Smart manufacturing in the process industries

Smart manufacturing requires the organic linking of process, equipment, technologies and people to achieve stable, sustainable and profitable operations.

Process industry manufacturers are under constant pressure to enhance profitability, improve capital efficiency and achieve sustainability in a business environment characterised by high volatility, uncertainty, complexity and ambiguity.

As a continuous and sustainable path towards growth and improvement, companies, partnerships and government agencies are leveraging the latest developments in digital technology, including new concepts such as smart manufacturing, Industry 4.0 and IIoT, in a process that requires collaboration among manufacturers, partners and other organisations.

Many advanced digital technologies such as artificial intelligence (AI), autonomous robots, cloud computing, intelligent sensor technology and augmented reality (AR) have become practical in terms of cost and value. The question is how to best take advantage of these technologies in an era of rapid technological change and massive competition. The goals of adopting these new technologies include total optimisation of efficiencies, flexibility and agility, safety and other operational improvements.

Background

The fast development and wide adoption of digital technologies are increasingly contributing to the efficiency and competitiveness of manufacturing industries. While digitalisation is still relatively new for many manufacturers, it has become a strategic imperative that underpins the survival of many organisations. All businesses must embrace digitalisation to be able to respond to rapidly changing market conditions and become part of a sustainable society.

For example, the camera film market has shrunk significantly since the introduction of digital cameras, thereby eliminating late adopters. Certain mainstream IT manufacturers have emerged as winners in the hypercompetitive world of PC manufacturing as a result of their supply chain reforms and implementation of one-to-one internet marketing. Thus, manufacturers must take note: the speed at which disruption is occurring means that manufacturers must continually transform their businesses and increase their competitiveness by applying digital technology and improving data utilisation. Waiting to undertake this type of transformation is not wise.

Survival in a continuously changing world requires the adoption of digitalisation on a scale that enables enhanced flexibility and agility that transcend the levels possible with conventional organisational silos. Implementation of vertical and horizontal integration and other means of collaboration are important enablers of these efforts.

Digital transformation can be defined as the novel use of digital technology to accelerate a company’s business strategy. It is about the application of digital technologies to empower people, optimise processes and automate systems of the organisation to radically reorient its business performance. It is, however, important to consider issues such as leadership, new competencies and change management to ensure a successful digital transformation journey.

From industrial automation to industrial autonomy

The process industries have benefited significantly from industrial automation, and many applications have become fully automated and unmanned. The process industries will advance further towards industrial autonomy by focusing on areas of innovation such as predictive maintenance, smart energy consumption, routing flexibility, remote monitoring and control, human-robot collaboration, digital performance management, real-time supply chain optimisation, advanced process control, digital quality management and data-driven demand prediction. We can define industrial autonomy as the state in which plant assets and operations have learning and adaptive capabilities that enable responses with minimal human interaction, empowering operators to perform higher-level optimisation tasks. Digital technologies sit at the core of realising self-optimising manufacturing processes and industrial autonomy.

Industrial autonomy is different from industrial automation in many ways. Industrial automation involves performing a sequence of highly structured pre-programmed tasks, each of which requires human oversight and intervention in case something goes wrong (see Figure 1). In addition, between their pre-programmed tasks, humans must perform other ad hoc tasks and are ultimately responsible for the safe and profitable completion of entire procedures. Examples of procedures requiring significant human oversight and intervention include start-ups, shutdowns, grade changes, quality control adjustments and the managing of abnormal conditions.

Figure 1: Industrial autonomy enhances industrial automation. For a larger image click here.

Industrial autonomy transcends industrial automation by adding layers of intelligent sensing and artificial intelligence (AI) to anticipate and adapt to both known and unforeseen circumstances. This removes the need for constant human intervention. In a fully autonomous operation, the industrial autonomous system is responsible for all aspects of operation, from start-up to shutdown.

Within a plant, any process or operation can potentially be made autonomous. This includes manipulating and controlling the process, as well as performing other activities such as manufacturing operating procedures, planning and scheduling, supply chain activities, margin optimisation and compliance measures. It is also possible to make the operation of devices, equipment, units and business systems — or ideally entire plants — autonomous, so each is self-aware and able to understand and adapt to the context in which it operates.

The most forward-looking companies are beginning to think about autonomous operations, and some are achieving unattended remote operations. Many of the steps necessary to achieve remote operations are also needed to achieve autonomy.

For field operations, this means moving from an absence of autonomy, where all tasks are performed by humans, to mid-level autonomy, where the system identifies tasks and guides operators on what to do and provides instruction on how to accomplish each task. Further along the autonomy journey, manual tasks must be converted to fully automated tasks, with human action only required as an exception. Fully autonomous operations require no human interaction. At this level, robotics plays a key role by conducting routine operator rounds and collecting samples and by performing monitoring, inspection and surveillance. Robots will thus perform all necessary field operations and maintenance tasks.

Process industry manufacturers will need to implement industrial autonomy sooner rather than later to improve productivity, flexibility and profitability. Industrial autonomy will also reduce human errors, remove people from hazardous environments and help compensate for the loss of experienced workers due to the ‘great crew change’.

In certain industries, completely autonomous plants are unlikely in the near future. However, it is reasonable to expect that certain functions will be autonomous based on the application, needs and cost/benefit ratio. In these cases, human intervention and decision-making will continue to be important as plant personnel learn to work alongside autonomous systems.

Smart manufacturing in the process industries

Smart manufacturing is defined by ISO and IEC as follows:

Manufacturing that improves its performance aspects with integrated and intelligent use of processes and resources in cyber, physical and human spheres to create and deliver products and services, which also collaborates with other domains within enterprises’ value chains.

The scope of smart manufacturing is broad and involves a journey. Connectivity beyond plants and enterprises at the IIoT and cloud levels provides seamless integration of data, including supply chain information, leading to smart manufacturing (see Figure 2).

To implement smart manufacturing, it is necessary to not only introduce new technologies as they become available but also prioritise the adoption of each solution based on a coherent vision of the future, along with analyses of the status quo and problems and their expected outcomes.

For many organisations, autonomous operations is the destination to achieve their smart manufacturing goals.

Figure 2: Connectivity beyond plants and organisations at the cloud and IIoT levels. For a larger image click here.

Integrating information to implement smart manufacturing

Many diverse problems hinder the realisation of smart manufacturing. For example, there may be difficulties with integration among systems because of cybersecurity requirements, organisational issues related to empowering employees and inconsistencies in information, demand prediction accuracy and other issues. From a broader perspective, there are many other issues that cannot be managed internally, such as drastic changes in economies and markets.

While there are many existing technologies for smart manufacturing implementations, no simple and general methodology has been established to dictate which technologies should be selected to enable smart manufacturing. Processes, technologies and people need to be linked to realise smart manufacturing, and by integrating systems and data, predictive and quick actions along with adoption of the latest technologies such as digital twins will lead to successful smart manufacturing goals (see Figure 3).

To transform data into information, organisations need to acquire accurate and reliable data and utilise it more quickly by turning it into usable and valuable information. This goal has been difficult to achieve with hierarchical systems, communications and organisations. By establishing vertical and horizontal integration and analysing information, the ability of process industry manufacturers to achieve these goals will be greatly increased.

Figure 3: Information integration and digital twins allow organisations to benefit from the latest technologies. For a larger image click here.

The integration of systems through a digital transformation platform

To realise smart manufacturing, solutions are required to measure and transform plant data into valuable information and intelligence, regardless of whether an organisation stores their data on premises or in the cloud. To effectively use these solutions and applications, the seamless integration of systems is important. Connectivity beyond plants and enterprises at the IIoT and cloud levels enables the seamless integration of business, production and supply chain data.

Traditionally, process manufacturing operations are built and engineered to include a wide variety of mission-critical equipment, control systems and human-machine interfaces. Most of these items incorporate software, and all require some form of system integration. Common items include distributed control systems, safety instrumented systems, tank gauging systems, programmable logic controllers and other automation components. These components receive inputs from instruments, analysers and other field devices. Software logic is applied to these inputs to drive outputs to valves, motors and other equipment.

For capital projects, most design efforts are completed up-front on a one-time basis, with little or no formal plan for changes over the entire process plant operation lifecycle.

To support long-term flexibility and continuous asset and technology updates, OT architectures are moving towards open, modular and interoperable frameworks with strong cybersecurity — such as the Open Process Automation Forum (OPAF) framework.

One outcome of this will be the decoupling of the hardware used for control from the software performing the control functions. This will enable radically different automation system architectures to be created using a small number of commercial off-the-shelf IT hardware in addition to software building blocks, creating a new generation of state-of-the-art automation software.

This new approach will create a high degree of interoperability at the plant floor level, which will also be needed across a company’s broader IT systems. Therefore, a comprehensive view of software architecture must be considered holistically across the organisation.

A new hybrid architecture is needed to integrate OT with IT systems and develop future applications. Such a digital transformation platform can also be used to connect with IIoT devices and cloud-based systems. Edge systems are an important starting point for the convergence of OT and IT. An ideal digital solution is realised by combining edge and cloud data hosted by an on-premise or cloud-based engineering and solutions application.

This digital transformation platform is used to create new opportunities, such as:

• Flexible and scalable connectivity: The traditional connectivity achieved through wiring and lots of software configuration makes data availability expensive. Digital transformation requires data and is characterised as flexible, scalable and affordable.
• Accessibility: The central deployment of data, information and applications makes it easy for the user for gain secure access to the digital system to obtain relevant information.
• Availability and sustainability: Digital transformation solutions have been proven to be reliable, and in conjunction with critical information that is still kept on premises, the solutions become sustainable.

The integration of plant asset information and the automation of engineering

In a smart process plant, information must be handled all the way from the equipment level to the plant level. Plant asset designs can change daily based on feasibility studies of front-end engineering design, detailed design, operation, and decommissioning. In many cases, plant data is designed separately for each purpose, but this data should be maintained consistently throughout the asset lifecycle.

With the right technology and planning it is possible to expand the automation of engineering and the automatic diagnosis of assets. The integration and management of disparate plant data and transforming the data into information provides knowledge and intelligence for multipurpose use throughout the asset lifecycle. This can be done by leveraging digital data, modelling technology, knowledge management, digital twins and information integration.

Conclusion

Smart manufacturing requires the organic linking of process, equipment, technologies, and people to achieve stable, sustainable, and profitable operations, utilising the latest technologies, integrated information systems, and applications such as digital twins as key elements. Taking advantage of these technologies in an era of rapid technological change and massive competition requires a future vision and an understanding of the reality of plant floor operations.

Top image: iStock.com/metamorworks