Process Mining Software

Process Mining: Data Science in Action

prof.dr.ir. Wil van der Aalst www.processmining.org

©Wil van der Aalst & TU/e (use only with permission & acknowledgements) Data scientists need to have data (see previous lectures), tools, and preferably also questions. Process mining toolbox Spectrum of tools

part of a bigger suite

process mining specific Disco ProM

simple, easy powerful, many to use, ... functions, extendible, ... ©Wil van der Aalst & TU/e (use only with permission & acknowledgements) Spectrum of tools

Knime data RapidMiner mining ... R

part of a bigger suite

process mining specific Disco ProM

simple, easy powerful, many to use, ... functions, extendible, ... ©Wil van der Aalst & TU/e (use only with permission & acknowledgements) Spectrum of tools

Knime data RapidMiner mining ... R

part of a bigger suite Perceptive Process Mining ...... Celonis process mining Discovery specific Disco ProM

simple, easy powerful, many to use, ... functions, extendible, ... ©Wil van der Aalst & TU/e (use only with permission & acknowledgements) Spectrum of tools

Knime data RapidMiner mining ... R

part of a bigger suite RapidProM Perceptive Process Mining ...... Celonis process mining Discovery specific Disco ProM

simple, easy powerful, many to use, ... functions, extendible, ... ©Wil van der Aalst & TU/e (use only with permission & acknowledgements) Checklist for a process mining tool

people organizations business “world” machines processes documents • Which of the 10 process mining information system(s) activities are supported? provenance event logs

“pre “post mortem” current historic mortem” • Which perspectives are data data supported? navigation auditing cartography • What modeling notations are check detect predict explore promote discover enhance compare diagnose recommend supported? models de jure models de facto models • What import and export formats control-flow control-flow are supported (event logs, data/rules data/rules

resources/ resources/ models, alignments, etc.)? organization organization

©Wil van der Aalst & TU/e (use only with permission & acknowledgements) Example: ProM

people organizations business “world” machines • Aims to cover the whole process processes documents

information system(s) mining spectrum. provenance event logs • Notations supported: Petri nets “pre “post mortem” current historic mortem” data data (many types), BPMN, C-nets, fuzzy navigation auditing cartography models, transition systems, Declare, check detect predict explore promote discover

enhance etc. compare diagnose recommend

models • Also supports conformance de jure models de facto models checking and operational support. control-flow control-flow • Many plug-ins are experimental data/rules data/rules

resources/ resources/ prototypes and not user friendly. organization organization

©Wil van der Aalst & TU/e (use only with permission & acknowledgements) Example: Disco

people organizations business “world” machines processes documents • Focus on discovery and performance

information system(s) analysis (including animation). provenance event logs • Powerful filtering capabilities for “pre “post mortem” current historic mortem” data data comparative process mining and ad- navigation auditing cartography hoc checking of patterns. check detect predict explore promote discover enhance compare diagnose

recommend • Uses a variant of fuzzy models, etc.

models • Does not support conformance de jure models de facto models checking and operational support. control-flow control-flow

data/rules data/rules • Easy to use and excellent

resources/ resources/ organization organization performance.

©Wil van der Aalst & TU/e (use only with permission & acknowledgements) Other commercial tools (1/2)

• Perceptive Process Mining (before Futura Reflect and BPM|one) (Perceptive Software) • ARIS Process Performance Manager (Software AG) • QPR ProcessAnalyzer (QPR) • Celonis Discovery (Celonis) • Interstage Process Discovery (Fujitsu) • Discovery Analyst (StereoLOGIC) • XMAnalyzer (XMPro) • ©Wil van der Aalst & TU/e (use only with permission & acknowledgements) Other commercial tools (2/2)

• Most tools are similar to Disco in terms of scope (but often less user-friendly and less performant). • Typical coverage:

control-flow control-flow and ... only time resources data . discovery L M conformance L+M D enhancement L+M M

©Wil van der Aalst & TU/e (use only with permission & acknowledgements) Three examples (next to ProM and Disco)

Knime data RapidMiner mining ... R

part of a 1 bigger suite RapidProM 3 Perceptive Process Mining ... 2 ... Celonis process mining Discovery specific Disco ProM

simple, easy powerful, many to use, ... functions, extendible, ... ©Wil van der Aalst & TU/e (use only with permission & acknowledgements) RapidProM

www.rapidprom.org

©Wil van der Aalst & TU/e (use only with permission & acknowledgements) Create a process mining workflow

©Wil van der Aalst & TU/e (use only with permission & acknowledgements) check conformance using alignments

copy objects create ProM (technicality) context (technicality)read event log discover a Petri net using the Alpha algorithm ©Wil van der Aalst & TU/e (use only with permission & acknowledgements)

©Wil van der Aalst & TU/e (use only with permission & acknowledgements) Another process mining workflow

©Wil van der Aalst & TU/e (use only with permission & acknowledgements) create dotted chart create Petri net using inductive miner

read event create social log network based on handover of work ©Wil van der Aalst & TU/e (use only with permission & acknowledgements) Results

©Wil van der Aalst & TU/e (use only with permission & acknowledgements) RapidProM

• Can be used to repeat process mining analyses (scientific experiments, reusable macros, ). • Can be used to combine process mining with data mining, text mining, etc.

©Wil van der Aalst & TU/e (use only with permission & acknowledgements) Celonis Discovery web-based

supports process discovery

©Wil van der Aalst & TU/e (use only with permission & acknowledgements) (but no concurrency) frequencies

©Wil van der Aalst & TU/e (use only with permission & acknowledgements) time information

©Wil van der Aalst & TU/e (use only with permission & acknowledgements)

animation ©Wil van der Aalst & TU/e (use only with permission & acknowledgements)

frequency- based filtering (1/3)

©WilBPI van challengeder Aalst & TU/e (use2013 only withlog permission & acknowledgements)

frequency- based filtering (2/3)

©WilBPI van challengeder Aalst & TU/e (use2013 only withlog permission & acknowledgements)

frequency- based filtering (3/3)

performance-related dashboards ©Wil van der Aalst & TU/e (use only with permission & acknowledgements) Perceptive Process Mining

web-based

supports process discovery

various ways of filtering e.g. based on frequency

replay event log

tokens colored based on flowtime

replay based on social network (top-3)

performance- related charts

©Wil van der Aalst & TU/e (use only with permission & acknowledgements) Related tools Although related, (beyond pure process mining tools)How about BI tools?classical data mining and machine learning BPM systems can tools typically do not be extended with support process process mining mining. capabilities.

Related to operational support, but typically not process-oriented.

• Possible definition (one of many): "BI is a set of methodologies, processes, architectures, and technologies that transform raw data into meaningful and useful information used to enable more effective strategic, tactical, and operational insights and decision making." [Boris Evelson, Forrester Research 2010] • Many products, e.g., IBM Cognos Business Intelligence (IBM), Oracle Business Intelligence (Oracle), SAP BusinessObjects (SAP), WebFOCUS (Information Builders), MS SQL Server (Microsoft), MicroStrategy (MicroStrategy), NovaView (Panorama Software), QlikView (QlikTech), SAS Enterprise Business Intelligence (SAS), TIBCO Spotfire Analytics (TIBCO), Jaspersoft (Jaspersoft), and Pentaho BI Suite (Pentaho).

• ETL (Extract, Transform, and Load). • Ad-hoc querying. • Reporting. • Interactive dashboards. • Alert generation.

©Wil van der Aalst & TU/e (use only with permission & acknowledgements) Tools are available, but process mining is still a relatively young discipline.

New tools will appear in Avoid the trap: coming years and process • process mining ⊆ data mining mining functionality will • process mining ⊆ BI be embedded in more • process mining ⊆ BI/BPM/DM suites. Use checklists! Part I: Preliminaries Part III: Beyond Process Discovery Chapter 1 Chapter 2 Chapter 3 Chapter 7 Chapter 8 Chapter 9 Introduction Process Modeling and Data Mining Conformance Mining Additional Operational Support Analysis Checking Perspectives

Part II: From Event Logs to Process Models Part IV: Putting Process Mining to Work

Chapter 4 Chapter 5 Chapter 6 Chapter 10 Chapter 11 Chapter 12 Getting the Data Process Discovery: An Advanced Process Tool Support Analyzing “Lasagna Analyzing “Spaghetti Introduction Discovery Techniques Processes” Processes”

Part V: Reflection Chapter 13 Chapter 14 Cartography and Epilogue Navigation

©Wil van der Aalst & TU/e (use only with permission & acknowledgements)©Wil van der Aalst & TU/e (use only with permission & acknowledgements)