Skip to content

Data required for a Process Mining study

The tables below describe the essential (mandatory) and / or optional (optional) data that the software needs to perform an analysis.

Any input format (file) must contain these fields, which will be used in the data feed API.

Name Description mandatory/optional
Case ID Allows to distinguish, in the traces, the case to which the activity relates (examples: client ID, which one seeks to analyze the succession of activities, production batch, client file number, ...) mandatory
Activity Description or name of an activity (can be combined with an activity ID) mandatory
Activity Start date Date on which an activity begins mandatory
Activity End date Date an activity ends recommended (*)
Resource Description of the resource involved in the activity (ex: Customer advisor, Production resource, ...) recommended (**)
Other information If the data source contains other information allowing, such as Resources for example, to complete the business analysis, to categorize, to understand or to deduce the behavior of the process, this information is obviously to be kept. They are not formally described as necessary data, but can add much to the analysis. Examples: promotion period, commercial context, production plant, ... optional

(*) If we want to take advantage of the time spent in activities

(**) If we want to obtain relevant information on the Resources associated with the process

Example of event log: industrial process

Fictitious example of an industrial process The second line indicates the business mapping corresponding to the necessary data

Case ID Activity Start date End date Resource Other
lot number step entry timestamp output timestamp robot id production line
1 cutting 13:46:00 13:46:24 R2 RENNES
1 stamping 13:46:26 13:46:28 R2 RENNES
1 marking 13:46:30 13:46:33 R2 RENNES
1 setting 13:46:36 13:46:40 R2 RENNES
1 labeling 13:46:45 13:46:46 R2 RENNES
2 cutting 13:46:05 13:46:29 R2 LYON
2 stamping 13:46:30 13:46:32 R2 LYON
2 marking 13:46:35 13:46:37 R2 LYON
2 labeling 13:46:40 13:46:41 R2 LYON

Other Example of event log: market process

Case ID Activity Start date End date Resource Other Other
order number activity date advisor product quantity
57021 order 28/01/2016 11:10:05 Ludovic iPhone 1
57022 order 28/01/2016 12:34:23 Sarah Hard disk 2 To 2
57023 order 28/01/2016 13:30:20 André Headphones 1
57021 stock check 28/01/2016 11:15:05 Ludovic iPhone 1
57022 livraison 28/01/2016 12:35:23 Sarah Hard disk 2 To 2
57023 cancellation 28/01/2016 13:33:10 André Headphones 1

Other Example of event log: hospital journey

Case ID Activity Start date End Date Resource Other Other
patient care entry date output date staff age cost class
883 radiology 10/12/2016 09:10:05 10/12/2016 09:23:09 Dr Durant 45 A
884 reception 10/12/2016 09:12:50 10/12/2016 09:20:30 J. Dupuis 65
884 blood test 10/12/2016 10:02:20 10/12/2016 10:23:12 A. Joyeux 65 C
884 billing 10/12/2016 10:45:05 10/12/2016 10:50:22 J. Dupuis 65
880 scanner 10/12/2016 11:10:00 10/12/2016 11:50:09 Dr Poisson 36 E