Unlocking the full potential of on-site workers through automation of repetitive, low-skill tasks
We use ultra-high-mix, low-volume production at the OMRON Ayabe Factory to manufacture approximately 20,000 items, ranging from sensors a few centimeters in size to inspection devices that are several meters large. At this assembly factory, high-quality products are efficiently produced by finely coordinating product design and process design activities. Processes within the production line that do not generate added value, such as transportation of parts, are automated to lessen the burden for on-site workers. Data obtained from the production floor is analyzed to visualize improvement points by uncovering procedural factors that affect production efficiency and quality. Tasks that traditionally required the skill and knowledge of veteran workers have been transformed through automation.
Automating parts transport with mobile robots to allow human workers to focus on product value improvement
Streamlining high-mix low-volume cell production requires transport that can deliver the right parts to the right place at the right time. Critical though it may be, this is not something that can directly improve product value. At OMRON Ayabe Factory, autonomous mobile robots have been implemented to automate the transport process. In this system, humans and robots work together to enable safe transport that can flexibly adapt to changes in floor layout. This allows on-site workers to focus on tasks with high added value, which in turn has led to reductions in lead time and work-in-process inventory.
Enabling optimally timed maintenance through predictive management of optical element alignment
Lens positioning is a crucial process in fiber sensor production, with a decisive effect on light emission capabilities. At the Ayabe Factory, image data, waveforms and other process data are collected using IoT devices to create big data, which is later analyzed for automatic position correction to deliver consistent quality. AI controllers enable predictive maintenance, allowing advance action to be taken based on signs of failure and defects detected at assembly facilities. By managing equipment based on their condition instead of their length of operation (as in scheduled maintenance), maintenance can now be conducted on an as-needed basis, minimizing production line downtime and allowing for timely maintenance.
Detecting the slightest shifts in robot hand position to reduce facility downtime and part loss
Intermittent downtime caused by errors due to equipment in autonomous operation can compromise utilization rate, quality, and stability. At the Ayabe Factory’s automated production lines, displacement sensors measure the slightest shifts in robot hand tip position. By monitoring changes in feature values, equipment anomalies can be detected and addressed before failures actually occur, reducing line downtime as well as part loss.
Enabling low-cost automation of inter-process material handling with compact autonomous transport robots
Material handling is a good example of a process which, while it adds very little value, is difficult to automate. At the Ayabe Factory, we use compact autonomous transport robots for flexible inter-process transport, automating material handling at a low cost. Designed to enable flexible and easy transport route modification, this system effectively cut transport facility costs by 50% and reduced the time required for its own design/construction by 20% by automating material handling in facilities where workpieces are processed individually and by using commercially available plastic rails.
Robotizing complex soldering into skill-free operation without compromising precision or speed
At the Ayabe Factory, soldering tasks that previously required the "sleight of hand" and "sixth sense" of skilled veteran workers have been automated by employing robots tuned using high-speed camera footage of these workers. By converting tacit knowledge confined to specific individuals into explicit knowledge that can be put to systematic use, we created a system that delivers a high yield of over 99.97% with no training necessary, as opposed to the 60 days required to train a human worker.