The first part of the Expert View Series “From machine constructor to platform operator” focused on what the status quo in the machine and plant construction industry looks like in the context of digitization, and the second part dealt with the protocols, standards, and technologies with which IoT platforms should be built. The third and final part deals with specific use cases for IoT platform activities and the resulting ideas for digital business models and processes in the industrial environment. It is aimed in particular at IT managers, IoT product developers, and service managers.
Because almost every company should now be asking itself: “Do we need an IoT platform in order to be successful?”
As always, the answer: “It depends.”
If the company is interested in achieving a larger margin through inward automation, it needs primarily Industry 4.0 technologies and not its own platform. Azure, Mindsphere, and the like are sufficient and feature a good project implementation for processes at this point.
But if the company wants to generate more revenue, new digital products or digital products and processes that complement the traditional products will certainly help. At this point, at the very least, an industry add-on to an industry-independent platform or even a specially developed platform is required.
In some circumstances, this results in two strategies: one for digital products and another for digital optimization/automation in manufacturing. For larger companies, this can also be followed by other different platform decisions.
Automation, in particular, is a good starting point for machine learning in increasingly complex process chains and IT architectures. Humans and machines work together to optimize operations in the company. With the easy-to-use cloud-based machine learning offerings (ModelCloud), it is becoming increasingly easy to process data into recommendations or predictions – even without considerable expertise. In addition, Edge Computing can ensure more powerful data transmission and local pre-processing. Through cloud native and analytics applications, completely new application possibilities can be generated.
Use cases in the industry
The wide range of technological applications offered by the IoT results in a broad spectrum of use cases and areas of application. In the context of IoT platforms, the footprint of German companies often spans not just one but rather several use cases. For example, the IoT activities of many companies focus primarily on the service and production areas as well as the measurement of all process activities. In this sense, the areas of digital service processes and digital production processes are currently among the most heavily pursued use cases for the IoT. Many of today’s PoCs and projected solutions have already made it into productive use. The primary aim is to increase the degree of automation in manufacturing and to reduce maintenance and service costs in downstream processes in order to increase profitability with digital business models in the sense of a platform.
These use cases include maintenance process optimization – predictive maintenance is one of the most relevant use cases in mechanical and plant engineering. For example, predictive maintenance uses streaming data from sensors and devices in order to quickly assess current conditions, detect warning signals, transmit alerts, and automatically trigger appropriate maintenance processes.
Maintenance becomes a dynamic and rapid operation – by performing tasks only when they are needed, Predictive Maintenance promises cost savings over routine or time-based preventive maintenance. Because the solution is to get the right information at the right time. Production managers can thus identify which machines or equipment need maintenance. Maintenance work can thus be better predicted and planned. The systems do not have to be taken off the grid – and production can continue. Other potential benefits include longer equipment life, increased plant safety, and fewer accidents that negatively affect the environment.
In particular, machine and plant manufacturers who service their equipment themselves after delivery now have almost all predictive maintenance offerings in their portfolio. However, the legal basis or the trust of customers that is required to make better use of production data is often lacking. Only those providers who change their core business from the CAPEX purchase model, for example, to a Full-service rental model can effectively perform data analytics today because they continue to own the machines in the field.
For example, a machine constructor that offers not only a predictive maintenance solution for new machines but also a retrofit for all of its customers’ existing machines – including machines from other manufacturers – creates significant value for its customers.
This not only changes the service processes for manufacturers or independent service providers. The supplier is also perceived as a digital integrator – and not just a machine builder. Predictive maintenance solutions can actually generate new revenue streams for many manufacturers.
Another use case that finds application in traditional manufacturing companies is the measurement of resource efficiency (asset efficiency). By means of analyses, it is primarily a question of monitoring the plant and production lines within a company.
In addition, asset efficiency analytics enable easy location and traceability of key assets, including along the supply chain (e.g., raw materials, finished products, and containers) in order to optimize logistics, maintain inventory levels, prevent quality issues, and detect theft.
One industry that relies heavily on resource traceability and monitoring is maritime shipping. At the macro level, sensors help track a ship’s location at sea. At the micro level, they can display the status and temperature of individual cargo containers.
Such real-time measurement data is particularly advantageous for refrigerated containers. The cargo of these containers must be stored at stable temperatures so that perishable goods remain fresh. Each refrigerated container must be equipped with temperature sensors, a processing unit, and a mobile transmitter. If temperatures deviate from the optimal mark, the crew can be notified and begin the necessary repairs.
Digital twin for thermal disposal and recycling processes
What is actually possible with IoT platforms is demonstrated by, among others, the Thermal Twin 4.0 project of the city of Vienna (CO2 and resource savings using physical and data-oriented approaches). In the Thermal Twin 4.0 project, an overall model of the combustion processes of a thermal waste treatment plant is being developed using a combination of physical (control and process engineering) and data-driven (machine learning) approaches.
Physical relationships are used to develop a model of the processes within the waste incinerator that can simulate the combustion behavior of the input streams.
Data-based approaches – both statistical and AI technologies – come into play in order to compensate for incomplete data on input streams.
Together, they form a digital twin of the combustion process as a whole. This can then be used to optimize energy efficiency and reduce CO2 emissions from the plant.
Recommendations for your own IoT platform strategy:
- Don’t talk about platforms too soon. In the early stages of digital products, creativity and exploration of the technical possibilities is important. This can happen in different teams and should not be hindered by technology specifications.
- Stimulate IoT innovation. The first IoT projects should not compete with each other but rather exchange ideas and technology experiences. The Chief Digital Officer (CDO) must create the appropriate network between IT, product, and service departments.
- Moderate IoT business models. The biggest challenge is to bring technology and business models together. In many cases, technology is what makes a particular business model possible in the first place. For example, cloud-native technologies sometimes achieve such low operating costs that viral go-to-market models become possible. This is where the CDO must step in.
- Structure IoT use cases. A digital product consists of various components. A company’s digital portfolio, on the other hand, consists of various digital products. After the initial experiments, a portfolio management system must be established for this purpose. This also structures the requirements for a platform.
- Balancing platform independence and platform synergies. As soon as you want to productively deploy an entire digital portfolio or a complex digital product, you should focus all your efforts on a technology stack and a cloud back end. Regardless of which vendor you choose, the question is how much vendor lock-in you want to achieve. High-quality PaaS services in the cloud and commercially licensed software components on the Edge initially accelerate projects. However, this can mean a lot of money and a long dependency on one provider later on.
- Match operations aspects to volume. Unlike the old enterprise IT, in the cloud, it is possible to use many different yet similar services. On the Edge, consolidation to fewer technology stacks still makes sense. If you have chosen primarily open-source frameworks when selecting cloud services, you can use “fully-managed” PaaS services at the beginning. If the volume increases or you want to switch to a different cloud infrastructure, it makes sense to run these services on the container management system Kubernetes itself.
- Use and build skills. While IoT projects in most companies start with external help, at least the parts that differentiate the digital product and implement proprietary intellectual property should be covered by in-house developers from an early stage. Even if skills are usually available in-house for a particular enterprise technology provider, CDOs and CIOs should not underestimate this and encourage a willingness to learn cloud-native technologies.
Sharing expertise – book your free initial workshop!
Book your free initial workshop today in order to be able to identify and expand the digitalization potential in your company.
Identification workshops on the following topics: Digital strategy and data-driven business models, IT infrastructure strategy, data science, process automation, custom production software, digital twin simulation, application modernization, and custom software development.
- Best practices
- Requirements analysis
- Target image/solution concept
- Ideas for proof of concept