Blog - Process Mining using Artificial Intelligence

• Read original article here

Only a few years ago, pen and paper were enough to map processes and try to find areas for improvement. Once a report was done, the process might have changed completely and the report was left obsolete.

Instead of using pen and paper, process mining technology (read: what is process mining) uses Artificial Intelligence and Machine Learning to automatically extract, visualize, and make sense of operational data from organizations' IT systems.

Process mining software like QPR ProcessAnalyzerhelps you understand this dynamic visualization (or "digital twin") of your organization and make interesting findings: where costs can be saved, processes can be simplified, automation needs to be improved most, etc.

Not only can you find root causes for your problems, but also predict problems and get automatic alerts when problems are about to arise - so you can proactively take action.

This blog opens up how Artificial Intelligence and Machine Learning are used in process mining: cluster cases, predict problems, and more. 

When your employees or software robots interact with IT systems – such as SAP, Salesforce, or Oracle - in your company, the activities leave a trace of data behind, referred to as an event log.

Process mining takes the data that exists in these information systems (using pre-built connectors) and uses it to visualize the real-life execution of your company’s processes together with other insights drawn from the event logs.

It points out what you should focus on in order to improve your efficiency, for instance:

Intelligent Process Mining uses machine learning algorithms to add artificial intelligence into traditional process mining. Intelligent Process Mining capabilities can be divided into four categories:

Process mining was first developed as a descriptive method to discover patterns and gain a deep understanding of real-life business operations. While both traditional BI and process mining help you monitor KPIs and targets, process mining adds to this by helping you quickly perform pre-built process analyses in real-time. These pre-built analyses easily reveal a variety of areas for optimization: bottlenecks, compliance violations, and process deviations, which are visualized based on your actual operations.

“We gave the data of the system, and right away, in 5 minutes, we saw the bottlenecks of the process." -Piraeus Bank

The following three analyses turn the traditional process discovery to intelligent process discovery:

After detecting a process problem, you want to know why it happened. Diagnostic process mining provides answers not only to “what is happening?”, and “when is it happening?” – but also “why is this happening?”.

Process mining reveals problem areas in your processes by highlighting them in flowcharts and ranking them based on how they contribute to your business outcomes. This tells you what to prioritize when you want to improve your business operations. 

The following functionalities show you why your problems occurred: 

Read a more detailed description about these categories of Intelligent Process mining in the blog: Intelligent Process Mining using Machine Learning.

In this third step, intelligent process mining predicts future problems.

Predictive process mining predicts what will happen next in any given ongoing case. By using the information from all past and ongoing cases, the machine learning system can predict the outcome of each case. The more data there is, the better the accuracy of the prediction will be.

The last step, prescriptive process mining with QPR ProcessAnalyzer, introduces an ML-based Intelligent Orchestrator to help you succeed in your operations. This Orchestrator will learn and become aneven better companion for you over time.

In the past years, QPR has invested heavily in building more AI-powered features in process mining and succeeded in making QPR ProcessAnalyzer one of the leading solutions in the process mining market.

QPR ProcessAnalyzer’s Predict & Act functionality takes the standard KPI monitoring concept much further with AI/ML-based prediction capability. It allows you to act already before potential issues or bottlenecks arise. This is effectively changing the way organizations think - from late fixes to preventive actions. 

With the new Predict & Act functionality, process mining can: 

Read a more detailed description about these categories of Intelligent Process mining in the blog: Intelligent Process Mining using Machine Learning.

Since 2012, QPR has been working closely with Aalto University to develop Artificial Intelligence solutions for today's process mining use cases. 

During the QPR Conference 2019, Teemu Lehto (DSc in Technology, VP Process Mining at QPR) had a joint presentation with Alexander Jung (Assistant Professor) from Aalto University, a leading institution in technology and science with a sharp focus on machine learning. The presentation discussed the use of AI to drive process mining and business optimization.

During the presentation, Teemu and Alex discussed the following challenges:

Check out the findings in the 3-page whitepaper and videos below!

To learn more about how process mining can be applied in your organization, feel welcome to send us a message or book a demo with one of our experts.