Internet of Things

Black Box IoT?

Who wants to buy a pig in a poke? Especially when you are entering completely new territory with it… The first stroller, for example: you are already aware that you need it, but the decision between a small pack format or the bicycle trailer combination is more difficult.

The last time I bought a baby carriage was some time ago (now I am an expert in buying skateboards). But if you’re just about to make a choice, I recommend a dark model, because you can’t see the dirt so clearly.

An IoT system is usually not as sensitive to dirt, but its selection is all the more complex. Especially since many companies are breaking new ground with it and can already despair of the selection of the right criteria.

Here I have two good news for you. First: Which IoT solution fits is easy to try out. A Proof of Concept is a very good way to test a solution without major risks before you decide on a system. And secondly, as much as IoT offers new opportunities, smart business processes are based on good old business virtues:

Are you sure what your IoT business will look like and that it will stay that way for a long time? Probably not. IoT is a very volatile market where a lot is happening right now. So your IoT system needs to be responsive and best adaptable by the business department itself. The keyword behind this is “low code”, which means no time-consuming programming to map your processes. And if that’s not enough, the system should be so modular that you can simply “reload” additional components.
Does your IoT business emerge on the greenfield or does it expand your existing business? If the second is the case, then your IoT system should be able to talk to the rest of the IT world in your company: Spare parts orders, for example, should generally have passed through financial accounting once. Open or even certified interfaces are the most important thing here as well.
Do you manufacture industrial goods? Do you also have to do with spare parts and maintenance? Then your new best friend in IoT new territory becomes the “Digital Twin”. But only if it’ s also suitable for industrial use: it must be able to map the components of your system in detail (preferably with the corresponding 3D model), know the current parameters such as software statuses and, in particular, be able to document changes after maintenance or conversion.

To be honest, the connection of devices is usually only a question of fiddling, but usually not a fundamental problem. Step by step, standard protocols for machine-to-machine (M2M) communication such as MQTT (Message Queuing Telemetry Transport) or OPC/UA (Unified Architecture) are gaining acceptance and making life easier for everyone involved.

So: Just try it out with a Proof of Concept! If you also pay attention to the three touchstones, you’re in the running. Then the possibilities of “Analytics”, “Big Data”, “Data Driven Processes” or “Predictive Maintenance” are open to you.

Getting started with IoT in 4 steps

Everyone talks about the Internet of Things (IoT) and the digital twin – they form the framework for new, digital business models. According to a forecast by PwC, digitization will bring the manufacturing industry an increase in turnover of more than 270 billion euros in Germany alone over the next four years.

Companies are hoping for sales growth through smart products and digital business models. This is also confirmed by our current I oT study, which was conducted jointly with the Fraunhofer Institute for Production Systems and Design Technology (IPK) and the Association of German Engineers (VDI). It shows that companies have high expectations, but at the same time makes it clear that there is still a certain reluctance to implement the new legislation in practice. Many companies are faced with the question: “How does it actually work with IoT?”

In my experience, companies often think the second way before taking the first step, which leads to restraint. Of course, it is good and important to have a vision. The picture, which is often published in blogs and forums, usually shows very sophisticated IoT scenarios. They don’t start where many companies currently stand with their business model and technology knowledge.

That’s why it’s important to gain your own experience and gradually approach new digital business models, true to the motto Think big, start small, act now!. Own projects, also together with technology partners, automatically expand the wealth of experience. So why not start using the new technology to support classic business?

With my contribution I would like to show how companies can realize an effective IoT scenario for their business in just 4 steps.

Step 1: The digital twin as a communication interface

The necessary data for the d igital twin is usually already available in the company. The first step is a simple serial number. It serves as a documentation interface and connects the data with the product. 3D data is added later. The data is often already available in PLM or ERP systems – for example from production, purchasing or development – and should be displayed in a dashboard.

Step 2: Generate data via sensors

Sensors are also often already available, for example for controlling devices, machines and systems. They record states such as power, pressure, consumption, etc. This data is now consistently recorded and stored suitably. In this way, the current status can always be viewed. In addition, limit values are defined, for example for excessive current consumption, whereupon warnings can be sent and errors rectified.

Step 3: Initiate smart maintenance work

A detailed damage and wear picture can be derived from the analysis of the data and measures such as maintenance projects can be initiated as early as possible. The digital twin serves as a documentation interface. All adjustments to the product thus remain traceable. This data history can later be used for the development of predictions (predictive maintenance). The digital twin as maintained supports the documentation of product changes, can link them with historical data and thus also prove in which configuration the product functions optimally. The classic product lifecycle is thus extended to its usage.

Step 4: Request spare parts

In addition, the information is used to request spare parts. With the help of compressed service parts lists or spare parts catalogs, the data is assigned to the affected component and the required spare part is delivered in the event of imminent damage. This data is also already stored in ERP systems. This process can be triggered manually or automatically on the basis of the device messages. In this way, companies avoid downtime in their own production.

In these 4 simple steps, an efficient IoT scenario has been implemented and a big step towards a digital business model has been taken. I am sure that many companies will be able to get started with the new technology in this way.

So: Get started and use the experience gained for digital business models!

Smart products have their price

IoT failures were the subject of my previous blog post, and what particularly surprised was a “Smoke detector with integrated microphone that allows monitoring in living rooms” that a well-known manufacturer launched on the market. The question of whether this is really a design flaw or whether we don’t have to put up with it for the comfort of smart products resulted in really interesting, sometimes controversial discussions. One question that emerged is not new, but the trade-off generally concerns users of smart devices:

How many and which kind of private data do I disclose for smart comfort?

In the case of the smoke detector, the advantages are obvious: the networking of the smoke detectors in the house offers greater safety in case of fire. If one detector is triggered, all other smoke detectors are informed and the alarm sounds throughout the house. In addition, the alarm can be forwarded, for example to a mobile phone, so that users are informed at all times. This functionality, does not require a microphone that allows monitoring. However, the high-resolution microphone is required if the smoke detector is to be used in addition to voice control for a “smart home”.

Advantage: Then I only have one device on the ceiling: smoke detector with voice control
Disadvantage: I need a smoke detector in every room, but there are rooms where I don’t want any voice control elements to listen.

Maybe during design of the smoke detector this has not been taken into account or simply an existing circuit design was reused. Here it becomes clear that for the development of smart products it is important to look at the whole package from the user’s point of view:

How should a smart product behave, what is technically possible and what should it not be able to do?

For some products it is not clear whether it’s a useful and safe product, like e.g. a jogging stroller that drives autonomously in front of the running track. Is autonomous driving safer than the person holding the stroller? Because he too could stumble and the stroller could roll onto a road …

Furthermore, the Internet of Things and the ongoing digitalization of different areas of life also offer the opportunity to develop sustainable products and solutions. I would like to drive an electric car whose route planner calculates the electric filling stations needed on the way and suggests filling up at a suitable time, naturally taking waiting times into account. Or, in general, smart home applications that save energy and offer greater safety.

Very interesting are also the possibilities in the industrial area, which can be reached by the use of digital twins of plants or machines: Operating states can be recorded at a glance and the machine can be controlled via apps. Algorithms calculate optimal resource allocations, bottlenecks can be detected, and real-time control becomes possible.

The exciting challenge I see in the design of IoT products is the interaction between hardware and software. What possibilities there are to design sustainable and sophisticated products and to optimize processes, if the overall system is considered! Complexity is a big challenge for designer and developer. And in addition verification, testing and validation of a solution are required to make sure, that products and systems behave as required.