-Process mining, i.e., a sub-field of data science focusing on the analysis of event data generated during the execution of (business) processes, has seen a tremendous change over the past two decades. Starting off in the early 2000's, with limited to no tool support, nowadays, several software tools, i.e., both open-source, e.g., ProM and Apromore, and commercial, e.g., Disco, Celonis, ProcessGold, etc., exist. The commercial process mining tools provide limited support for implementing custom algorithms. Moreover, both commercial and open-source process mining tools are often only accessible through a graphical user interface, which hampers their usage in large-scale experimental settings. Initiatives such as RapidProM provide process mining support in the scientific workflow-based data science suite RapidMiner. However, these offer limited to no support for algorithmic customization. In the light of the aforementioned, in this paper, we present a novel process mining library, i.e., Process Mining for Python (PM4Py), that aims to bridge this gap, providing integration with state-of-the-art data science libraries, e.g., pandas, numpy, scipy and scikit-learn. We provide a global overview of the architecture and functionality of PM4Py, accompanied by some representative examples of its usage.
The field of process mining [
Both ProM and Apromore put a significant emphasis on
non-expert usability, i.e., by means of providing an easy to
use graphical user interface. Whereas such an interface helps to
engage non-expert users and, furthermore, helps to showcase
process mining to a larger audience, it hampers the usability
of the tools for the purpose of large-scale scientific
experimentation [
To bridge the aforementioned gap, i.e., the lack of process mining software that i) is easily extendable, ii) allows for algorithmic customization and iii) allows us to easily conduct large scale experiments, we propose the Process Mining for Python (PM4Py) framework. To achieve the aforementioned goals, a fresh look on the currently available programming languages and libraries indicates that the Python programming language2, along with its ecosystem, is most suitable. In particular, the data science world, both for classic data science (pandas, numpy, scipy . . . ) and for cutting-edge machine learning research (tensorflow, keras . . . ), is heavily using Python.
Other libraries, albeit with a lower number of features, exist
already for the Python language (PMLAB [
The bupaR library [
Lowering the barrier for algorithmic development and
customization when performing a process mining analysis
compared to existing academic tools such as ProM [
Allow for easy integration of process mining algorithms with algorithms from other data science fields, implemented in various state-of-the-art Python packages. 1http://rapidminer.com 2 http://python.org 1 from pm4py . a l g o . d i s c o v e r y . a l p h a import v e r s i o n s 2 from pm4py . o b j e c t s . c o n v e r s i o n . l o g import f a c t o r y a s l o g c o n v e r s i o n 3 ALPHA VERSION CLASSIC = ’ c l a s s i c ’ 4 ALPHA VERSION PLUS = ’ p l u s ’ 5 VERSIONS = fALPHA VERSION CLASSIC : v e r s i o n s . c l a s s i c . apply , 6 ALPHA VERSION PLUS : v e r s i o n s . p l u s . apply g 7 d e f apply ( log , p a r a m e t e r s =None , v a r i a n t =ALPHA VERSION CLASSIC ) : 8 r e t u r n VERSIONS [ v a r i a n t ] ( l o g c o n v e r s i o n . apply ( log , p a r a m e t e r s , l o g c o n v e r s i o n . TO EVENT LOG ) , p a r a m e t e r s ) Create a collaborative eco-system that easily allows researchers and practitioners to share valuable code and results with the process mining community.
Provide accurate user-support by means of a rich body of documentation on the process mining techniques made available in the library.
Algorithmic stability by means of rigorous testing.
The remainder of this paper is structured as follows. In Section II, we present the architecture and an overview of the features provided by PM4Py. In Section III, we present some representative examples (process discovery, conformance checking). Section IV discusses the maturity of the tool and Section V concludes this paper.
In order to maximize the possibility to understand and re-use the code, and to be able to execute large-scale experiments, the following architectural guidelines have been adopted on the development of PM4Py:
A strict separation between objects (event logs, Petri
nets, DFGs, . . . ), algorithms (Alpha Miner [
Most functionality in PM4Py has been realized through factory methods. These factory methods provide a single access point for each algorithm, with a standardized set of input objects, e.g., event data and a parameters object. Consider the factory method of the Alpha Miner, depicted in Fig. 1. The Alpha (variant=’classic’) and the Alpha+ (variant=’plus’) are made available. Factory methods allow for the extension of existing algorithms whilst ensuring backward-compatibility. The factory methods typically accept the name of the variant of the algorithm to use, and some parameters (shared among variants, or variant-specific).
In the remainder of this section, we present the main features of the library, organized in objects, algorithms, and visualizations.
Within process mining, the main source of data are event data, often referred to as an event log. Such an event log, represents a collection of events, describing what activities have been performed for different instances of the process under study. PM4Py provides support for different types of event data structures:
Event logs, i.e., representing a list of traces. Each trace, in turn, is a list of events. The events are structured as key-value maps.
Event Streams representing one list of events (again represented as key-value maps) that are not (yet) organized in cases.
Conversion utilities are provided to convert event data objects from one format to the other. Furthermore, PM4Py supports the use of pandas data frames, which are efficient in case of using larger event data. Other objects currently supported by PM4Py include: heuristic nets, accepting Petri nets, process trees and transition systems.
The PM4Py library provides several mainstream process mining techniques, including:
Process discovery: Alpha(+) Miner [
Conformance Checking: Token-based replay and
alignments [
Measurement of fitness, precision, generalization and simplicity of process models.
Filtering based on time-frame, case performance, trace endpoints, trace variants, attributes, and paths.
Case management: statistics on variants and cases. Graphs: case duration, events per time, distribution of a numeric attribute’s values.
Social Network Analysis [
NetworkX: static representation of social networks. Pyvis: web-based, dynamic representation of social networks (see Fig. 4).
using Alpha Miner and visualize the process model. The factory methods that are needed (XES importer, Alpha Miner and Petri net visualization) are loaded (line 1-3). Then, an XES log is imported (line 4), the Alpha Miner is applied providing the log object (line 7), and the visualization is obtained: a factory method is applied to layout the graph (line 8), and the In this section, we provide some examples of the use of result is shown in a window (line 9). The result is shown in PM4Py. 1 from pm4py . a l g o . c o n f o r m a n c e . a l i g n m e n t s import f a c t o r y a s a l i g n m e n t s 2 # a l i g n m e n t s a c c e p t s a l o g and an a c c e p t i n g P e t r i n e t , i . e . 3 # a P e t r i n e t a l o n g w i t h an i n i t i a l ( im ) and a f i n a l ( fm ) m a r k i n g 4 a l i g n e d t r a c e s = a l i g n m e n t s . apply ( log , n e t , im , fm ) 5 f o r i n d e x , r e s u l t i n enumerate ( a l i g n e d t r a c e s ) : 6 p r i n t ( i n d e x , r e s u l t [ ’ a l i g n m e n t ’ ] ) [ ( ’ r e g i s t e r r e q u e s t ’ , ’ r e g i s t e r r e q u e s t ’ ) , ( ’>>’ , None ) , ( ’ c h e c k t i c k e t ’ , ’ c h e c k t i c k e t ’ ) , ( ’ examine t h o r o u g h l y ’ , ’ examine t h o r o u g h l y ’ ) , ( ’>>’ , None ) , ( ’ d e c i d e ’ , ’ d e c i d e ’ ) , ( ’>>’ , None ) , ( ’ r e j e c t r e q u e s t ’ , ’ r e j e c t r e q u e s t ’ ) ] of a trace on an example log and model is reported. the result. First, the alignments factory method is loaded (line 1). Then, the alignments between a log object and a process model are obtained (line 4). For each aligned trace (line 5) the alignment result is displayed on the screen (line 6). The alignment of a trace is reported in the lower part of Fig. 3.
by 200 students in the “Introduction to Data Science” course held by the Process and Data Science group in the RWTH Aachen University. Already two academic projects have been supported by PM4Py and are publicly available:
Usage of probabilistic automata for compliance checking Prefix for event features. There are some integrations of the PM4Py library in bupaR R process mining library uses PM4Py to handle alignments and get models using the Inductive Miner. A data analytics web interface was written in Vue.JS In Fig. 6, some statistics taken from Google Analytics are reported about the number of accesses to PM4Py web site during the month of February 2019. In Fig. 7, some statistics about the downloads of the PM4Py library from PIP are reported. with maximum score, has been awarded to the PM4Py library.
PM4Py process mining library (http://www.pm4py.org) has been introduced. PM4Py supports a rapidly growing set of process mining techniques (discovery, conformance checking, enhancement . . . ). A video presenting the library and some example applications (log management, process discovery, conformance checking) has been made available3. The library can be installed4 through the command pip install pm4py. Extensive documentation is provided through the official website of the library. Moreover, the Github repository supports a collaborative eco-system where users could signal problems or contribute to the code. 4Additional prerequisites, available at the page http://pm4py.pads.rwthaachen.de/installation/ have to be installed.