- Integration of the data from many lines and plants.
- Statistics of downtime and predictions of lines failures.
- Comparison of the teams and changes.
- Operational and analytical reporting
Nowadays, production on a large scale is usually automated to a very large extent. Machines and production lines operation is driven by the industrial drivers, and the role of an employee is only limited to supervision over and additional operations (setting-up, supplement of materials, reaction to problems, elimination of faults). By driving the machine operation, the drivers collect many up-up-date important parameters describing the production process, which can be provided to other applications.
The information available in the drivers’ memory can provide valuable knowledge of the production process, its quality, performance, efficiency, downtime (including the shortest ones). In certain plants, such data are analysed to some extent, in others they are not analysed at all.
The process data from one machine include the information on the narrow part of the production course. The information collected from all machines in one places allows to analyse the whole process. An example can be the analysis of the maximum electricity demand by the entire plant to the nearest quarter. If there are several machines in the plant, execution of this analysis without proper integration of the data is extremely time consuming.
The process data collection for a long period of time allows to track the slow changes in the production process. Some of them can affect production quality or reliability. Analysis of these data can potentially predict the imminent failure.
Example: The refrigerant pump in the power plant is driven by an electric motor. The engine is used to measure the temperature of the bearings and its supply line has an independent electricity meter. These data are currently transmitted to the OPC server and transmitted for further analyses. Observation of the changes of the maximum temperature of the bearings and the average daily power shows slow but consistent growth of both values, approx. 1-2% per month. This may suggest commencing of the bearings wearing process resulting in increased temperature and increased current consumption. It is better not to wait until the failure occurs, but plan the pump service at a convenient time.
Integration of the process data of the machines within the entire plant allows to gain valuable knowledge. There can also occur the need to analyse the data in the entire plant, which has many geographically dispersed locations. In this situation, a good solution is to collect the necessary data in one place, to which the authorized persons will have access.
Collection of the process data from many locations allows to execute the comparative analyses covering all plants together. It can turn out that although having the same equipment conditions, in one of the plants, the production capacity on Mondays till 12:00 is by 15% lower than in other plants. It is worth to check why this is happening. In other plants, the number of products, which do not comply with the quality standards, regularly increases in the last hours of operation. The number of possible to execute summaries is limited only by the number of the available data (variable) and… our imagination.
Collection and analysis of the industrial data can be entrusted to the system. This is a dynamically developed platform, the purpose of which is to provide rich analytical and visualization possibilities for the data collected from the production lines.
The proposed solution architecture provides the possibility to develop the system to meet the growing needs of the customer. In the initial phase, the data can be transferred from one machine only (e.g. on the solution piloting stage). Adding the additional data sources can be realised for example by adding their support to the existing OPC server. In addition, one can also start another OPC server connected to the Gateway module to communicate with the infrastructure of the supplier. Another possibility is to start another Gateway module with an independent data source (e.g. independent OPC server for the machines in another location). The system architecture allows to transfer the data from any number of the data sources. The data within the individual Gateway modules are grouped in the logical channels, configured independently. One channel can correspond to one machine or a dedicated sub-group of the data from many machines. This depends only on the accepted design assumptions and the reporting needs.
WHERE DO WE COLLECT THE DATA?
The solution, which is very well suited to the above scenario, is an external server located in the so-called cloud. The data are transmitted to the external server of the complete solution supplier. On the server, it is possible to control the status of the data collection process (observation of status of the modules installed at the customer, status of the process supplying the reporting system, configuration changes). In the same way we can access the finished reports based on the data provided form the plants, and which is important, access is possible almost from anywhere.
WHY A CLOUD?
This is the solution, in which the customer does not have to bear the excessive costs. Now there are more and more suppliers, which, at a reasonable subscription fee, offer the opportunity to use all or a part of the server for own use. For this reason, the customer does not need to maintain the expensive equipment by itself, take care of safety and backups, and only uses the infrastructure offered by the external supplier. Within implementation of the system, only the modules collecting the process data and transmitting them to the external service are installed. Further processing is already on the external server, and the customer has access to the finished reports through a web browser or application on the phone/tablet.
I DO NOT WANT MY DATA PRODUCTION DATA TO LEAVE THE PLANT, THE SOLUTION BASED ON A CLOUD IS NOT NEEDED. ARE THERE ANY ALTERNATIVES?
If, for a variety of reasons, the customer does not want to transfer the data to a cloud, then the server collecting and processing the data can be placed in the infrastructure of the customer. Such a solution has the same architecture as the solution based on a cloud, with the difference that the physical data are not transmitted outside the internal network of the plant.
It should be noted that in such an approach the customer is obliged to provide the data processing server with support (service, backup).