PROCESS MINING – next level of business analytics

8 October 2019

Data – not until only recently a deficit material willingly replaced with heuristic methods and “expert” opinions. Currently, in the era of Big Data and millions of unanalyzed pieces of information, they overwhelm with their size, dynamic growth, and accessibility. They evoke the feeling of powerlessness. They give the illusion of control and effective management. In reality, however, they become a graveyard for dormant knowledge, which has been requiring deep exploration for many years. How to extract the said effectiveness, combine it with efficiency and transparency of your business, and embrace it in a single simple analysis?

 

Process Mining – process seen with the eyes of your data

Process Mining is a relatively young field of knowledge which allows obtaining a detailed image of actual processes taking place in enterprises. It makes it possible to build, verify, and extend models of these processes based on logs (event logs). The said events, taking the form of distributed data, are a key starting element for each analyzed process. Instead of a static diagram, they represent a factual course of individual business operations.

 

How does Process Mining work?

The effectiveness and usefulness of the Process Mining analyses, just like for all other analytical methods, largely depend on the quality of input data. The better the data are, the better the output results one can expect. In Process Mining, input data are referred to as Event Logs. They need to be composed of at least the following elements:

- token identifier (unique number);

- process activity name;

- name of user performing activity;

- activity start and end date.

What is important – thanks to their simplicity – input data can be generated and retrieved from various sources (databases), which allows their comprehensive analysis via specially designed algorithms and processing tools.

 

In order for the Process Mining analysis to bring satisfactory results, three key assumptions need to be met:

  • DISCOVERY, that is retrieving data and generating a model using algorithms;
  • CONFORMANCE, that is comparing actual processes obtained with reference models (e.g. Business Process Modeling Notation) to identify differences and deviations;
  • ENHANCEMENT, that is streamlining available reference models based on the processed data from the actual process image.

 

Advantages of employing Process Mining in business

Using Process Mining analyses can bring value to organizations undergoing digital transformation. The most important advantages include:

  • The possibility to merge distributed data in various business applications;
  • Full (detailed) knowledge of the factual course of the process;
  • Profound process analysis, based on basic data (activity start and stop, resource, role), presented from various perspectives (e.g. efficiency-related, organizational).
  • Automatically generated visualizations of actual processes;
  • Simple and fast generation of the models of actual processes based on data analyzed;
  • Possibility of daily observation of changes in a dynamic process through models generated;
  • Possibility of identifying the best practices to apply at the corporate level;
  • Possibility of extracting the entire added value from business processes in an organization.

The above functionalities allow implementing effective optimization activities in various business areas and reacting dynamically to variability (with the implementation of preventive measures and corrective actions).

 

Reach the higher level of analytics

The OptiBuy group of consultants have used the Process Mining analysis at one of their clients – AmRest Holdings SE. It is the largest independent company running restaurant chains in Central and Eastern Europe.

- Using the Process Mining analysis in the project carried out for AmRest allowed us to verify the process image arising from the available documentation and discussions with internal customers, as well as significantly streamlined the course of the entire project – says Mariusz Krzysztoń, Manager at OptiBuy.

- The Process Mining analysis used in the consultancy project carried out by OptiBuy for AmRest Polska made it possible to specify and define purchasing processes in our organization in a feasible manner. On that basis, bottlenecks and ineffectivenesses occurring in each of them were identified, and their elimination reduced the meantime of the processes and allowed the unification of similar workflows – adds Irina Dudnichenko, Procurement Process Manager at AmRest Holdings.

 

More information

Back to list