Data validity as the basis for determining key figures in digitally networked production plants is indispensable. The path to success leads through the highest possible data quality.
After the production of new machines and plants, it makes sense to check whether all relevant machine sensor data is available from the machine manufacturer (OEM) and whether the data is valid. This aspect is often neglected when purchasing new machines. However, valid machine data is essential for the determination of key figures in order to make far-reaching company decisions.
Specifications, construction criteria and defined standards exist for the construction of machines and plants. However, suitable validation options for the implementation of machine sensor data are lacking for the verification of such standards, both internally and externally. Much emphasis is placed on the digitization and modernization of production plants, but ultimately the effort is of little use if the database is incomplete.
In the beverage bottling industry, the customer is usually guided by the Weihenstephan standard. He expects that the data modules with their addresses have been correctly stored after completion of the machine.
Although the OEM checks the sensor data before the machine is delivered, the values are rarely all valid. As a result, machine sensor data does not provide correct and complete data after commissioning.
A standardized process plan of the factory acceptance test before delivery of new machines and systems should be the goal of every production operation. This provides certainty that valid data will arrive right from the start.
Quantis has already successfully conducted a factory acceptance test for a customer at an OEM's site. The customer was so convinced by the procedure as well as the results that a process concept for data validation was developed. Quantis supports its customers both in the execution of the test and in an advisory capacity regarding the validation results.
The Factory Acceptance Test can be performed at the OEM's site, but it is not a must. The customer decides what exactly should be tested. The testing of the data is based on sent data. This is done by an edge connection of the machines using hardware or software. Using the data scanning method, Quantis reads the control units (PLC) and searches for the data blocks with their addresses. In the process, values are checked for presence and for validity.
After the test, the results are made available to the customer in the form of a report. From this, Quantis derives recommendations for action as to which values need to be adjusted in order to achieve high data quality. The customer can then decide together with the OEM which data points are to be improved.
The process plan for data validation essentially contains the following steps:
Step 1- Determination of the general conditions The conditions for a factory acceptance test are set in a meeting. An inventory of the customer's conditions and requirements is made. Among them, KPIs to be tested are defined per machine, system requirements are checked, and a plan for coordinating the measures is made.
Step 2 – Implementation The customer provides the virtual environment and infrastructure. The Edge is connected and the Pocket Factory is activated. Missing data is searched for and KPIs are configured.
Step 3 – Test
After the system test, data from the machine is collected. The machine is monitored via dashboard. The data is validated and documented.
Step 4 – Stabilization After data evaluation, the results are handed over in the form of a report, which is discussed with the customer and recommendations for action are made.
Step 5 – Goal The factory acceptance test of new machines is thus completed, weak points have been identified and an improvement can be made by the OEM in a targeted manner. After delivery, the new machine is implemented in the existing plant. It makes sense to carry out a permanent data validation even during operation, as the data can deviate due to integration into existing systems. In addition, updates or the replacement of touch computers make a renewed validation of the machine data necessary.
credits to Unsplash Foto Headway