x2Log research project: Process Mining for the manufacturing industry
Utilise process data and work more efficiently
About the project:
What is behind x2Log?
The x2Log research project aimed to develop a software framework that improves the collection and analysis of process data. The aim was to make it usable for process mining. A reference architecture was then to be created. This is a step-by-step guide on how the x2Log framework can be used in practice. A reference implementation at the participating application partners should demonstrate the potential for cost savings and quality improvements.
Which consortium partners were involved in the realisation of x2Log?
The following partners have formed a consortium to realise the research project:
– Fraunhofer Institute for Applied Information Technology FIT (research partner & consortium leader)
– soffico GmbH (technology and implementation partner)
– RAPA GmbH (application partner)
– targenio GmbH (technology and implementation partner)
What is process mining?
Process mining describes the use of process data generated during the execution of business processes in IT systems to gain insights and information about processes within an organisation. The main objective is to make real process flows comprehensible and transparent in order to enable data-based analysis of the processes.
In this way, companies can understand how their processes actually run – and where they deviate from the planned sequence.
The data obtained can then be used to derive insights into workflows that can contribute to optimisation and increased efficiency.
The challenge of process mining
Many companies already rely on process mining, but quickly reach their limits in practice – namely in one of the key success factors: data preparation. This is because the data required for well-founded process analyses is often available in different formats, comes from heterogeneous systems and is also often stored in data silos. This diversity makes it considerably more difficult to merge data in a consistent and usable way. The necessary consolidation and transformation of these data sources is usually associated with considerable manual effort – an effort that many companies shy away from.
As a supposedly simpler solution, companies therefore often fall back on already consolidated data sources in the appropriate format. However, this approach comes at a price: a large proportion of potentially relevant information remains unused, which leads to so-called ‘blind spots’ in the analysis process – and significantly limits the informative value of process mining.
This is exactly where Orchestra comes in: The solution enables companies to intelligently integrate different data sources and formats – and thus offers a much more comprehensive and in-depth view of existing processes. This creates real added value, particularly for companies that are already using process mining but have only been able to gain limited insights to date.
Requirements for the x2Log project
The aim of x2Log was to use process mining to enable end-to-end process analysis by bringing together previously unused structured and unstructured data as automatically as possible and preparing it for further analyses.
As soon as data preparation is no longer an obstacle, companies can exploit the full potential of data-based process optimisation – with tangible benefits in terms of costs, time and quality.
Realisation of the x2Log research project
In the first step, the consortium partners developed a theoretical framework. This is capable of automatically connecting data sources of all kinds, correlating them with process data and thus enabling end-to-end process analyses.
In the second step, the framework was ‘translated’ into a reference architecture for industrial use. The result is a blueprint that serves as a guideline for future implementations.
In the third phase of the research project, the results of the work were finally tested on real business processes. For this purpose, the application partner RAPA GmbH provided real process data that was used for process mining.
The company, which specialises in the production of valves for the automotive industry, medical technology and the manufacturing and process industry, wants to use the improved database to increase the efficiency of its manufacturing processes.
The role of soffico in x2Log
As a technology and implementation partner, soffico played a central role in the research project. This is because the reference architecture was implemented with the help of Orchestra.
Orchestra is the integration solution from soffico GmbH. It enables data integration and process automation, which are essential for AI, automation and data room initiatives.
With the Orchestra Designer, scenarios can be easily compiled graphically or programmed. Process model templates can also be created and made available in the same way.
Such process model templates are provided in Orchestra for the individual data integrations and components of the reference architecture. In this way, not only can the reference architecture be implemented efficiently, but the mapping of the individual scenarios in the Designer also enables seamless data integration and transformation. This is essential for effective and precise value stream mapping.
Process mining shows how processes actually work and not just how they are defined on paper.
It helps to identify bottlenecks or delays in processes that can affect productivity and efficiency.
By analysing standard processes, companies can identify repetitive and time-consuming tasks that can be automated to save resources.
Organisations can make informed decisions based on actual data, making it easier to plan and implement process improvements.
By eliminating manual analysis activities, our process engineers are relieved of several hours of work every day.
Markus Benndorff
