithms – usually trained in the data center or in the cloud – to perform those tasks. Deep Learning is an even more advanced phase: machines are trained directly by computers and do not need human intervention to work autonomously. This requires a lot of computing and processing power and companies like Eurotech are able to help by delivering products that move the paradigm of data center processing capabilities from the Cloud to the Edge. 

Another challenge facing process engineers today, as they implement IoT solutions and Industry 4.0 applications, is security. This is true not only when introducing cloud platforms but also for the integration of automation networks and IT systems on premises. 

Another prerequisite for this type of implementation is the need for interoperability between the various disparate devices and applications. OPC UA (Unified Architecture) is an interoperability standard addressing these challenges. It is increasingly accepted and adopted globally, including the major players in the automation as well as the IT industry. OPC UA leverages security standards, which have been proven in the IT world since years and defines ways to integrate these standards flexibly in IoT solutions.

Recently NAMUR – the standardisation association for measurement and control in chemical industries – has approved the use of OPC UA technology for data exchange and has defined the NAMUR Open Architecture (NOA) model addressing the need for standardised Industry 4.0 application in the process industry. “The outcome of these activities includes a specific OPC UA Companion Standard as well as the introduction of a so-called ‘NOA diode’ as a system architecture component, for security reasons enforcing strict read-only access to data from field devices and PLCs,” said a spokesperson at Softing Industrial Automation.

 

The spokesperson went on to explain that, in Germany, the interest group for sets of engineering rules (Interessengemeinschaft Regelwerke Technik) represents approximately 30 companies from the chemical and pharmaceutical industries and its service providers. One of its tasks is to promote digitisation in the process industry. As part of this task, the group initiated the development of an NOA model test facility providing complex processes with all types of process and diagnostic data from the various operating and failure states under real conditions for testing purposes.

“Bilfinger Maintenance was contacted to build this test facility, which consists of two containers and their automation, representing the procedural aspect of the facility together with measuring instruments transmitting machine diagnosis data to the automation system. It offers a dedicated internet line including various firewalls and security protections and can be connected to one or more clouds,” continued the Softing spokesperson. “As a result, the test facility can be adapted flexibly to different requirements allowing to be used easily and without a complicated approval procedure by participating companies. This provides a way to produce realistic fault conditions in a reproducible manner to address undesired operating states in the process and failures in the device technology.”

Softing’s dataFEED Secure Integration Server was included in the test facility to guarantee the NOA diode behaviour and to provide the full set of security aspects based on OPC UA in one central component. For example, it allows different access rights to be specified or different applications and users and to restrict access to individual data to specific applications. Thus, the access to the data in the field devices and PLCs can be defined as read-only.

Digital technologies for quality management (Quality 4.0) can also create substantial added value. However implementation is anything but easy according to consulting firm BCG and the German Society for Quality (DGQ) , who surveyed 221 companies from manufacturing industries worldwide on their quality management strategies. 

Nearly two-thirds of the survey participants believe that the application of Industry 4.0 technologies to enhance quality will significantly affect their operations in the coming years. The greatest advantage cited was the reduction in defect rates, and consequently in costs. However, there are still many challenges to overcome including a shortage of necessary skills, and in many companies a lack of willingness to implement the technologies and a reluctance to formulate clear quality goals. 

Today engineers are facing very different requirements and challenges than they were just a few short years ago. This has been acknowledged by Deutsche Messe, organisers of the Hannover Messe in April 2020, which will focus on the many changes that are currently taking place in industry, as it moves towards Industry 4.0 under a lead theme of ‘Industrial transformation.’ "The world is embarking on a period of change, the likes of which we haven’t seen for a long time," said Dr Jochen Köckler, managing board chairman at Deutsche Messe. 

The challenge for industry now is to take control and actively shape this transformation process, and that means investment in areas like Industry 4.0, Logistics 4.0, artificial intelligence, and 5G. It is vital that engineers seek inspiration and fresh ideas for the ongoing changes they need to make in their operations.