Process Mining

Process Mining is a revolutionary new development by Pallas Athena in association with the Eindhoven University of Technology. In most organisations there is a wealth of information hidden away in the system logs, audit trails and management information of the existing line of business applications. Process Mining now makes it possible to extract valuable process data from this information and present it in a meaningful way so you may understand what has been happening within your organisation over a period of time based on real performance data.

What is Process Mining?

Process Mining allows you to start ‘mining’ your historical data. You first need to identify some simple parameter fields in your current databases, these should include:

A record identifier;
An activity name;
Date and time of execution;
The user ID of the person executing the activity (optional).

Based on these data the process model is built by the process mining software. This is called Process Discovery. The model is built automatically from scratch when you have identified the required data and timeframe. The resulting process may then be animated and performance analysed.

Example

This example illustrates the detailed results you can get when mining a process model from historical data.

Banner - Process Mining animation

An invoice process 'mined' with Process Mining

Please note that the shown process model is automatically generated from historical data. The invoice cases are shown as coloured dots that run through the process model. The colour of a dot represents whether or not the SLA for that specific invoice case is met. This gives you a unique insight into the execution of your business process!

What does Process Mining help you achieve?

Automatic generation of the process, controls, dates, organisation and staff structures
Improve existing models
Analyse the differences between the process you defined and the actual implementation (are you in ‘control’?)
Identify bottlenecks at a glance
An objective overview
Roll forward the actual historical data to future simulation for future efficiency and cost savings