Key Takeaways
- Deploying the Internet of Things in the industry needs not be complicated nor expensive.
- Skoda Auto reports addressing more quickly and efficiently problems occurring in the production line with a custom IoT solution engineered in two weeks
- Toyota Peugeot Citroën Automobile can make early predictions of production stoppage and take remedies thanks to an under-€500 device manufactured in two weeks
- IoT also helped optimize the productivity and minimize the stress of dairy cows through an inexpensive device which can operate for 5 years on a battery in an outdoor environment
Tomáš Morava, CFO and co-founder at Hardwario, recently presented at WebExpo 2019 in Prague three practical, industry-oriented applications of IoT in the automotive and agriculture sectors. In his presentation, Tomáš Morava illustrated how deploying the Internet of Things in the industry needs not be complicated nor expensive. He discussed solutions from Škoda Auto (how to respond more quickly to production problems), TPCA (how to pilot predictive maintenance), and IFRAMIX (how to get online data about heat stress in chicken farms and other agricultural premises).
InfoQ interviewed Morava and obtained additional details on the aforementioned industrial applications. Answers have been edited for clarity and length.
InfoQ: For those of our readers who would be somewhat unfamiliar with IoT, can you quickly explain what it is and what it encompasses?
Tomáš Morava: The Internet of Things (IoT) encompasses the world of hardware devices connected to each other and to the users via the internet. It is a very common word in today’s digital era and the meaning is very broad. The main purpose is that all the devices (such as sensors, actors, measures, displays) provide data which can then be analyzed in a holistic way and interpreted by end users to make better decisions, resulting in increased control, user experience and such.
InfoQ: Your bio mentions that you have been managing investments and finance for private equity investors for 15 years. What led you to create Hardwario? What is your vision of IoT, and its field of applications for business today?
Morava: I love doing new things and I have always dreamed about building my own company, and leverage the experience I have accumulated over the years. The latest technological innovations together led me, together with great friends (Hardwario’s co-founders) to the world of IoT with one simple vision: to enable everyone, every kid, teacher, maker or manager to build their own, simple IoT devices. That is why we focus that much on education and simple modular solutions in IoT in a range of industries.
InfoQ: You gave three illustrative examples of actual, practical applications in the industry: Škoda Auto, Toyota Peugeot Citroën Automobile , and IFRAMIX. May I ask you for each of them to present quickly key data about the company, the problem they wanted to solve, the IoT solution, and the results brought out by the solution, if they can be shared. Let’s start with Škoda.
Morava: Skoda Auto, for those who do not know of the company, is a Czech automobile manufacturer, with a production of around 1.25 million automobiles in 2018, and production plants in Europe, Asia and Africa.
Skoda wanted problems in production to be addressed more quickly and efficiently. When an error would occur on the production line, operators would call their supervisors on their personal mobile phones, or wave their hands to catch the supervisors’ attention on the other side of the production plant.
In that context, we engineered an IoT solution inspired from previous experience in Japan where the Andon system for visual checks of production was developed. Display boards are located near the production lines, providing information on the current state of the workplace, including immediate notifications of issues occurring during the production process.
Škoda thus needed a device to control the display boards directly from the production lines. The device had to alert the relevant operators via the display board to immediately address the issue. At the same time, it had to notify the management. Thanks to the alerts, the management may set priorities in a better way and improve the processes as well as the operation of the line.
Our device, Škoda Andon, was built based on our BLOX IoT Kit, combining four modules: a Core Module, a Battery Module, a LoRa Module, and three external buttons:
The entire device was put into industrial control boxes equipped with the three buttons, distinguished by colours. RGB LED strips were used for the visual design of display boards, since they can be readily used to display numbers, text and symbols.
We have implemented the device with buttons as well as the display boards into all production sectors. Display boards show changes in real time once a button has been pressed. If the line operator encounters an error, they press the "Warning" button. The technicians and superiors can thus immediately see that there is a problem in the given production sector which needs to be addressed. If a serious problem occurs on the line, the operator presses the "Defect" button. Their supervisors then stop the line until the defect is repaired. As soon as the problem is resolved, the operator presses the "Production" button.
The entire system may be integrated into software solutions used in Škoda, such as MS Azure IoT, MS Teams and MS Office 365.
It took us about 2 weeks to implement the Škoda Andon prototype and to collect the first pieces of data. The new system was then fully implemented within 2 months. This is to illustrate that an IoT project can be realized in a short time frame.
About results, I would mention increased safety, better management and productivity of the production line. Additionally, Škoda Andon collects data on warnings and defects during the production process in the long term. This provides the management with valuable input when changing the production flow, redesigning processes as well as when introducing technical improvements and operator quality assessment.
InfoQ: The second example you gave was Toyota Peugeot Citroën Automobile.
Morava: Indeed. TPCA is a joint venture between Toyota and the French Groupe PSA. Their manufacturing is also here in the Czech Republic. In 2017, they produced over 3 million cars.
Here, the problem was to help the TPCA automobile plant to prevent production stoppage. To give a little bit of context, such large plants need to operate as long as possible close to its maximum capabilities or capacity. Production stoppage in a part of the production line may lead to stopping the entire car mounting process, and result in huge losses. TPCA had already identified that an overheated mounting line engine may be one of the possible causes of such stoppage. They wanted to start an IoT pilot project which would monitor the engine temperature in a non-invasive way and would report early increased temperature values. They wanted a device which would operate on battery for a long period of time (6 months or more). Additionally, for the reasons mentioned in the beginning, the device had to be installed in a very short time during a planned stoppage of the line. Importantly, temperature data had to be sampled and sent at a very high frequency.
To give an idea, a mounting line looks like this:
Solution-wise, we modified our standard IoT kit, which already fulfilled partially the requirements (wireless communication and low-cost), to use a modified radio protocol within the 868 MHz band. We also added to the solution external temperature sensors connected to a separate module and a Raspberry Pi system with radio dongle as a communication gateway.
So we prepared a total of 3 temperature sensors, which TPCA technicians then connected, in a non-invasive way, to the line engine casings. These sensors send the measured temperature figures wirelessly at very short intervals to the Raspberry Pi system and further to the car manufacturer’s IT system. The sensors are powered by conventional alkaline batteries.
The gateway, i.e. the Raspberry Pi system with radio dongle, receives data from the sensors. The LAN network then sends them to the Grafana application for their further processing and visualisation.
At this moment, a notification service sends a warning e-mail to the maintenance staff if the predetermined temperature is exceeded. The technicians then inspect the engine and propose measures to prevent defects and unplanned line stoppage.
There again, the development of the entire solution, including the modification of the radio protocol, took two weeks. for a cost under €500. This again goes to show than IoT projects can be built within a reasonably short-time in a cost-effective manner.
InfoQ: Third example was IFRAMIX. I found it interesting to see this example, as it comes from a field, agriculture, which some people may not have in mind when they think about applications of IoT.
Morava: IFRAMIX is a Czech company and a member of CCPA, a market leader in animal feed. Prevention of animal stress provoked by for example inappropriate temperature and humidity in farms is one of their current topics.
Dairy cows, for example, are very sensitive to ambient temperature and air humidity, especially at high production levels. These two factors produce thermal stress, referred to as the THI (Temperature Humidity Index). Thermal stress significantly reduces performance and can cause death. So, the dairy cows produce less milk and its quality deteriorates. The animals do not grow enough, and this is reflected in the farmers’ lower profits. To give some figures, under good conditions, a cow may produce 45 liters of milk per day. On the other hand, that can go down to 20 liter per day under poor conditions, that is 55% of the optimal output.
Therefore, for efficient farming settings, farmers must know in detail the weather forecast and climatic conditions within the farm. Often, they are aware of them, but take only a few factors into account or measure them irregularly and do not reverse engineer the data. Such data is inaccurate and misleading. For example, ambient temperature itself is not predictive and air humidity still has to be taken into account when measuring.
IFRAMIX has been dealing with the issue of animal thermal stress for a long time, with solutions for all animal categories, such as feed supplements based on plant natural extracts. With accurate prediction about the level of thermal stress, farmers can prepare recommended farming measures in time. For example, special ionic and herbal products can be added to the food, they may adjust the ration or the respective premises can be ventilated more.
IFRAMIX needed a set of mobile sensors to get the necessary information about the breed by constantly measuring temperature and humidity. The device was required to be wireless with a long battery life to allow farmers to move it as needed. Communication also had to be ensured in areas not covered by the standard GSM network (as may be the case in remote farms).
That is about the problem. The solution involved temperature, humidity and CO2 sensors from CHESTER and COOPER devices; communication within Sigfox and NB-IoT (Narrow-Band IoT) networks; and a custom API to enable connection to the client’s internal system.
About the Interviewee
Tomáš Morava is CFO and co-Founder @ HARDWARIO. Morava has been managing investments and finance for private equity investors for over 15 years, and his experience comprises pure financial investments as well as industrial strategic oriented companies. His background is hence financial and M&A with strong orientation on value-generating technological companies and innovative projects. In 2017 he decided to start his own business and became a co-founder of HARDWARIO - a start-up company bringing IoT kit for makers called BigClown and helping other companies with the development of IoT devices and bringing the digital innovations into life.