Data analytics plays an absolutely crucial role in industry optimization. With relatively small investments, industrial CEOs can achieve significant improvements in almost all areas of their company.
Collecting and analysing production data allows you to monitor the performance of the entire production process (e.g. production cycle time, breaks, machine utilisation, material or energy loss or waste) and identify areas that need to be optimised. By analysing data from production equipment and sensors, machine failures and the need for timely maintenance can be predicted. This can prevent production downtime, which can cost a company up to several million dollars in fines.
Using data analytics, companies can forecast demand and better manage their inventory and component supply, optimise production schedules and space utilisation. Tracking material flow, time data and space requirements allows you to identify areas of over or under utilisation and make appropriate adjustments to maximise production capacity.
Industrial companies are aware of the benefits of data analytics and the majority (83 percent) of large enterprises and half (55 percent) of small and medium-sized enterprises use it in a more or less advanced form to monitor and optimize the efficiency of production lines.
"Currently, we can observe an increase in investments in areas related to digitalization and robotization of production processes, tracking of commodities throughout the entire production process (from delivery of basic raw materials to export of the finished product), streamlining and automation of warehousing and implementation of other technologies facilitating demanding and repetitive tasks that were previously performed by employees," says Jakub Lichnovský, partner at PRK Partners law firm.
Understanding supply and purchasing chains
Data analytics and software to optimise shipping routes and delivery schedules are used by 68 per cent of large and 18 per cent of small companies. Another quarter (23 percent) of large and 28 percent of small companies plan to implement this method of logistics management within three to five years.
František Podzimek from Siemens sees data analytics as the key to competitive and sustainable supply chains. "Without modern software solutions, it is difficult to track and coordinate all steps: from the initial idea for a product to its actual production, and to quickly and easily find relevant information for decision-making. Today, it is possible to view the history of a product or component, for example, verify its origin using blockchain, track the carbon footprint of a product in the supply chain, monitor lead times and generally optimise logistics and warehousing processes based on accurate and up-to-date data," he explains.
Jan Hirš of T-Mobile agrees, adding: "Based on historical data and trends, it is also possible to forecast demand for products and resources, which helps, for example, in production planning and inventory optimization."
Majority (75 per cent) of large enterprises and a third of SMEs (36 per cent) use software to monitor delivery times and use data analytics to identify areas for improvement. When it comes to monitoring inventory and planning product replenishment, 84 percent of large and 43 percent of small companies use data analytics to optimize logistics processes.
Companies collect data but do not integrate it with each other
Data incompatibility is still a big problem in manufacturing companies. Companies very often collect data from different sources, but no longer integrate them and work with them separately. Only two-fifths of large firms (43 percent) and one-quarter of small firms (23 percent) in the survey said they integrate the data they collect into a centralized database or platform.
Data integration is crucial to the overall efficient operation of a manufacturing company. It enables the connection and collection of information from different systems and sources within the production process. "Managers have an overall view of production, can monitor key performance indicators (KPIs) and analyse operational data. Overall supply chain management is improved. Data linkage also enables automation of information flow and minimizes manual operations. This reduces the likelihood of errors and speeds up production flow. Thanks to data integration, it is also possible to better identify areas of inefficiency and implement improvement measures," says Alena Burešová, Senior Industry Manager at the National Industry 4.0 Centre.
The most common errors in data evaluation
Unfortunately, Czech manufacturing companies very often collect incomplete or inaccurate data. Hence their evaluation is based on insufficient information. This can lead to wrong decisions and poor analysis. According to a survey by the National Centre for Industry 4.0, Czech manufacturing companies make the following mistakes most often when evaluating data:
- Data quality is not sufficient. If the data are contaminated, contain errors, missing or duplicated, this can significantly affect the results of the evaluation.
- Companies assume incorrect relationships between data or base analysis on inaccurate prior knowledge, often leading to erroneous conclusions.
- There are many different analytical methods and algorithms. Firms sometimes have difficulty choosing the one that fits their needs. If a firm chooses the wrong method for evaluating specific data sets, the results may be inaccurate or irrelevant.
- Evaluating data requires the ability to interpret the results correctly. If a firm cannot interpret data correctly, erroneous conclusions and incorrect actions may result.
- Firms do not always update data sufficiently and do not consider current trends in their industry, then they may miss important information and opportunities.
- Finally, firms do not evaluate data in context. This leads to misunderstanding and misinterpretation.
"Quite often companies skip the analysis of what data to collect and why to collect it - typically they collect data that is simply available - and then think about how to use it. The right thing to do is to first analyse which parameters are important - both for the company and, above all, from the perspective of the end customer, and choose which data needs to be collected and evaluated accordingly," Jiří Bavor from Eviden and Atos shares his experience from many years of practice.
"Very often we encounter following question in practice: 'We collect data, we have a lot of it, but we don't know what to do with it, can you advise us?' Without this initial analysis, they will hardly get information relevant to their business," concludes Jan Hroch, Executive Director of the Czech Maintenance Society.
More information can be found in the Czech Industry Analysis 2023 / 1, which is published on the occasion of the National Industrial Summit 2023. The research was conducted between March and April 2023 and was attended by 207 directors of manufacturing companies from all over the Czech Republic.