技术
- 分析与建模 - 机器学习
适用功能
- 离散制造
用例
- 自动化制造系统
服务
- 软件设计与工程服务
挑战
由于日本等老龄化社会劳动力的减少,确保足够的人力资源变得越来越困难,这反过来又增加了对能够支持高效机械化操作的人工智能的需求。然而,由于预先学习和实际车间环境的差异,新的生产设施带来了特殊的挑战,导致在全面实施人工智能之前需要花费大量时间来教授人工智能。
解决方案
三菱电机的新 AI 在短暂学习后自动为实际环境创建控制程序,以支持优化操作。
三菱电机开发了一种人工智能 (AI) 技术,能够使用模拟器快速逐步学习,从而在相对较短的时间内有效地完成运动学习。新技术结合了公司最新的专有 Maisart 紧凑型 AI 技术和强化学习,使机器能够通过高效的试错法探索最佳行动。在公司智能控制人工智能技术的支持下,新的人工智能技术能够快速灵活地适应不断变化的条件,实时学习和响应实际环境的变化,使机器运行顺畅。
运营影响
Case Study missing?
Start adding your own!
Register with your work email and create a new case study profile for your business.
相关案例.
Case Study
Plastic Spoons Case study: Injection Moulding
In order to meet customer expectations by supplying a wide variety of packaging units, from 36 to 1000 spoons per package, a new production and packaging line needed to be built. DeSter wanted to achieve higher production capacity, lower cycle time and a high degree of operator friendliness with this new production line.
Case Study
Robot Saves Money and Time for US Custom Molding Company
Injection Technology (Itech) is a custom molder for a variety of clients that require precision plastic parts for such products as electric meter covers, dental appliance cases and spools. With 95 employees operating 23 molding machines in a 30,000 square foot plant, Itech wanted to reduce man hours and increase efficiency.
Case Study
Fully Automated Visual Inspection System
Tofflon has developed a fully automatic machine that uses light to inspect vials, medicine bottles, or infusion containers for glass fragments, aluminum particles, rubber grains, hairs, fibers, or other contaminants. It also detects damaged containers with cracks or inclusions (microscopic imperfections), automatically removing faulty or contaminated products. In order to cover all production processes for freeze-dried pharmaceuticals, Tofflon needed to create an open, consistent, and module-based automation concept.
Case Study
SAP Leonardo Enabling Rocket Science
At times, ULA has as many as 15 different operating systems dedicated to overlapping processes, such as rocket design, testing, and launch. Multiple systems created unnecessary costs and unwanted confusion among workers at offices, factories, and launch sites in different location. In order to improve collaboration and transparency during vital activities that directly influence mission success, ULA wanted to improve data sharing and streamline manufacturing processes.
Case Study
IIC Smart Manufacturing Connectivity for Brown-field Sensors
The discrete manufacturing domain is characterized by a strictly hierarchical structure of the automation systems, commonly referred to as the automation pyramid. Data acquired by a sensor typically flows through an IO-module into a Programmable Logic Controller (PLC) which manages the local real-time control system. As all process data are concentrated in the PLC, re-programming the PLC and thus, implementing interfaces to access these data appear to be the natural choice to transfer them to the IT system. However, for brownfield installations this choice has proven impracticable for the following two reasons:In brownfield facilities, PLC usually operate within a once-specified environment and are rarely re-programmed. That is why the active staff is often not familiar with the code and lacks of the competence to modify the existing implementation in a reasonable amount of time.Furthermore, for cost reasons, any PLC was selected to exactly match the requirements of the environment within which it was intended to operate. That is why it cannot be assumed that a PLC will be able to support additional tasks such as communicating data through additional interfaces.
Case Study
Human–Robot Control
Industry 4.0 is changing the way manufacturing industry operates. Increasingly more manufacturers are leveraging advanced technologies such as robotics and automation systems to improve productivity and efficiency. As a result, human–machine interfaces (HMIs) are becoming more important in their role in the digital connectedness of humans and machines. However, using the wrong HMI can lengthen development times and increase implementation costs.