|
|
|
Machines form the basis of a successful manufacturing company. Companies can evaluate how effectively the equipment operates based on various parameters. In the context of Industry 4.0, digitalization, and the Industrial Internet of Things (IIoT), MDE, or Machine Data Acquisition (English: Equipment Data Acquisition, EDA), primarily provides important insights into production data.
What is Machine Data Acquisition (MDE)?
MDE serves as the interface between machines, production technology, and data processing:
Digital machines capture and store production-relevant data, share this data with each other if necessary, and evaluate it in real-time.
The information obtained can either be visualized using specialized software, for example, in a clear production dashboard. Additionally, it can also be used for machine control, production planning, or process control. This contributes, among other things, to making production processes more transparent, tapping previously potentially unused capacities, identifying potential bottlenecks or disturbances early, or even avoiding them altogether, thereby increasing productivity.
Depending on the plant and the company, the machine data acquisition occurs automatically via sensors that can directly transmit the measured parameters to the corresponding software for further processing.
What Machine Data Is Available in Manufacturing?
The captured machine data generally includes all information that can arise in a production facility and can be captured and made available by programmable logic controllers (PLC) or sensors.
This data can then be used or aggregated for various applications. For example, the throughput of the system can be monitored. In such cases, these are referred to as performance data. Additionally, it is also possible to integrate throughput data and thus monitor the material consumption of the plant — this falls into the category of consumption data.
Machine data includes, for example, the following information:
Operating data (for example, start and stop times)
Performance data (for example, speed, pressure, or temperature)
Condition data (for example, wear or vibrations)
Maintenance data
Error and disturbance data
Energy management data (for example, energy efficiency and energy consumption)
Production data (including quantities, throughput rates)
Sensor and actuator data (raw data collected during monitoring and control of the machine)
Environmental data (such as temperature, humidity, air pressure)
Operating parameters (for example, machine settings)
Which machine data is relevant and should be captured, processed, and possibly combined ultimately depends on the respective application case.
Machine Data Acquisition vs. Operating Data Acquisition
Closely related to machine data acquisition is so-called Operating Data Acquisition (BDE). However, the two terms do not mean the same thing: While MDE focuses on the information that machines provide about the production process and the products in manufacturing, the scope of BDE is broader.
The operating data of a company includes both organizational data, such as orders and personnel, as well as technical information about machines, materials, and tools. Thus, machine data acquisition provides a part of the operating data and flows into the operating data acquisition of the entire company.
Why MDE is Consistently Underestimated in Manufacturing and Projects Fail
One of the main reasons why MDE projects backfire is that many do not understand what all is involved in machine data acquisition.
Because machine data acquisition is often confused with the mere connectivity of machines. However, connectivity only translates the "language" of the machine into an open protocol (e.g., OPC UA), allowing data to be read.
To use the data successfully (the desired end result), four components are necessary that are often forgotten or underestimated in their complexity:
Connectivity: From a proprietary protocol to an open protocol.
Data preparation: Making usable data from confusing variables (correct units, standardized naming, etc.)
Data storage: Making prepared data accessible long-term, safely, and centrally.
Data usage (e.g., analysis tool): Using data to optimize manufacturing.
We illuminate each of these components in our article on MDE software in detail and explain how you can successfully implement machine data acquisition.
4 Examples of Using Machine Data with the ENLYZE App
Machine data captured and provided through MDE can be utilized in many different areas and for various purposes in data-based manufacturing. Essentially, MDE, like many other production analyses, aims to optimize manufacturing and its processes.
A useful tool is, for example, a central production dashboard into which all captured machine data flows. This provides workers with real-time information about which jobs are currently running on which machines and whether they are being produced according to specifications.
To take advantage of the benefits of digital MDE, we at ENLYZE digitize all relevant machine data, link it with the order data from MES and ERP, and consolidate everything in the ENLYZE Manufacturing Data Platform™️. This results in various ways to use machine data from manufacturing. Here are some examples:
Example 1: OEE Management
OEE stands for Overall Equipment Effectiveness, which in German means Gesamtanlageneffektivität. This metric measures the productivity of industrial facilities and machines in manufacturing. The OEE takes into account the availability of the plant as well as its performance and the quality of the produced parts and relates them to each other. OEE is considered the gold standard for manufacturing companies to evaluate the effectiveness of their shop floor.
The ENLYZE platform also offers the opportunity to keep an eye on and control the OEE of the machines through an OEE dashboard. MDE, in turn, enables the accurate and automated recording of OEE along with the individual losses based on machine data and visualizing it over time. This enables the identification of potential for improvement based on key figures, deriving appropriate measures, and increasing overall equipment effectiveness.
How to calculate the OEE of your machines is explained in this article. Information on what to consider when selecting the appropriate OEE software can be found in this article.
Example 2: Production Monitoring
Using MDE opens up far-reaching possibilities for monitoring one’s own production. All machine data from manufacturing, as well as information from ERP and MES, flow together in the ENLYZE platform, where they can be analyzed. This information can also be visually represented in various dashboards. This way, companies maintain an overview of the current status of all machines in manufacturing, based on real-time data.
Example 3: Traceability
Even on the best shop floors, there are always process variances and production errors. Being able to trace these later is important to avoid them in the future. However, this is only possible with the corresponding process and product data from MDE. Digitally captured and comprehensively recorded machine data enable continuous traceability in case of quality deficiencies. This not only facilitates troubleshooting in case of complaints but also identifies points for improvement measures.
Example 4: Calculation of the Product Carbon Footprint (PCF)
In the context of sustainable manufacturing, the CO2 balance of a product (English: Product Carbon Footprint, PCF) is an important metric. The calculation depends on the raw materials used and the technical settings on the production line. Since these can vary significantly depending on the product and order, the PCF must always be calculated on an order-specific basis. Here, machine data, as well as energy and order data, are included to determine CO2 emissions very accurately for each order and thus each product.
How this works exactly is explained in our webinar with Julius Scheuber, founder and product manager at ENLYZE, and Dr. Karsten Riest, Sustainability Manager at KAP AG.
You can watch the recording of the webinar here.
Machine Data Acquisition with ENLYZE: Use Data from Day 1 Without an IT Project
Today, at least four different products must be married for the use of machine data. This integration costs time and money.
We bypass the problem at ENLYZE by bundling the four products into one product. This way, you have no lengthy IT project with high costs, but a product that works from day one.
Our connectivity solution enables heterogeneous machine parks to be digitized quickly and easily.
Our data platform ensures that data is efficiently collected, machine, order, and product information is aggregated, and relevant metrics are accurately and correctly calculated.
Our analysis software provides turnkey applications for typical challenges (OEE management, traceability, etc.) on the shop floor to deliver value and ROI faster.
To avoid becoming a new data silo, all this data can be exported and shared with other systems via modern interfaces.
In short: We empower you with your data instead of working for your data so that you can focus on creating value.
Reading Machine Data
What Hardware Is Available for Machine Data Acquisition?
Today, machine data is usually measured by sensors and collected via a central PLC. This PLC uses this data to control and regulate the machine.
The data is then integrated into modern factories through interfaces into SCADA or DCS systems, extending to MES and ERP systems, to better inform decisions in manufacturing and business planning. Through corresponding software, insights into quality assurance and machine maintenance can also be derived from this data.
The connection of this data is usually done through standardized interfaces (more on this under “Exchange of machine data from the PLC to other systems”). If these are lacking, the connection becomes more complex.
Here you can find out how we at ENLYZE read your machine data using Sparks.
What Software Is Available for Machine Data Acquisition?
There are specialized software programs available for machine data acquisition. The devices used usually have manufacturer-specific programs, such as Simatic S7 for Siemens PLC hardware. In addition, there are several providers (including ENLYZE) that enable the capturing and integration of machine data for a variety of PLCs.
Exchange of Machine Data from the PLC to Other Systems
Usually, these collected data should not only stay at the machine. The machine data can enable and support analyses of the processes in other business areas, e.g., in
SCADA or DCS systems to simplify process management
MES systems to better plan production
in ERP systems to display machine availability
to inform warehouse logistics about quantities of raw materials and end products
to enable process analyses for quality assurance
to enable predictive machine maintenance
The transmission of machine data from the PLC to ERP, MES, and other planning systems occurs through interfaces. Common interfaces include standards like the OPC standard (“Unified Architecture” OPC-UA, “Distributed Architecture” OPC-DA) or the Ethernet standard “Profinet”.
Most providers of digitalization handle these standards very well. The connection of very old machines, which often lack standardized or even any interfaces, becomes more difficult. Here, the connection possibilities are examined on a case-by-case basis, and special solutions are developed.
Process of Automatic Machine Data Acquisition
The process of automatic machine data acquisition is simple, provided they are modern systems. Most of these have open interfaces for connecting to all ERP, MES, and other planning software. Data is automatically captured by the machines and transmitted via a central system (e.g., cloud systems) or directly to the recipients, which is, however, significantly more complex.
If it is a historically grown machine park, reading machine data presents an even greater challenge. Since these machines were originally only meant to be connected to manufacturer-specific systems, they often do not have open interfaces for integration into modern external planning systems.
The connection of these machine parks is then done through special software that reads the data from the control units and prepares it for analyses of productivity and other optimizations and integration into MES and ERP systems.
Advantages of Automatic Machine Data Acquisition for Companies
Traditional data collection, e.g., with analog, handwritten protocols, is prone to measurement gaps. These are usually filled out directly by the worker or by the production manager. Due to illness or unforeseen obstacles in daily life, this can be forgotten.
In contrast, automatic machine data acquisition reliably and continuously collects data about production, making it easier to analyze problems and causes. In addition, human measurement errors are avoided, which arise from manual recording by workers or quality inspectors.
The data is also collected automatically faster and without delay and directly input into the needed systems. Thus, automatic machine data acquisition is a very efficient method for monitoring production and planning continuous improvements.
Advantages and Disadvantages of Automated MDE:
What About Machine Data Acquisition and Data Protection?
Machine data acquisition is a sensitive topic, as many companies have concerns regarding data protection. These concerns are entirely justified, as any form of machine data acquisition makes their facility vulnerable.
At the same time, this risk must be weighed against the ever-increasing competitive pressure and the rapidly advancing digitalization and automation. It's important to have a well-thought-out digitalization strategy with careful consideration of reasons, expectations, necessary scope, and responsibilities for digitalization involving strong partners.
Many cloud providers, for example, contrary to common belief, have the highest IT security due to their years of experience and enormous technical and financial resources. The greatest risk for data protection lies with end users who have not been sufficiently trained or otherwise exhibit risky behavior. A clear strategy for the internal handling of data protection is crucial here.
