The manufacturing industry is experiencing an increasingly rapid paradigm shift. The wave of rapid digitalization that began with Germany’s Industrie 4.0 strategy is now overturning preconceived notions by innovating the structure of industry and how companies do business. There is also a pressing need to work on ensuring a sustainable future by improving the energy efficiency of the energy-intensive manufacturing industry. We asked Kenichi Fujita and Yoshiaki Suenaga to discuss the current work and future tasks for Industry 4.0 in Japan and around the world, and to examine ways of using energy more efficiently. Fujita is the President and CEO of Siemens Japan, a technology provider that assists manufacturers with data use. Suenaga is the Director in Charge of Market/Product Strategies and LSI Development at ROHM Semiconductor, a provider of semiconductors designed to help save energy in industrial equipment.
It has now been about seven years since the Industry 4.0 concept got started in 2011. How do you think data use at manufacturing sites has advanced during that time?
I see a different industrial structure when I look at overseas markets in comparison to the Japanese market, and it seems progress has been different in each. Industry 4.0 is a concept that originated in Germany and is rooted in the idea of digital interconnectedness. Japan is home to a number of companies known as keiretsu that are affiliated with major corporations and have interlocking business relationships and shareholdings. These companies maintain close business relationships with each other. The Japanese auto industry is the prime example. But with 95% of Germany’s private-sector companies being small and medium-sized businesses, there are a countless number of small companies with advanced technologies. The country’s auto industry also lacks the hierarchical industry structure seen in Japan. Carmakers such as Volkswagen, BMW and Daimler do business with subsystem and part suppliers such as Bosch and Continental on an equal footing.
This environment inevitably results in a large number of supplier combination choices when creating product supply chains in Germany. But since the country does not have the same sort of predefined business channels as Japan’s keiretsu companies, German manufacturers need to start the process of creating new products by first creating a network to connect each company taking part. The original aim of Germany’s Industrie 4.0 strategy was to make this process more efficient and effective through the use of digital data. So German companies have been very eager to take part in Industrie 4.0 initiatives. In China and other countries with extensive new plant construction, data use is advancing due to concurrent investment in digitalization. These countries have been able to install new production equipment from the ground up, letting them promote digitalization without having to limit themselves to existing equipment specifications.
But I think Industry 4.0 initiatives will take some time in Japan. Many plants in Japan have always enjoyed a precisely functioning keiretsu system and been able to avoid major problems. I think they’re now ready to work on gathering the data they need to be using. Production lines can have different types of existing manufacturing equipment procured from multiple suppliers, which can make it difficult for manufacturers to create methods of gathering data from sites.
What are the obstacles to digitalization for production sites of Japanese companies?
ROHM uses a vertically-integrated production system in which all processes are done uniformly in-house. These processes range from pulling the silicon ingots used as the starting point for semiconductor chip production, to designing the circuits and layouts within chips, creating masks, processing wafers, testing and packaging. This system is crucial for enabling responsible management of product supply and quality, and we consider it our strength. It’s the reason we have been able to supply industrial equipment and the high-quality products demanded by carmakers for many years.
But with so many processes to be worked on, we use a wide range of different design and production equipment and systems. We also handle many different types of products, resulting in a massive volume of data to be handled in process sequences. And as you mention, different data needs to be analyzed for each system used in production, making data-gathering difficult.
The use of data on production sites is going to become increasingly important, and we are actively embracing it. We’re going to be using data for predictive maintenance. For example, our headquarters will be able to monitor plant operations and use the results to make management decisions, or we’ll be able to detect problem precursors so that problems can be anticipated and prevented before they occur. Implementing preventive maintenance will require us to determine the type of data we need to analyze and visually depict as a resource for management decision-making, as well as the type of data we need to acquire so that we can create methods of detecting potential production line problems before they occur. I imagine that Siemens is examining a large number of sample practical applications for the manufacturing industry. What benefits have you obtained in this area?
On production sites, we’re using an IoT platform called MindSphere that analyzes data acquired from the site to provide visual depictions of trends. Using MindSphere lets us immediately identify information such as the operating rates of each production line equipment item, and differences in production conditions or production efficiency between plants.
To provide predictive maintenance for example, we monitor production line equipment and view historical data on durability and data correlations while using AI to measure damage levels. MindSphere has enabled us to predict replacement times of consumable parts. Power plant gas turbine blades that were previously replaced once in three years regardless of damage levels are now replaced in conformance with their operation load history. This change has created dramatic benefits, letting us extend the periodic inspection interval by 30% and reduce inspection cost by 16%.
Software and AI alone are not enough to advance manufacturing industry digitalization.
Gathering data accurately from sites is no easy task. I feel that many people mistakenly believe that manufacturing industry digitalization only requires software or AI used for analysis or creating visual depictions. But in reality, the most important element is the connectivity that enables the data for analysis or visual representation to be accurately gathered from the site and sent to the cloud. Siemens has worked on over 170 jobs for sites representing markets as diverse as electric power, industrial equipment, public transport/transportation, medicine and telecommunications. We pride ourselves on the understanding of sites this experience has given us, which has enabled us to excel at bringing connectivity to them. No matter how advanced a company’s IT resources are, it will not be able to digitalize its plants without understanding what happens on-site. Does ROHM experience any difficulties in acquiring data?
Plants are special environments that contain many different types of equipment in continuous operation. So data can be hard to gather or transfer in some cases. For example, when installing new data-gathering sensors, new power supplies need to be installed that will reach the location of the equipment. So while plants should ideally enable production lines to be shut down for power line installation, sites with that much flexibility are rare.
ROHM’s plants also don’t enable easy installation of sensors used to acquire required data. But since we’re also a sensor supplier, we feel it’s our duty to develop and provide solutions that can solve this problem. So we have devised IoT devices enabling easy installation.
Wireless communication that has low power consumption, eliminates the need for wiring and enables aftermarket installation can be installed right away to enable immediate improvements. Production processes can be made visible by compiling information on the cloud from data gathered from various sensors using wireless networks such as Wi-SUN (developed by ROHM) and Bluetooth. These sensors include ROHM’s acceleration, gyro and other motion sensors, and environmental sensors that measure data such as temperature and air pressure.
Improving productivity and saving energy: The twin pillars of sustainability
Along with digitalization, environmental adaptation is another crucial requirement for modern plants. Companies are being pressured to ensure global sustainability and the UN’s Sustainable Development Goals (SDGs). So what sort of adaptations do you think are needed by plants that consume large quantities of energy?
Siemens is using three approaches to working on improving the manufacturing industry’s energy efficiency—namely, electrification, automation and digitalization. The German town of Amberg is home to a Siemens plant that began operation over 25 years ago and produces plant control equipment. The plant has automated 75% of its production processes by implementing measures conforming to the three approaches mentioned. It has also improved productivity by analyzing site data every day to change layouts and takt times. Its productivity has also improved by more than a factor of ten relative to when it began operating. We are helping improve energy efficiency by improving productivity and improving the quantity and quality of products that can be produced for the same power consumption.
So your methods dramatically increase productivity with the same power consumption while improving energy efficiency. ROHM is working toward the same goal, but we are helping improve energy efficiency by providing energy-saving devices. For example, power devices are one of the applications we provide. These devices are vital components for power conversion and motor drive. We have developed and supplied new silicon carbide (SiC) -based devices that provide greater power efficiency improvements, and are different from conventional silicon (Si) -based devices. We were the world’s first company to release SiC-based MOS transistors. About 50% of worldwide power consumption is reportedly used for motors. If SiC power devices can be used to improve efficiency by even 1 or 2 percent, they will create a corresponding reduction in energy use and CO2 emissions.
That’s amazing. Improving energy efficiency by just a few percentage points can have a major impact. Power plants make strenuous efforts to improve the efficiency of conversion from thermal to electric power by just 0.1%. For a large thermal power plant, a 1% efficiency improvement is enough to meet the power needs of nearly 10,000 households.
So Siemens and ROHM have differing but complementary approaches to improving energy efficiency.
SiC power devices that help save energy have characteristics that are still a bit rough-and-ready. Making the most of their potential capabilities will require peripheral devices matched to these characteristics, such as gate drivers and shunt resistors. We mix and match the characteristics of devices needed for applications such as power supply, power conversion and motor drive, refining them into easy-to-use forms. We then provide high-quality products that can help make power systems more efficient, more energy-saving, smaller and lighter on a permanent basis.
I’ve gotten the sense that if Siemens makes effective use of the devices that ROHM provides to help save energy, the result could be some powerful system solutions.
That’s right. There could be a lot of areas we can work together on. I’d like to work actively together on helping manufacturing industry digitalization and ensuring global sustainability.
Saving energy at production sites with ROHM’s SiC devices
Plants use a wide array of AC-powered auxiliary devices, such as controllers and data gathering cameras used to ensure precise operation of factory automation. The drive power for these auxiliary devices is a crucial requirement. SCT2H12NZ is an SiC MOSFET developed by ROHM to meet this requirement. It has a voltage resistance of 1,700 V and helps improve auxiliary device drive power efficiency. It’s a power device that provides a highly efficient auxiliary device power supply with characteristics tuned to the high voltage resistance and low current needed for auxiliary device power supply on production lines. It can reportedly reduce conduction loss to just one-eighth the level of conventional Si MOSFETs.
BD7682FJ-LB is an AC/DC converter control IC provided by ROHM for SiC MOSFET drive. AC/DC converters are crucial components for converting the high-voltage AC power supply used in plants to the DC power supply used for driving equipment such as auxiliary devices. Installing SiC MOSFETS for use in this application can reduce energy consumption, but AC/DC converter configurations have conventionally required combinations of large numbers of discrete components. The use of BD7682FJ-LB in this case enables a simple circuit configuration and a compact, energy-saving design.
Combining SCT2H12NZ and BD7682FJ-LB can increase energy efficiency by up to 6% relative to AC/DC converters with conventional Si MOSFETs. Equipment can also be made dramatically smaller and lighter since the number of heat-dissipating parts can be reduced. These devices support a wide range of industrial equipment. In addition to the 400 VAC power supply used by standard industrial equipment, they also support the 690 VAC high-voltage power supply that maximizes the benefits of SiC MOSFETs. They also have multiple protective features that increase the reliability of continuously driven industrial equipment.
ROHM plans to continue providing solutions of high added value that provide high performance, and improve ease of use and reliability.