The National Institute of Advanced Industrial Science and Technology (AIST) has opened the Cyber Physical Systems Research Facility in the AI Research Center at Tokyo Academic Park Waterfront. The facility, which started operations in April, is equipped with simulated manufacturing, logistics, and drug discovery environments for human-machine collaboration.
One of Japan’s top priority societal challenges is the shortage of workforce due to an aging population and declining birth rates. The introduction of robotics and artificial intelligence (AI) broadly in the industry and society is considered an important step to mitigate the challenge. According to the Ministry of Internal Affairs and Communication (MIC), the working-age population has declined from 60 percent to 54 percent of the total population since the peak in 2008.
This has a huge impact on the society, which not the least can be seen within the service industry. The 24 hours convenient store chains have today an acute shortage of labor, and the franchise owners have a difficult time finding enough employees. The owners have been forced to work long hours to keep the stores open 24/7 and to avoid penalty from their head offices. As the interest of improving working conditions has been growing in Japan, the customers are now conscious of the impact of the labor shortage crisis and show understanding of short operation hours. The major convenient stores are executing pilot studies on new logistics systems, investigating the impact and planning the strategy to pursue productivity.
In 2018, there were 153 cases of bankruptcy due to the shortage of labor and the number of cases increased by 44 percent compared to the previous year. Especially the population decline has a huge impact on the production, logistics and labor-intensive industries. The statistics of the Ministry of Health, Labor and Welfare (MHLW) in 2017 showed that 58 percent of Japanese workers feel high stress from work. Among those workers, 63 percent answered the cause of stress was the amount and quality of work.
The turnover rate for new university graduates in Japan to leave their work within 3 years is about 32 percent, and for the service industry, it is about 50 percent. In the situation of the current labor shortage, AI is considered to play a big role in improving the working environment and keeping personnel. In addition, elderly people are encouraged to continue working as long as they are healthy and can participate actively in the society. As diversity now is promoted and expected to increase in Japan, it is necessary to secure production processes in line with the diversity trend, and transaction from mass production to high-mix low-volume to fit the needs of society.
The new AIST Cyber-Physical research facility at Tokyo waterfront aims at addressing these societal challenges, and promotes cutting-edge R&D in bio and information technologies, such as drug discovery utilising robotics and computational biology, medical use of genomic data, AI research to create value from big-data, digitisation of human functions and behaviors, etc. Their activities will include international collaboration, public-private-partnership, technology transfer, fostering of young researchers, and other outreach activities.
Test bed for retail environment: Convenient store
The work for the employees at convenient stores in Japan includes displaying, cleaning, cooking, selling tickets, and parcel shipping, with high expectations to provide high quality service. A test bed for robotic support at convenient stores is part of the Cyber Physical System (CPS) facility to develop and introduce advanced technologies in retail. The goal of AI and robotics is to support the efficiency of work, and in the facility researchers from universities, institutes and companies can develop robots for stores, including i.e. sensor cameras in the ceiling and advanced data processing.
Creating a healthy environment for manufacturing
The centre also includes a research lab that focuses on human-robotic processing and assembling plants. Research on machine processes and assembling is important for manufacturing. For this, knowledge and experience of highly skilled workers are important assets, and by modelling these, data and AI can be used for the training purposes. The aim for this manufacturing research is to grasp the current situation and develop processes to enhance global competitiveness of industry, and at the same time provide the best environment for workers at the manufacturing sites. A strong field within AI and robotics in Japan is sensors, and at the CPS facility, dynamic data of workers is gathered and analysed, and optimization, simulation and deep learning technologies are applied for promoting the cyber-physical interactions and support. Consortia for R&D and businesses, including application and training, are also established for creating new industrial applications.
Bio robot MAHORO for drug discovery to improve accuracy and speed
The MAHORO is a laboratory work robot. In a bio-research environment, an AI robot can advance the drug discovery research productivity. With technology advancement in analysis, the capacity of analysing data has vastly increased. The lab droid MAHARO support automation of cell culture processes with hands, brain and eyes using robotics, AI and microscope. Cell information close to the molecular level is analysed by the highspeed sensors and AI. The robot can operate for 24 hours a day, and will be able to obtain accurate data without errors that often occur by humans in the complex and routine oriented laboratory work. With such support, researchers can instead concentrate on more advanced work that needs to be done by humans. By shortening the experiment period, it is expected to impact and promote drug discovery in pharmaceutical industry. Furthermore, the establishment of a bio research consortium in collaboration with government, academia and industry is planned, and AIST is aiming to advance the drug discovery.
Minimal fab – small-scale, high-added value Semiconductor Factory
The “minimal fab” is a series of small-scale semiconductor factory that replaces a traditional cleanroom. It is a production system that consists of a set of minimal equipment, a group of small manufacturing units, such as for wet process, exposure, etching, film formation, etc., with a streamlined, common design. Rather than a large wafer with a diameter of 300 mm, a wafer with a diameter of 12.5 mm is employed. The standard manufacturing process is one wafer flow, and the processing in each tool is carried out in minutes. For traditional semiconductor manufacturing, huge factories made up of expensive, large-scale manufacturing equipment and cleanrooms used to be necessary, with high cost for low-volume production. With minimal fabs, manufacturing costs are reduced to 1/100 (to 1/1000) of the traditional cost. AIST has three connected centers of the minimal fab, and at the Tokyo site, the custom-made device is developed as a connecting point to industry. There are also R&D centers in Tsukuba and Kyushu for executing pilot studies. The development and trial manufacturing of semiconductors and devices becomes radically simplified with the minimal fab.
The research institute AIST is focused on R&D for industrialisation, and contribute to bridge technology development with industry. The new Cyber Physical System facility of AIST at the waterfront in Tokyo constitutes a bridge to the future.