Edge Computing software from Mitsubishi Electric connects relevant business processes and helps companies that digital transformation implement. This increases productivity, improves product quality and optimizes system availability. In order to make modern production lines fit for the Industry 4.0 the automator now has the data analysis tool Melsoft Mailab presented which KI supported helps to digitize production.
07.02.2023 | The edge computing software Melsoft Mailab (Mitsubishi Electric AI Laboratory) supports companies in the digital transformation their production and promotes their increase in productivity. The data science tool is intuitive and operator-centric. The intelligent platform uses artificial intelligence (AI) to optimize automated processes. For example, through preventive maintenance, waste can be avoided, rejects reduced, downtime reduced and energy consumption reduced.
Future-proof, highly efficient and fast-reacting systems will only be possible if the software for data analysis large amounts of data also processed correspondingly quickly. However, companies often lack the resources to achieve this goal on the way to the smart factory. There is a lack of budget for qualified data analysts and AI specialists. And the ability to efficiently process the large amounts of data is often not available. There is also a lack of time for the development of precise models.
With the edge computing software Melsoft Mailab, Mitsubishi Electric now offers a solution to these problems. The intelligent Data analysis tool acts as a dedicated virtual AI data scientist. Mailab helps companies implement their future-oriented manufacturing strategies. Mailab is up and running quickly – with minimal training and prior knowledge. The edge computing software supports users step by step in creating models that are derived from historical data.
Various algorithms are used for this machine learning (ML) used. Data acquisition, the creation of forecast models and the analysis of large amounts of data can thus be automated. Melsoft Mailab breaks the barriers to getting started with Industry 4.0 applications that require advanced data analysis. Strategies to improve production with a quick return on investment (ROI) are also supported.
The installation of the edge computing software is user-friendly. Mailab can be installed locally on a PC or on a central high-performance PC/server. Access is possible from all Internet-enabled devices in the network also by multiple users possible at the same time.
The intuitive user interface helps users to analyze the data and supports them in all phases of a data analysis project. The data records to be processed can be visualized in various ways. Analysis models can be created based on the targets selected by the users. The processes within the AI analysis tool use the Corn type AI by Mitsubishi Electric (Mitsubishi Electric's AI creates the State-of-the Art in technology)
Mitsubishi Electric developed Melsoft Mailab software to support a wide range of different applications – also specially tailored. The software can be used offline so that existing data can be used to generate or optimize predictive models, and adjusted with Python scripts if necessary. The generated prediction models then allow a real-time diagnosis. The data generated during plant operation is made available to these models and provides insights into plant health, plant performance and optimization.
In addition, new information can flow in to Accuracy of the prediction models and continuously improve results. In this way, companies can increase their productivity step by step.
15.12.2020 | The digital transformation of industry has connection of OT | operational level and IT | superior information technology gave new impetus to the corporate level. In order to be able to fully exploit the potential of digitization, however, more than just a network connection between these two levels is required. Edge computing from Mitsubishi Electric offers the solution for this. Relevant business processes can be connected via the intermediate level with intelligence.
The digital transformation should create a basis for increasing productivity, increasing product quality, optimizing plant availability and maximizing plant utilization. The goal of these improvements is to serve customer demand, ideally in real time. The basis for this is increasingly being formed networkable components at the operational level, which will become part of the IIoT or industrial IoT or Internet of Things.
However, data is data. Why shouldn't the direct vertical connection between production and higher-level business systems be enough to achieve the goal of digital transformation? The answer is found in the way how the platforms of IT and OT have developed. They handle different types of data and data processing requirements differently. OT today processes data in real time with process speeds of seconds or less. IT, on the other hand, processes this with much longer sampling times from minutes to hours or longer.
Behind the model of Industry 4.0 there is a need for the OT and IT rooms to use the data that the other can provide. But they are not designed to translate and interpret the data received from the other in a manner or in a timeframe that can have a direct impact on the plant's operations in real time. As the digital transformation progresses, the result of direct IT / OT integration is often simply the delivery of large amounts of unfiltered data instead of the relevant information required for real operational improvements.
Some have argued that perhaps the cloud could provide an environment for managing this large amount of data by providing a platform for efficient data aggregation, filtering, and analysis. And certainly protocols like OPC UA a direct connection from production via the higher-level systems to Cloud manufacture.
But although the cloud offers an ideal platform for developing knowledge about plant operations, it is not the right platformto put knowledge into practice for production operations. While it can provide useful and essential data analysis capabilities, it does not provide the real-time aggregation and analysis needed at the OT/IT bridge.
A more suitable solution for the OT / IT integration offer the new intelligent edge computing technologies. These technologies form an intermediate layer between the shop floor and the higher-level business systems. They offer a simple interface between IT and the OT world as well as new options for the place where the data analysis takes place.
The edge computing solution Melipc from Mitsubishi Electric offers this functionality and is OPC UA compatible. Data can be pre-processed and aggregated locally to produce valuable information for the systems that require and process it. Edge computing also seamlessly connects the production area with higher-level IT systems such as MES and ERP platforms.
By performing sophisticated data analysis Real time through edge computing, the increasing use of AI | Artificial intelligence Algorithms and machine learning for a more intelligent handling of data can improve the Melipc's efficiency in production. The costs of data processing can also be reduced considerably, since only necessary and relevant information is passed on from one company level to another.
The Melipc Edge Computing offers one Real-time data collection and processing to a robust industrial standard. From a data processing point of view, it includes a number of analysis tools, including multiple regression analysis, the Mahalanobis Taguchi system and statistical process control (SPC) as well as AI functionalities. This includes, for example, the detection of similar waveforms that provide feedback to the factory floor in real time.
As a result, Edge Computing can perform functions of data acquisition, filtering, processing and analysis using operational know-how with diagnosis and feedback for applications such as the predictive maintenance combine. This takes place within a real-time flow of information that can influence the decisions of the production systems.
Edge computing provides a basis for the digital transformation of the company and offers a platform for connecting machines and devices. So can Manufacturing processes react faster and smarter to changes in production. This happens regardless of whether production is based on the plant, demand or supply.
The increasing amounts of data from the operational level and their requirement to use this data more intelligently, enable the digital transformation of industry. At the same time, they also represent a challenge. Edge computing as an intermediate layer between IT and OT solves this challenge and paves the way for the event-controlled architecture that defines Industry 4.0.
This gives production a key with which it can be converted into an intelligent operation. So edge computing builds that natural bridge between OT and IT in a format that both sides can bridge the gap.
02.12.2019 | Companies that aim to connect the operational technology (OT) of their production environment to their IT systems now have a new option from Mitsubishi Electric: the edge computing solution Melipc opens up optimization potential through preventive condition monitoring (predictive maintenance) or quality assurance with real-time data evaluation and immediate feedback to the operator.
The Melipc solution enables quality and productivity increases in Industries again Automotive industry, food technology and beverage industry or life sciences. Here the real-time process software provides local diagnostics, predictive feedback and visualizations for the analysis of the production environment.
Analysis algorithms for predictive maintenance and quality assurance are already integrated in the Melipc series. In continuous production processes, this solution can optimize processes and adapt them to variables. Efficiency is increased by creating a real-time prediction model improved, which can be updated and refined based on new requirements from the process. In this way, errors in packaging machines in the food industry, for example, can be detected, better still prevented and thus quality improved.Melipc uses artificial intelligence (AI) to identify Machine status anomalies Real time. Detailed status information is reported back to the automation level so that machine operators can proactively carry out adjustment or maintenance measures at an early stage.
In addition, the decentralized edge computing Benefits in terms of data security and process integrity. Mitsubishi Electric's Melipc solution is also used in life science applications, where it stores and analyzes sensitive data at the OT level instead of sending it to the cloud. This accelerates the diagnosis to real-time speed and reduces costs and requirements of the IT infrastructure.
The new Melipc solution complements the existing portfolio from Mitsubishi Electric on edge computing modules such as Maps Scada, data loggers and controllers.
The authors are Christian Nomine, Strategic Product Manager Visualization (l.) and Jonas Roski, Junior Product Manager HMI - MPLC, both Mitsubishi Electric Europe BV, Ratingen