Categories
News

What AI brings to production – Device Insight in the SZ column

What AI brings to production – Device Insight in the SZ column

The AI breakthrough in local production is still to come. Why is that? This is the question dealt with by SZ columnist Helmut Martin-Jung. He spoke with experts from Device Insight, KUKA and Sentian. An overview of the status quo of the “Software of the Gods” in the industry.

AI in production

Süddeutsche Zeitung: “The artificial brain needs data”

Is there any room for improvement in the production halls of German car manufacturers and mechanical engineers? Haven’t things been running like clockwork for a long time now, SZ editor Helmut Martin-Jung asks in the column “Silicon Future”. In fact, with the help of artificial intelligence, also being referred to as the “Software of the Gods”, there is still room for improvement. As Dr. Christian Liedtke from KUKA explains, the potential lies “in the combination of individual production processes”. The robot manufacturer from Augsburg is also subject to the trend towards a relentless increase in efficiency and knows about the advantages of combining AI and IoT.

This is why KUKA, “under its subsidiary Device Insight, has joined forces with the Swedish AI specialist Sentian. While Device Insight ensures that all processes involved supply data, it is up to Sentian to interpret the data and draw from it the right conclusions,” explains SZ editor Helmut Martin-Jung.

The biggest challenge would be that the artificial brain needs data in order to be able to provide useful information for, for example, process optimization. “Values from sensors alone are not enough,” emphasizes Cloud & AI specialist Martin Dimmler from Device Insight. And therein lies the crux: data generated varies from company to company. AI therefore had to be adapted accordingly. “The one blue-print solution, which is simply placed in the company’s data center and switched on, does not (yet) exist”, is the final conclusion of the SZ.

Ultimately, the decisive question is: “What’s the real point of having AI in production?” Each company has to decide that for itself. When it comes to weighing up the costs and benefits, manufacturers are not alone. Feel free to contact us to find out more about the concrete business value of AI & IoT for your company.

Mockup Predictive Maintenance EN 1280x960

Would you like to be inspired by the examples of other companies?

Learn more about real use cases from paper, pharmaceutical and sensor industries in our whitepaper.

Mockup AIoT EN
Download now for free

Recommended posts

Image
2020/12/01
Press releases

How the combination of artificial intelligence and IoT makes the smart factory a reality

Combining AI and IoT can pave the way to the smart factory.
Image
2020/11/20
News

Artificial intelligence – where does the industry stand?

Artificial intelligence is controversially discussed, but how far is the industry really?
Image
2020/11/03
News

Device Insight and Sentian in Handelsblatt – Linking IoT and AI

We describe in Handelsblatt how AIoT can automate production intelligently.
Categories
Press releases

How the combination of artificial intelligence and IoT makes the smart factory a reality

How the combination of artificial intelligence and IoT makes the smart factory a reality

Augsburg / Munich – Part 4 of the digital press conference series “Join us for a coffee…” took a look into the future of manufacturing and showed how industrial companies can increase efficiency, product quality and revenue by linking artificial intelligence and the Internet of Things – this time with the expertise of KUKA, Device Insight and Sentian.

Smart Factory

When it comes to combining artificial intelligence and IoT, industry has so far clearly focused on predictive maintenance. “A mistake,” says Dr. Christian Liedtke, Head of Strategic Alliances at KUKA, with conviction. As the expert made clear at the beginning of the virtual discussion round: “If companies focus exclusively on predictive maintenance, they can only achieve better availability of a single machine, which shouldn’t fail anyway.” What end users are really interested in is generating more revenue. “To achieve this, however, all those involved in the process must work better together and individual processes must interlock seamlessly.”

Smart production with Artificial Intelligence of Things

One approach enabling such a holistic optimization of production is the combination of artificial intelligence and the Internet of Things to form an “Artificial Intelligence of Things” (AIoT), as created by KUKA subsidiary Device Insight and AI specialist Sentian. Here, the aim is to continuously reduce deviations from the optimum within a manufacturing process and to automate improvements. As initial applications of AIoT show, it is precisely the fine adjustments of industrial production that can exploit enormous potential to increase the quality of goods produced and overall yield. According to McKinsey, this will enable an increase in efficiency of up to 30 percent. The key is therefore to synchronize AI and IoT technologies.

"If you think of AI as the brain of Industry 4.0, the Internet of Things functions like the nervous system in the body of a factory. Only when both systems work together, i.e., when well-prepared data is applied in large quantities to state-of-the-art machine learning methods, can the full potential of AIoT be exploited.“
Martin Dimmler
Martin Dimmler
Head of Solution Strategy & Solution Engineering, Device Insight

„In 10 to 15 years, artificial intelligence will be in every production process.“

This is why IoT pioneer Device Insight has joined forces with the Swedish AI specialist Sentian. Together, they are now able to accompany companies on the way to intelligent production – away from individual solutions and selective improvements, such as those possible with predictive maintenance, and towards a holistically optimized smart factory.

„In 10 to 15 years, artificial intelligence will be in every production process”, says Martin Rugfelt, CEO of Sentian. “In fact, AI is already important for many industrial companies. It can reduce energy consumption in the chemical industry, cut waste in the pharmaceutical industry, handle variation in paper production or optimize production lines in discrete manufacturing. For example, JUMO, a German manufacturer of automation and sensor technology, has been able to increase the proportion of its sensors in the highest quality class by 8 percent.”

Mockup Predictive Maintenance EN 1280x960

You want to learn more about AIoT?

Find out in our white paper how you can turn your production into a smart factory by combining AI and IoT.

Mockup AIoT EN
Download now for free

Recommended posts

Categories
News

Artificial intelligence – where does the industry stand?

Artificial intelligence – where does the industry stand?

Artificial intelligence (AI) is a much-discussed topic that can sometimes be seen as a double-edged sword. On the one hand, it revolutionizes the industry and fundamentally improves common processes and manufacturing methods, but on the other hand, AI is already feared as the big job killer that could make thousands of industry workers obsolete. What is the true status quo in terms of AI and where is this technology already in use?

Mockup AIoT EN
Whitepaper "Intelligent Automation with AIoT"
Read now!
Artificial-Intelligence

Judgment: highly expandable

If you look closely, AI has already arrived in people’s everyday lives – regardless of the industrial context – in the form of voice assistants in smartphones or home systems including Alexa and Google Home, but also in intelligent route planning that is able to incorporate road works and traffic jams in real time, or as product recommendations in e-commerce.

In industry, in contrast, AI is far from being as widely used as you might think. An example of this is predictive maintenance: according to a recent bitkom Survey, 15 percent of respondents believe that predictive maintenance is already an industry standard. In fact, predictive maintenance is currently used in a mere two percent of companies. The enormous potential of artificial intelligence is therefore far from being exhausted in the manufacturing industry and its use is still in its infancy in Germany.

Indispensable for the future

All agree that AI is a future technology of the highest relevance and indispensable for sustainable business success. These are the results of a study by Deloitte, for which AI experts were surveyed worldwide, including 200 German decision-makers. The will to use AI is definitely there. There is however a wide lack of implementation. Above all, there is a general lack of technical know-how and company resources. AI specialists and skilled workers, on the other hand, are scarce and are correspondingly sought after in the labor market.

One way out of this situation is the “AI-as-a-Service” approach or the purchase of ready to use AI solutions. They make it possible to approach and implement AI projects, even if there is a lower level of knowledge or fewer staff in the company. This approach does have its disadvantages: With “off the shelf” AI, a company’s individual framework conditions and challenges can only be addressed insufficiently. Basically, the focus of companies is on AI projects with small, feasible use cases that deliver rapid added value. On average, projects in this country pay for themselves within 1-2 years, faster than the international average.

Despite all the positive developments, certain reservations remain. Skeptics fear that the technology could subjugate people instead of supporting them. In fact, repetitive tasks can be taken away from people, freeing up time to focus on other, more creative processes. In reality we are still a long way from ideas from the world of science fiction in which intelligent machines “take over” power. At present, it is generally the case that the AI is given a clearly defined application framework with precise tasks and objectives – and it does not deviate from this.

Where AI is actually being used

What exactly are the current applications for AI in industry? By far the most common and well-known use case is – as mentioned before – predictive maintenance. Thanks to predictive maintenance, technical errors and problems in the production process can be detected and eliminated before machines fail and production comes to a standstill.

Another frequent field of application is quality inspection and assurance. Here, by means of intelligent image recognition, optical inspections by AI applications can help to carry out error or process analyses. Concrete examples include surface inspection, measurement of size and shape, completeness check, object recognition as well as position recognition.

In industrial product design, for example, components are already designed in a highly optimized way by means of complex AI calculations based on Big Data. For example, weight is reduced during the development phase, raw materials are used optimally, and costs are saved from the outset.

It depends on the data

How well these and other AI use cases work in practice, however, depends on the available data, keyword Big Data. Without data, AI application algorithms cannot be trained. The systematic linking of IoT and artificial intelligence has, correspondingly, great potential. Device Insight builds on this knowledge together with Swedish AI specialist Sentian, offering an innovative approach to optimizing production processes in an integrated manner with its joint “Artificial Intelligence of Things” (AIoT) approach. Machine learning can be used to reduce deviations in the production process and significantly increase overall quality and yield.

Mockup Predictive Maintenance EN 1280x960

Curious to learn more?

The whitepaper “Beyond Predictive Maintenance – Intelligent Automation with AIoT” provides a well-founded insight into this subject, including practical examples and recommended steps for implementation.

Mockup AIoT EN
Download now for free

Recommended posts

Categories
News

Device Insight and Sentian in Handelsblatt – Linking IoT and AI

Device Insight and Sentian in Handelsblatt – Linking IoT and AI

“Pioneer of the Internet of Things for Industry” meets Swedish AI specialist – this is the new AIoT partnership between Device Insight and Sentian. In an interview with Handelsblatt, Martin Dimmler, Business Development and Solutions Lead Cloud & AI at Device Insight, and Martin Rugfelt, founder of Sentian, describe how the innovative combination of IoT and AI is enabling the path to the smart factory.

Smart Factory AIoT Handelsblatt

Handelsblatt: “AIoT to help networked factory breakthrough”

In today’s edition of the Handelsblatt, Martin Dimmler explains that the possibilities of the Internet of Things are far from being exhausted. “Most industries have relied on predictive maintenance only to improve availability in production. It’s not enough.” The technology’s potential is enormous, especially when IoT and artificial intelligence are integrated. Production efficiency can, for example, be increased by up to 30 percent.

This is why Device Insight has joined forces with the Swedish AI specialist Sentian. The joint solution “Artificial Intelligence of Things” (AIoT for short) is based on the core competencies of both partners and could become the next game changer in the industry. “Device Insight contributes the experience gained in connecting machines to the Internet of Things and in managing the data obtained, while Sentian contributes the algorithms that are designed to show deviations within individual production processes or entire plants,” describes Handelsblatt journalist Axel Höpner. Sentian founder Martin Rugfelt is also convinced that the joint AIoT approach points the way to the future: “In 10 to 15 years, artificial intelligence will be part of every factory’s production process”.

Mockup Predictive Maintenance EN 1280x960

Would you like to learn more about AIoT?

Further information can be found in our free whitepaper.

Mockup AIoT EN
Download now for free

Recommended posts

Categories
Whitepaper

Artificial Intelligence of Things – Why predictive maintenance is not enough

Artificial Intelligence of Things – Why predictive maintenance is not enough

For a long time it was considered the number one scenario for AI and IoT applications in industry: predictive maintenance. However, companies that focus exclusively on predictive maintenance leave the enormous potential of AI and IoT largely untapped. The full value contribution of these technologies can only be realized when the entire production process is optimized on the basis of AIoT (Artificial Intelligence of Things). Device Insight and Sentian, a Swedish specialist in industrial AI, are pooling their expertise for this innovative approach. In our joint whitepaper we show companies how they can transform their production into a Smart Factory with the help of AIoT.

Mockup AIoT EN
Whitepaper "Intelligent automation with AIoT"
Read now!

The combination of IoT and artificial intelligence offers companies numerous advantages.

Why predictive maintenance is not enough

Predictive maintenance is one of the most important and most discussed technologies at the interface of internet of things and artificial intelligence. Failures and maintenance requirements are to be predicted precisely, thus preventing the failure of machines and significantly reducing service and maintenance costs. However, if one takes a closer look at the industrial manufacturing process, one finds that predictive maintenance does not cover all AI scenarios by far. Here, AI and IoT merely act as “observers” and do not go beyond mere alarming in case of damage.

At the same time, AI and IoT technologies can contribute a much greater value to industrial production if one moves away from the focus on predictive maintenance and instead looks at the entire production. In fact, it is precisely the gradual improvements and fine adjustments of the manufacturing processes with AI and IoT that offer promising business value, with the prospect of an efficiency increase.

Better performance, higher quality und more yield

This is where Artificial Intelligence of Things (AIoT), e.g. the smart combination of artificial intelligence and IoT, comes in. The basic goal of the innovative approach of Device Insight and Swedish AI specialist Sentian is to significantly minimize variation from the optimum within the production process. Fewer deviations mean improved machine and system performance, less waste and lower costs – and above all, more highest-quality products. The result: income and profit, as well as customer satisfaction will increase noticeably. Production will be transformed into a smart factory. Companies that optimize their production processes with AIoT are able to increase the efficiency of their production by up to 30 percent.

AIoT Benefits EN

Use Case: JUMO goes smart factory

One example is the German manufacturer of automation and sensor solutions JUMO. JUMO operates highly automated high-tech production facilities with advanced machines and many robots. Nevertheless, the company found that small but significant fluctuations occur within the production process. This led to deviations in the quality of the sensors produced and inevitably to a lower yield. With the help of intelligent automation – above all a special imputation model for missing data – the sensor accuracy could be increased significantly. In this way, JUMO was able to increase the proportion of sensors in the highest quality level by 20% and developed its production into a smart factory.

AIoT = Sentian x Device Insight

In order to use the full potential of intelligent automation, two technologies must be combined: Artificial intelligence and IoT. Only if well-maintained, high-quality data is available centrally, ML models can be applied on this basis for initiating a “learning process” within the production processes and deriving automated measures. To this end, Device Insight and Sentian bundle their know-how and accompany companies in a 5-step process on their way to becoming AIoT pioneers – from the analysis of the status quo and the IoT and AI readiness to the use case design, implementation, validation and company-wide integration and scaling.

The clear focus here is on the proof of value, i.e. the actual business added value that companies achieve through AIoT, and not – as is so often the case – on the purely technological feasibility, the proof of concept. At several points in the AIoT process, Go / No Go milestones are therefore provided at which – on the basis of clearly defined KPIs – a decision is made as to whether the proof of value has been achieved or whether adjustments need to be made. Through this clear documentation and an iterative, agile process of small steps, we ensure that companies really achieve their goals and the corresponding business value.

Mockup Predictive Maintenance EN 1280x960

Curious?

The free whitepaper will tell you more about the innovative AIoT approach and provide insights into real use cases.

Mockup AIoT EN
Download now for free

Recommended posts

Categories
Press releases

Device Insight and Sentian launch the era of “Artificial Intelligence of Things”

Device Insight and Sentian launch the era of “Artificial Intelligence of Things”

Munich / Malmö – Device Insight, established provider of IoT and IIoT solutions, and Sentian, a Swedish specialist in industrial AI, are pooling their expertise to help companies optimize their production processes along the lines of a smart factory. This cooperation combines both of the most important current fields of technology, AI and IoT, to form an “Artificial Intelligence of Things” (AIoT) and at the same time take the intelligent automation of industrial manufacturing processes to a whole new level, enabling companies to increase the efficiency of their production by up to 30 percent.

Device Insight and Sentian - Artificial Intelligence of Things

Until now, most industrial companies have concentrated on predictive maintenance, leaving the opportunity to optimize their core processes with the help of artificial intelligence unused. In fact, it is precisely these gradual improvements in production processes that offer promising business value, enabling companies to significantly increase their product quality level as well as the efficiency of their operations.

AIoT makes industrial production “smart” – consistent and durable

The goal of the innovative AIoT approach is to continuously reduce deviations from the optimum within manufacturing processes. Fewer deviations mean improved machine and system performance, less waste and lower costs – and above all, more highest-quality products. The result: income and profit, as well as customer satisfaction will increase noticeably. Production will be transformed into a smart factory.

Linking IoT know-how and AI expertise

For the implementation of AIoT projects, Device Insight brings its expertise in connecting machines, aggregating and managing IoT data and linking AI applications into the partnership. Additional added value is created by the Munich-based IoT pioneer’s many years of expertise in the analysis and visualization of evaluations based on high-performance IoT components. Swedish AI specialist Sentian contributes its advanced algorithms and solutions that help reduce deviations within individual production processes or even entire plants. Sentian’s mathematical optimization approach is groundbreaking, allowing fast and extremely precise planning as well as flexible replanning throughout production. Another special component is Sentian’s novel, model-based approach to “Reinforcement Learning” – the latest development in deep learning.

Why predictive maintenance is not enough

Thanks to this unique combination of AI and IoT, Device Insight and Sentian are now able to accompany companies on the way to intelligent production – away from individual solutions and selective improvements, such as those possible with predictive maintenance, and towards a holistically optimized smart factory.

“Predictive maintenance is still very important for the industry. When it comes to process optimization, however, predictive maintenance can only be of limited help. The real challenge within industry lies elsewhere. These days, many control systems are outdated and not very adaptable, while at the same time machines are becoming increasingly complex. This is the conflict area where we begin with AIoT. Together with our partner Sentian, we want to help companies fully exploit the hidden potential for better efficiency, higher quality and ultimately more profit.”
Marten Schirge, Managing Director Device Insight
Marten Schirge
Managing Director and CSMO at Device Insight

“Bringing AI to the core of production enables companies to truly benefit from AI” says Martin Rugfelt, CEO at Sentian. “The potential of AIoT and our cooperation to deliver fully scalable solutions provides proof of value rather than just technical proofs. AI is ready to be operationalized.”

Mockup Predictive Maintenance EN 1280x960

Do you want to learn more?

Learn more about the innovative AIoT approach and get insights into real use cases in the free whitepaper.

Mockup AIoT EN
Download now for free

Recommended posts

Categories
News

A perfect match: AI & IoT

A perfect match: AI & IoT

It is supposed to change everything: artificial intelligence. According to a  Bitkom study , 60% of the companies surveyed are convinced that AI is the most important future technology. At the same time, experts now agree that AI and IoT should not be considered separately. Why is that?
Mockup AI in IoT EN 800x400
Whitepaper "Artificial intelligence in IoT practice"
Read now!
AI & IoT

Through the Internet of Things, enormous amounts of operational and production data are collected and evaluated in many areas. In order to generate real added value from this IoT data, companies need AI and analytics tools to help quickly identify problem areas (anomaly detection), derive decisions and automate optimization measures. Only the combination of AI & IoT can bring that decisive breakthrough in digital transformation.

AI & IoT in practical application

When it comes to the practical application of AI & IoT, some companies are already ahead of their competitors, as shown in the Bitkom study 2019. Swabian technology enterprise Voith uses KI & IoT in the service and maintenance of large facilities. Important factors here include a metadata model that brings together all the information an engineer needs to commission a machine as well as the use of machine learning to analyze raw data on which basis smart services can then be implemented.

Starting with simple ML algorithms

All the hype about AI leads many companies to imagine the application of AI and machine learning to be highly complex. In fact, practice shows that simple ML algorithms, such as linear regression, often have a significantly stronger prognostic power than complex ones, such as Gaussian distribution or k-Means.

“In many use cases, linear regression is sufficient as a statistical analysis method to make predictions about when machine parts or tools need to be replaced"
Hendrik Nieweg
Hendrik Nieweg
Vice President Solution Management at Device Insight

For example, Device Insight implemented an algorithm for robot manufacturer KUKA that predicts when the next maintenance is due for a specific robot type.. The forecast is based on usage data from the robot and uses linear regression as well as a “Generalized Additive Model” (GAM)-based forecasting method. The advantage of this is that the robot is only maintained as required, rather than too often or unnecessarily. From a company point of view, such predictive maintenance naturally makes much more sense.

More efficiency, more sales – more satisfied customers

Many companies are still in the early stages of developing an IoT application that gives them a transparent insight into their operating and production processes. However, the next stage – increasing reliability, efficiency, and productivity – is already imminent. This is where sophisticated AI functionalities come into play, especially ML algorithms.

Used together, AI & IoT open up numerous opportunities for companies: amounts of complex data generated by IoT projects can be evaluated by AI at high speed; knowledge gained flows directly into improving production processes and product quality, which in turn leads to more efficiency in production, more sales and returns and, last but not least, to higher customer satisfaction.

Mockup AI in IoT EN 1280x960

Curious?

Learn how to implement AI in your company with simple steps in our white paper.
Mockup AI in IoT EN 800x400
Download now for free

Recommended posts

Categories
Press releases

Three steps to a successful AI project

Three steps to a successful AI project

Artificial intelligence applications are becoming more and more tangible – particularly in the industry sector. This raises hopes as well as fears. Companies therefore need to know exactly what’s behind the hype, carefully weighing up all opportunities and risks involved. In an exclusive white paper, Device Insight provides IT leaders, CIOs and production managers application-oriented insights into “AI in IoT”.

IoT Trend: Artificial Intelligence

AI is becoming more and more important for industry. This is mainly due to the fact that applications of human-machine interaction and collaboration are becoming ever more tangible. This includes remote monitoring and control of production processes, which are becoming simpler and faster thanks to modern augmented and virtual reality applications. At the same time, consumers are worried that this could lead to “Robot Apocalypse”, destroying millions of jobs for people.

Consultants at PricewaterhouseCoopers have predicted that 35 percent of professional activities in Germany could be automated by 2030, particularly in the transport and logistics, production and sales sectors. This is precluded by a shortage of up to four million mainly highly skilled workers, but this is precisely where AI also offers perspectives. Forecasts predict that digital technologies will lead to entirely new types of jobs, currently unknown to us.

Higher expectations, greater need for information

According to Hendrik Nieweg, Head of Solution Management at Munich-based IoT platform provider Device Insight, “Machines are very talented, but they will not be in a position in the near future to develop intuition or be creative”. Industry expectations are high, Nieweg continues, but it must be explained that artificial intelligence and machine learning are not a universal remedy and that it is better to start with simple algorithms and clearly-defined applications rather than going straight for the big leap.

Predictive maintenance: a clear case for machine learning

As part of AI, machine learning offers industry companies new possibilities, using mathematical algorithms to recognize patterns to deduce new possibilities for action. A prominent area of application for machine learning is predictive maintenance. Device Insight core competencies include the collection and storage of operation and production data provided by sensors in an IoT platform which can then be analyzed using ML algorithms to precisely predict critical machine conditions and plan maintenance intervals as demanded. Consequently, the more extensive the data basis, the better it can be predicted when a service case arises or the “end of life” of a component is reached.

IoT and machine learning in three steps

Munich-based IoT specialist Device Insight uses a three-step ML approach: First, data is collected to show where action is required. Next, in step two, these findings are put into practice in a rule engine. Finally, step three involves the automation of processes using statistical methods and ML algorithms as well as predicting maintenance assignments.

In our white paper “Artificial Intelligence in IoT – Use Cases and Success Factors”, you can see how this works using concrete pplications.

Device Insight shows here how companies can start optimizing their production processes using simple algorithms and clearly-defined use cases. Are you interested in more applications or would you like to talk to our expert Hendrik Nieweg, Head of Solution Management? We are naturally always at your service should you have any further queries.

Mockup AI in IoT EN 1280x960

Curious?

Learn more about artificial intelligence in IoT practice
in the free whitepaper.

Mockup AI in IoT EN 800x400
Download now for free

Recommended posts