The path towards industrial autonomy has to be taken step by step. Increasing autonomy also means a decline in human intervention. If, until now, decisions have often been made based on experience, a new decision-making basis will now enter the spotlight: Data. Harnessing data isn’t new, but the requirements placed on data quality are totally different when using them for autonomous operation.
Data is accumulating in ever greater volumes, and these volumes will continue to grow thanks to digitalisation and the IIoT. Data will also become more significant for business processes. While data was once saved specific to an application in individual databases, in future, it won’t be possible without networking and joint evaluation. Big data analyses offer a wide range of possibilities for linking different data types to generate new business models or streamline and gradually make processes more autonomous.
Insufficient data quality as a cost driver
The quality of the data doesn’t just refer to the data sets themselves, but their permeability across all divisions of a company, their adaptation to different use cases and how they are visualised. Of course, all of this has to take place complying with data security and all legal specifications. Imagining this complexity, it’s no wonder that many data-driven projects are still extremely cost-intensive or even fail.
Costs incur right from the start due to all the efforts required to put together the necessary data and get it in the form and quality needed for analysis. If incorrect data are used as the basis or relevant information is missing, there is a high risk of inaccurate evaluations. Later, additional costs can result from the online implementation of data-driven applications in unsuitable infrastructures, not to mention the maintenance of such systems. The effort required for individual projects may be acceptable compared with restructuring an entire data management system. Extensive system autonomy cannot be achieved with this.
Providing the necessary infrastructure
The 3B Maturity Benchmark of the Singapore Economic Development Board based on the Smart Industry Readiness Index (SIRI) highlights just how important data availability is for companies in a very descriptive way. A SIRI assessment illuminates the central areas of a company in terms of its Industry 4.0 maturity. While the gaps between the three observed groups, “Bottom Performers”, “Broad Middle”, and “Best-In-Class”, are relatively constant in all areas, the category of data connectivity stands out quite a bit. The leading companies are far ahead of the other comparison groups in terms of networking production. All components and computer systems are interoperable and securely connected. Communication primarily takes place in real-time.
Companies have mainly invested in providing the necessary infrastructure ahead of other activities such as vertical and horizontal integration or implementation of intelligent systems. After all, like data quality, reliable and fast interoperability across all levels is essential for autonomous systems.
High data quality as a cost driver?
Often, the required efforts and costs are also off-putting. Is ensuring a high data quality also a cost driver? Let’s take a look at the other side. The primary target of an intelligent data-driven system is to achieve savings. So, savings are directly linked to the data used in the application. As a simple formula, the greater the savings, the higher the value of the data. The higher the data quality, the better the application and, therefore, most likely the savings. In this way, the cost for providing data can be balanced out with the benefits of their processing. Thus, high data quality is not a cost driver; it represents a value that should compensate for the cost of providing quality data. So, investing in good data quality pays off because it will be vital on the path to industrial autonomy.
Secure and maintain
Such an implementation should always be accompanied by good data governance. Securing a comprehensive framework is crucial to maintaining good data quality and administration, protection, security, and compliance. Company-wide guidelines and data management organised centrally with standardised processes and methods help systematise the data infrastructure. And to establish a continuous process that makes the costs predictable and therefore included in the balance in advance.
Smart manufacturing: boundaries fade – requirements stay the same
