Experience the demonstration as an opportunity to think about deploying solutions at your company - "NEC DX Factory" Report 2

Experience the demonstration as an opportunity to think about deploying solutions at your company - "NEC DX Factory" Report 2

NEC Corporation (hereafter, NEC) opened the co-creative experience space "NEC DX Factory" in 2018. IoTNEWS visited in 2018 when it opened and reported on the experience.

This time, we covered the new "NEC DX Factory", including the newly added demonstrations and the new local 5G lab attached to the NEC DX Factory.

This article introduces various solutions that can be seen at the NEC DX Factory. By actually observing the demonstration at the NEC DX Factory, I was able to deepen my understanding of the characteristics and use cases of the solution, and I felt that it would be easier to imagine how it would actually be installed in the factory.

Contents

NEC multi-robot controller

When considering the introduction of AGV, which is a transfer robot, in order to realize high-mix, variable-volume production, it is difficult to introduce and operate the robot. There is a problem that it is difficult to control vehicle allocation according to the situation.

The NEC multi-robot controller is a solution that solves the problems of AGV for high-mix, variable-volume production. It has three major features.

First, optimal route generation. When you set a start point and a goal point, the route between them is automatically generated by an algorithm. It is possible to set a route according to the purpose, such as the shortest route or the route with the lowest cost. Since there is no need to set the route and teach one by one, the hurdles to introduce it will be low.

The second is centralized control. Multiple AGVs from multiple vendors can be controlled. Currently, we have partnered with AGVs of Topy Industries and Nidec-Shimpo, and can control the AGVs of the two companies.

The third is external system linkage. Since we provide an API for upper systems, it is possible to link with business applications such as MES and WMS.

The NEC multi-robot controller has a management screen that allows you to grasp the situation at the site at any time. On the management screen, each AGV is color-coded, each route is automatically generated, and it is possible to grasp which AGV is where at any time, and to control it sequentially wirelessly to the goal.

On the management screen, it is also possible to flexibly set driving rules such as temporary stops and prohibited areas.

Next Generation Factory Security

Security is said to be of greatest interest to companies visiting the NEC DX Factory.

Interest in data ownership and security is increasing due to the influence of collaboration between companies, but various incidents and accidents are occurring in the field of OT both in Japan and overseas.

In addition to malicious attacks, even if there is no malicious intent such as an employee connecting a smartphone to the equipment and being infected with a virus from there, the production line may stop.

Because it is also targeted from the outside, security measures are very important. In some cases, OEMs are also the target of attacks, so even small and medium-sized companies are being forced to take countermeasures.

What NEC is proposing is support from risk analysis to operation establishment. When introducing new systems and equipment, there are many opportunities to consider security measures, but it is necessary to visualize the current status of existing assets and equipment and consider what measures to take from the planning stage.

Process Design Simulator

In the environment surrounding the manufacturing industry, uncertainty is increasing. In particular, demand fluctuations are becoming more intense, and in order to seize opportunities and increase sales, it is necessary to quickly start up production lines.

There are some issues to realize the rapid start-up of the production line.

First of all, there is a problem in the work to prepare for production among the work to start up the production line. The main causes of the problem are that the load on the production site is very high because analog work is the main work, and that production preparation cannot be started until the design is completed. In particular, it is common for design to fail to meet delivery deadlines, such as when dealing with quality issues. In the end, work becomes busy in order to absorb design delays and meet delivery dates at the production site.

On the other hand, there is also the problem that the request from the production site to change the design to improve productivity does not reach the design stage. It is necessary to provide such feedback without omission.

In order to do that, it is necessary to digitally link design and production. In order to enable efficient product development, it is important to connect the supply chain and engineering chain, including production preparation, with BOP.

Solution by experiencing the demonstration

NEC is trying to digitally link design and production using a tool called a process design simulator.

The process design simulator is a tool for designing and improving the line through digital simulation when starting up the production line, and finally feeding back the best results to the actual line. By incorporating BOP information and equipment parameter data into the simulator, the accuracy of the simulation can be improved.

In fact, when the demand for home routers increased due to the impact of the new coronavirus infection, NEC asked if it was possible to use a process design simulator to produce home routers at a factory that produces other products. was considered. In the past, NEC has been greatly affected by floods in Thailand and natural disasters in Japan. Based on that experience, we have standardized QMS, systems, equipment, etc., so that production transfers and factory transfers can be done smoothly. It is said that a good foundation has been established.

In the background of the Japanese manufacturing industry, each factory has been improving productivity and making individual ingenuity so far, and the factories in the same company have been competing with each other to increase productivity. However, this background does not go well with management that is conscious of resilience. This is because it is not possible to spread the results of considerations made with the process design simulator to all factories at once, or to horizontally deploy improvements made in other factories.

By standardizing the parts that can be standardized, in the event of an emergency such as the inability to cooperate with overseas factories such as the new coronavirus infection, we will use the process design simulator to produce at a place where we can produce, I felt that it would be possible to prevent loss.

Video Management System

In the use of videos in the factory, the idea is to set up a camera and take a picture, and if a problem occurs, look back at the video and check the problem. be done. However, it takes a lot of time and effort to search for a specific part of a video that has been recorded for 24 hours.

Therefore, the video management system is a solution to simplify the search.

In the demonstration, the state of the production line was filmed, and using the motion capture function, when the patrol light turned on abnormally, the system could search for what happened there and play the video. It is possible.

By motion capture, if motion is detected, it will be displayed in the list, so you can check the movement such as unlocking the door or entering the restricted area.

In addition, it is linked with face recognition, and if you register your face in advance, you can display the image of each person. You can later check what a particular person was working on.

AI video work analysis

When assembly work is done manually, mistakes may occur. If something goes to the next process without making a mistake, there is a possibility that it will affect the next process and develop into a market problem.

AI video work analysis is a solution that AI detects work mistakes using video.

You can take a picture of the assembly work with a camera and check the real-time video on the system. If you do something wrong, an alarm will go off to let you know what you did wrong. By stopping the line at this point, it is possible to prevent the product from being sent to the post-process. In the case of correct work, the work analysis result is derived from the image of the time spent performing the work.

The correct work is filmed in advance, a model is created based on the image, and a judgment is made.

Autonomous Robotics

At NEC, we are thinking about cooperation between humans and robots. It is an image of replacing one person with a robot in the work that has been done by two people so far, and the person and the robot work together. We are considering whether workability can be improved by having a robot support heavy objects and work with both hands, or by having a robot carry parts to the hand.

In the demonstration, a depth camera is installed, the position of the hand is measured by the camera, and the trajectory of the robot to that position is calculated to carry the part. Workers wear gloves with AR markers printed on them, and the position of their hands can be detected by the AR markers.

The gloves have AR markers printed on the back of the hand and the palm of the hand, respectively, and when the worker faces the camera, the robot can be instructed to stop. Afterwards, when the back of the hand is turned to the camera, the robot is programmed to bring the next part.

Currently, we are using AR markers to give instructions to robots, but we are considering using other interfaces such as finger gestures, and a world where humans and robots work together while coexisting. is being considered.

Remote control of robots

If a robot on a production line breaks down or stops working in a factory, currently a production engineer comes close to the robot and uses a teaching pendant to deal with it. This method has become a common business process.

However, NEC believes that such work should be done remotely using wireless communication in the future. In an unmanned factory with no workers on site, if a problem occurs, the business process will be to remotely control the robot over the network. It will be necessary to check the image of the camera to confirm the situation of the trouble.

In the demonstration, two cameras are installed on the top of the device, one connected to Wi-Fi and the other connected to 5G so that you can experience the difference.

The video captured by the camera can be checked on the system. The operator uses a teaching pendant to operate remotely, but the pendant contains software that can be run on a PC and can be controlled over a network.

Remote operation support system

The robot remote control solution assumes an unmanned factory. Timing is assumed.

In particular, skilled workers will retire more and more, and the number of skilled production engineers will decrease, and it will become impossible to assign personnel to each factory.

In such a situation, NEC assumes that the business process will become such that skilled workers will remotely support multiple factories.

The remote work support system is a solution for supporting work from a remote location. On-site workers use devices equipped with cameras such as smartphones, tablet terminals, smart glasses, etc., and remote experts can see what is happening on-site and ask for support. .

A remote expert will be able to use a pointer on the screen to give advice such as whether the cable wiring is bad.

By using 5G, robot remote control and remote business support systems will be able to send even higher definition images without delay. It is expected that it will be possible to perform work and support in the same manner as when there are workers on site. However, in reality, there may be use cases where the current communication is sufficient for work and support. It is necessary to consider what kind of video is necessary in what kind of situation.

Related article

Shunsuke Obata

Worked at a manufacturer after graduating from university. As a production engineer, he is involved in the launch of new lines and kaizen work within the factory. Joined the company in July 2019 and is exploring the possibilities of IoT with a focus on the manufacturing industry.