PROCESS MINE-R
Authors: Noel Dempsey
Abstract: This project showcases an open-source process mining application built in R Shiny. Users are able to upload their own event log to explore in an activity map and examine throughput time as well as relative activity presence, activity antecedents / consequents, and absolute or relative frequencies.
The app was built primarily to aid those who want an easy to use tool for initial process mining exploration without having to pay for commercial alternatives.
Full Description:
This project showcases an open-source process mining application built in R Shiny. Users are able to upload their own event log to explore in an activity map and examine throughput time as well as relative activity presence, activity antecedents / consequents, and absolute or relatively frequencies.
The app was built primarily to aid those who want an easy to use tool for initial process mining exploration without having to pay for commercial alternatives.
What is process mining?
Process mining is the deep-dive analysis, discovery, monitoring and improvement of as-is processes. It takes all of the data related with a process and "mines" it for insight on potential improvement, focusing on finding better, more efficient pathways in operations.
Every time a process is completed, it creates data. For example, any time a customer service request is received, you know when the call was received, who handled it, how long it took, whether the issue was resolved, just to name a few. It could also include time-stamped logs of purchases made at a register. Or invoices sent on a particular date.
Common business processes leave digital footprints in enterprise systems. When using simple analytics tools, you can look at this data for processing, mining the information to understand some general trends - how long things take, the total volume of processes completed, who is handling what, etc.
Using a process mining platform, you can upload all of this data - both historical and real-time - to fully understand a process.
Usage
Upload an event log and explore the activity map. An event log is a tabular dataset grouped by cases and their associated activities and timestamps. Each row of the dataset corresponds to a case-activity. To work within the app, any event log must be a .csv file and have only the following three variables:
- case_id: the unique case indentifier
- activity_id: the activities within each case
- timestamp: the date and time of when the activity occurred in
YYYY-MM-DD HH:MM:SSformat
This tool is not yet fully optimised - it may run slower when using very large datasets, or when datasets contain a large variety of processes, compared to commercially available tools.
The folder example_eventlog in the Github repo contains an additional example eventlog called financial_eventlog.csv which can be uploaded to the app.
Shiny app: https://ndrshiny.shinyapps.io/process_mining_application/
Repo: GitHub - dempseynoel/process-mine-r: A process mining application built in R Shiny.
Thumbnail:
Full image:
