Process mining and process discovery are relatively new fields on the horizon, but in recent times the activity and awareness in the space have increased tremendously. The process mining terminology is increasing as well with the new entrants into the process mining software space. This blog post is an attempt to demystify and provide clarity on some of the process mining terminology. (Of course, different companies may try to use different terms to denote something similar as a part of their branding effort or to portray subtle differences in how their process mining software works.)
Process Mining Terminology - Glossary of Essential Terms:
Process: A business process is an end-to-end flow of work to accomplish a business objective.
Process Map: A visual depiction of a process using standard process mapping symbology.
Process Mapping: Mapping a process is an effort to document how a process works – typically by a human analyst through interviews, observation, workshops, and other methods.
BPMN: Business Process Mapping Notation: A standard set of notations by Open Group to represent process steps.
UML: Process architects use Unified Modeling Language as another notation to represent process mapping steps.
Process Mining: Process mining, pioneered by Dutch professor Wil Van der Aalst, is a method of analyzing application event logs to decipher a process map. Applications crate the event logs as a part of process execution.
Process Discovery: Process discovery is a method to record user interactions with digital systems to compile a process. Process discovery is the next evolution of process mining. (At Skan, we call it CPX – Cognitive Process Extraction, and we create a process metamodel using computer vision and machine intelligence to observe work at scale and high precision and map processes out of the digital exhaust generated by user actions.) Other companies have used the term task mining or client-side process discovery as variations of this term.
Process Variant: Despite the myth of a golden process, enterprise business processes are an amalgamation of process runs with permutations and subtle differences. A Process Variant is a particular and differentiable set of process runs that rise above the threshold of noise and constitute a substantive portion of the process runs.
Runs: A run is an end-to-end execution of a business process.
Abandoned Process: A process that did not reach a logical expected conclusion and stopped mid-way.
Reference Process or Standard Variant: A Reference Process or Standard Variant constitutes the process variant that conforms to the company’s preferred operating procedures. (One may term this as the preferred variant or even a golden process - though we believe a golden process is a misnomer.)
Explainable Variant: An explainable variant follows a decision fork in a process and one which can be explained by the logical paths in a process.
Process Drift: Process drift accounts for all processes that are not explainable based on a decision-driven path of a process.
Process Rerun: A process with the same case ID that is being rerun (due to an error, exception management, or other reasons)
Cycle Time: The time it takes for one process run to complete the full set of activities to accomplish a business process.
Process Quality Score: An overall score for a process that takes into consideration a variety of vectors (start time, finish time, time taken, error rates, etc.) to normalize and assign a quality score.
If you think we are missing any of the key process mining terminology, drop us a line.
Satya Iluri is Vice President of Strategy, Marketing, and Customer Success at Skan, an AI-powered process mining and process discovery platform.
“One of the things you don’t ever want to do is to automate a bad process. You are just going to make bad things happen faster, and that is not what anyone wants.”
Digital Enablement and Automation Services Lead, AVP FM Global