Smart manufacturing is the use of digitization and connectivity to increase quality, service levels and/or operational efficiency. Industry 4.0 (often referred to as the Industrial Internet of Things1 ) indicates a stage in smart manufacturing where not only machines are connected and operations automated, but the entire production line is visualized and can make decisions on its own.
Smart manufacturing and Industry 4.0 concepts are not new; their technical capabilities and business benefits have been acknowledged and were referred to as a new trend back in 2017.
And in 2016, an article in the IEEE Sensors Journal stated, “While software used to be only part of the production process, nowadays software increasingly defines the production process itself.”2
More and more technical machines and production plants are becoming increasingly intelligent and autonomous. Equipped with network capabilities, they’re able to consume and supply data to others.
In 2018 KPMG estimated that the value of the Industry 4.0 market would be worth US $4 trillion by 2020, while Gartner predicted IoT’s value at close to US $3.7 trillion.
Looking at the speed at which technology is developing, it’s time organizations started piloting and scaling in this area in order to stay competitive.
Digitization and connectivity means that physical systems become connected with cyber space. Smart manufacturing and Industry 4.0 systems and architectures are therefore referred to as “cyber physical.”
Cyber-physical systems integrate software & computing, networking and physical processing. They manage interconnected systems between their physical assets and computational capabilities.
To develop smart manufacturing and Industry 4.0 solutions successfully, the lifecycle of hardware operations and software development needs to be integrated. This is a challenge because the dynamics of the two differ substantially.
The question therefore is: how can organizations set up an integrated environment for scalable and successful smart manufacturing and Industry 4.0 pilots?
Smart manufacturing and Industry 4.0 are important prerequisites for manufacturers to become more customer adaptive, increase both quality of service and operational efficiency, and to stay competitive. Machines are connected with each other via a network of sensors (IoT); data from machine performance can be used (in real time) to optimize processes or to alter product configurations, enabling manufacturers to provide mass customization of their products. You don’t need a lot of imagination to envision the numerous advantages in the areas of operational efficiency, supply chain optimization, reduction in use of resources, time-to-market, energy consumption, predictive maintenance and customer service.
In addition, the increasing amount of software related to physical products provides manufacturers with the opportunity to move up the value chain and provide services alongside or in addition to their physical products (most commonly used in predictive maintenance). These services in return will provide information on product usage and are therefore valuable for continuous enhancement.
Despite the numerous opportunities, related costs to develop solutions are substantial. Investigating the challenges provides valuable insights on factors to take into account when planning for smart manufacturing and Industry 4.0 solutions.
Piloting and scaling smart manufacturing and Industry 4.0 solutions is challenging because of the necessary integration of hardware operation and software development lifecycles. As the lifecycles of the two separate worlds become more intertwined, their development and operations will become increasingly dependent on each other. However, there are several factors organizations need to take into account:
Testing integrated hardware, software and networks is costly because of the involvement of physical components. Piloting and testing are therefore often performed in a small-scale setting. However, these settings are no guarantee the solution will also work if it needs to be scaled within a factory or across factories.
Software developers and engineers have a different mindset to production operators and engineers because the dynamic of the development environment differs substantially.
The tools and techniques used in hardware operations and software development environments differ from each other, which poses a challenge in the integration of the two. Furthermore, network capacities are key and not all network technologies support smart manufacturing and Industry 4.0 solutions.
On top of that, the technical integration also poses specific challenges. In manufacturing, legacy systems are deeply ingrained in functional silos and consist of a large part of proprietary platforms.
Another technical challenge is the connectivity. Networks are still mostly wired (80%) and do not support IoT capabilities (high density of connected devices, security and privacy and differentiated quality of service levels) necessary for smart manufacturing and Industry 4.0. 5G is a promising connectivity technology. Connecting machines with each other, to the cloud and to applications, but also connecting manufacturing sites over a wireless network, is possible with 5G. You can read more on networking for Industry 4.0 by downloading our Industrial IoT services and 5G paper.
Taking into account the difference in mindset, the technical challenges and the costs related to developing smart manufacturing and Industry 4.0 solutions, the question is, how can organizations overcome these challenges and reap the benefits of these solutions?
Industrial DevOps & Software Defined Labs
In software development, the separation of development and operations leads to issues in the area of collaboration, communication and integration. DevOps is a methodology to overcome these issues and shorten the time-to-market and increase the quality of software.
DevOps can also be used in industrial environments, integrating the work of software developers and production operators and engineers. This form of DevOps is referred to as industrial DevOps. Industrial DevOps is the application of continuous delivery and DevOps principles to the development, manufacturing, deployment and serviceability of significant cyber-physical systems to enable these programs to be more responsive to changing needs while reducing lead times.4
DevOps for smart manufacturing and Industry 4.0 is a great way to integrate development and operations of hardware and software lifecycles on an organizational and technical level.
An important prerequisite for using DevOps and to increase the intensity of collaboration between software developers and production operators is the automation of hardware operations and the virtualization or emulation of hardware components. This method reduces hardware costs as there is no need to use the physical component in the development process. It also enables an integration with the software layer and therefore supports the development and deployment of end-to-end solutions. As a result, piloting smart manufacturing and Industry 4.0 solutions becomes less costly, easier to deploy and the quality of the end product increases.
Virtualization and emulation are also important building blocks for digital twins. A digital twin is an exact copy of a physical component in a digital environment. Digital twins exist simultaneously with their physical components, behave exactly the same and change at the same time in the same way. However, this type of integration is costly and not always necessary for smart manufacturing and Industry 4.0 solutions.
In many cases it is more feasible to use a stripped down version of a digital twin, in the form of a laboratory. In such a lab all software and hardware components are virtualized. This enables the piloting of new solutions without the use of physical components, and the entire environment can be set up in the lab. This means you don’t need to work with a subset of the reality (as is the case when building a smaller environment with real physical products). The lab simulates the entire environment. When piloting solutions in such an area, scaling to in-process will be less risky.
Luxoft has built such a solution: the Software Defined Lab (SDL). SDL is a lab solution to automatically test end-to-end solutions. It connects with different proprietary hardware and software products. Network components can also be integrated into SDL. SDL integrates with existing continuous integration and continuous deployment environments and therefore supports the DevOps way of working. Moreover, SDL can be used in all stages of solution development: pre-deployment, deployment and in-process. As such SDL can be used to test pilot solutions, and at the same time it supports scaling and monitoring of solutions to and in the in-process environment. Test benches can be set up in a few hours and numerous different configurations can be tested. The advantage of this is that different setups can be tested for one type of deployment, supporting any number of configurations that customers may use in their own environment. SDL can also be used for solution development and for scaling smart manufacturing solutions.
Moving to smart manufacturing means the behavior and operations of hardware can be changed fast, and even in real time, using software. Piloting such solutions would be costly if you chose to use the physical components (machine and network).
Follow these steps to create viable smart manufacturing solutions:
- Automate machine operation processes.
- Set up a DevOps environment with continuous integration and continuous deployment as a basis to increase time-to-market and continue improvement of the quality of end solutions.
- Use this DevOps environment to support the collaboration between software developers and production operators.
- Enhance the DevOps environment with automated testing based on a lab solution. A lab solution works for piloting and scaling without the immediate need for digital twins (which are costly and more complex). And last but not least, a lab works best with simulated and/or emulated hardware components.
1 S. Jeschke, C. Brecher, H. Song and D. B. Rawat, “Industrial Internet of Things: Cybermanufacturing Systems,” Springer, 2017
2 J. Wan, S. Tang, Z. Shu, D. Li, S. Wang, M. Imran and A.V. Vasilakos, “Software-Defined Industrial Internet of Things in the Context of Industry 4.0,” IEEE Sensors Journal, vol. 16, no. 20, pp. 7373–7380, Oct 2016. DOI: 10.1109/JSEN.2016.2565621
3 KPMG 2017, Beyond the Hype: Separating Ambition from Reality in Industry 4.0