Process mining as a link between business process modelling and data mining

3 min read

Jul 10, 2018 12:00:00 AM

As the number of processes running daily is rapidly increasing, it is critical to understand these processes and their interactions so we can improve them in areas such as businesses that affect our daily lives. Information systems record a lot of data about events that are happening. Many of these processes are being recorded in event logs. All these event logs form a large amount of data that can be discovered, analyzed and understood through process mining. So what is process mining? The core idea of process mining is to analyze data from a process perspective . Process mining will help you answer questions such as:

What does this process currently look like?
What are the sequence of steps in my process?
Are there any unnecessary steps that could be eliminated?
Are there any bottlenecks? At which step?
Are there any deviations from the defined process model?
How much time does it take to execute a process?

And there are many more questions can be answered by process mining techniques. Information systems are becoming more connected with the operational processes that they support. Organizations have problems extracting value from this collected data. Here comes the role of process mining: it uses event data to extract information related to those processes. Process mining is not only about collecting data; it is also about analyzing this data to improve processes. Process mining has been used in analyzing and studying processes in many areas such as healthcare, business, finance and software development. Process mining plays a major role in bridging the gap between process model analysis and data analysis techniques. Process mining is about extracting meaningful information about the processes using the collected event logs. There are three types of process mining: discovery, conformance and enhancement.

Discovery: the process discovery technique produces a model from event logs.
Conformance: compares an existing process model with an observed behavior in the event logs. It checks if what is recorded in event logs or reality conforms with the model.
Enhancement: improves and extends an existing process model based on information extracted by process mining.

Several tools have been developed for process mining. ProM is an extensible framework that supports many process mining techniques in the form of plug-ins. ProM is well known among researchers and is the most popular and frequently used framework in process mining implementations. Additionally, there are several commercial tools, for example, Celonis, QPR ProcessAnalyzer, Disco and ARIS Process Performance Manager. Why process mining? How is process mining different from data mining and process management techniques? All these are questions that you might be asking. To have a better idea, I will briefly compare process mining to process management and to data mining. Business process management (BPM) combines approaches from design, execution, control, measurement and optimization of business processes. The main focus of BPM is on process design and implementation. Additionally, process modelling plays an important role in the design phase and contributes to the configuration and implementation phase. Process modelling is more model-driven without taking into consideration the information hidden in the data. BPM focuses on monitoring, adjustment, diagnosis and requirements phases. These phases are more data-driven and process mining is mainly used in these phases of the BPM lifecycle. In this case, process mining can help to better understand the processes in more depth. Thus, we can see that process mining easily integrates into BPM. Basically, process mining can be applied to any process in which events can be recorded. Data mining techniques aim to analyze data sets to find relationships and to summarize the data in ways that are understandable and useful to the user. Like process mining, data mining is also data-driven. However, unlike process mining, the data mining techniques do not analyze data from a process perspective. Process models cannot be discovered or analyzed using the main data mining tools. When using process mining techniques, you study the data from a process viewpoint in mind. Data mining techniques do not consider end-to-end processes. To conclude this blog, process mining can be seen as the “missing link” between data mining and business process modelling. Process mining is an important tool for modern organizations that need to manage operational processes. On the one hand, there is an incredible growth in event data. On the other hand, processes and information need to be aligned perfectly in order to meet requirements related to compliance, efficiency, and customer service.

On this page

Ready to unlock value from your data?

With Pythian, you can accomplish your data transformation goals and more.

Speak with Pythian consultants now →

What We Do

Process mining as a link between business process modelling and data mining

Share this

Share this

More resources

Where in a Modern Data Platform Should You Process Your Data?

Mitigating the Problems Created by Tombstones

Big mining gets a boost from big data

Ready to unlock value from your data?