Process Mining Workshops: ICPM 2020 International Workshops, Padua, Italy, October 5–8, 2020, Revised Selected Papers (Lecture Notes in Business Information Processing, 406)

This book constitutes revised selected papers from the International Workshops held at the Second International Conference on Process Mining, ICPM 2020, which took place during October 4-9, 2020. The conference was planned to take place in Padua, Italy, but had to be held online due to the COVID-19 pandemic. The conference focuses on the area of process mining research and practice, including theory, algorithmic challenges, and applications. The co-located workshops provided a forum for novel research ideas. The 29 papers included in this volume were carefully reviewed and selected from 59 submissions. They stem from the following workshops: 1st International Workshop on Event Data and Behavioral Analytics (EDBA) 1st International Workshop on Leveraging Machine Learning in Process Mining (ML4PM) 1st International Workshop on Streaming Analytics for Process Mining (SA4PM'20) 5th International Workshop on Process Querying, Manipulation, and Intelligence (PQMI) 3rd International Workshop on Process-Oriented Data Science for Healthcare (PODS4H) 1st International Workshop on Trust and Privacy in Process Analytics (TPPA) Preface Organization Contents 1st International Workshop on Event Data and Behavioral Analytics (EDBA) First International Workshop on Event Data and Behavioral Analytics (EdbA’20) Organization Workshop Chairs Program Committee Visually Representing History Dependencies in Event Logs 1 Introduction 2 Related Work 3 Visualization Techniques 3.1 Visualization Directly-Follows Graph 3.2 Visualization Additional Rectangle 3.3 Visualization Additional Arc 4 Design Setting 5 Evaluation Results 6 Discussion 7 Implementation as ProM Plugin 8 Summary and Outlook A Rigorous Definitions References Analysis of Business Process Batching Using Causal Event Models 1 Introduction 2 Motivational Scenario 3 Related Work 4 Batch Analysis Based on Causal Event Models 4.1 Determine Causal Event Models for Event Log 4.2 Batching Analysis 4.3 Implementation 5 Results and Evaluation 5.1 Setup and Dataset 5.2 Visualization of the Results 5.3 Data Analyses 5.4 Discussion 6 Conclusion References Process “Prospecting” to Improve Renewable Energy Interconnection Queues: A Case Study 1 Background 2 Methodology 2.1 Gather DEC Regulatory Filing Data and Convert It into MS Excel Format 2.2 Assess Process Performance and Generate an Event Log CSV File 2.3 Conduct Petri Net Behavior and Event Log Conformance Analysis 2.4 Conduct Event Log Visualization and Directly Follows Graph Analysis 2.5 Analyze Key Date Behavior of Project Developers (Installers) 3 Results 3.1 Assess Process Performance 3.2 Petri Net Behavioral Analysis 3.3 Conformance Analysis 3.4 Event Log Visualizations 3.5 Directly Follows Graph 3.6 Key Date Behavioral Analysis: Solar Project Developers 4 Discussion 4.1 Petri Net Behavioral Analysis 4.2 Conformance and Event Log Analysis 4.3 Key Date Behavioral Analysis 4.4 Approach Viability 5 Conclusions References Automated Discovery of Process Models with True Concurrency and Inclusive Choices 1 Introduction 2 Background and Related Work 3 Approach 3.1 Refined Directly-Follows Graph Discovery 3.2 Refined Concurrency Discovery 3.3 Heuristic Improvement 4 Evaluation 4.1 Dataset and Setup 4.2 Results 5 Conclusion References A Novel Approach to Discover Switch Behaviours in Process Mining 1 Introduction 2 Background 3 Preliminaries 4 The Switch Process Tree 5 Discovering Switch Process Trees 5.1 The Switch Exclusive Choice Cut (Line 3) 5.2 Verifying the Exclusive Choice Switch Cut (Line 10–18) 5.3 Removing Incorrect Switch Behaviours 6 Evaluation 6.1 Evaluation Using Artificial Data 6.2 Evaluation Using Publically-Available Data 7 Discussion and Conclusion References Process Model Discovery from Sensor Event Data 1 Introduction 2 Related Work 3 Translating Sensor Data to High-Level Traces 3.1 Event Correlation 3.2 Activity Discovery 3.3 Event Abstraction 3.4 Process Discovery 4 Evaluation 4.1 Set-up 4.2 Results and Discussion 4.3 Limitations 5 Conclusion References Unsupervised Event Abstraction in a Process Mining Context: A Benchmark Study 1 Introduction 2 Related Work 3 Methodology and Experimental Setup 3.1 Evaluation Method 3.2 Data 3.3 Experimental Design 4 Empirical Results 4.1 The Effect of Abstraction on Model Complexity 4.2 The Effect of Abstraction on Model Fitness and Precision 4.3 Discussion 4.4 Limitations 5 Conclusion References 1st International Workshop on Leveraging Machine Learning in Process Mining (ML4PM) 1st International Workshop in Leveraging Machine Learning for Process Mining (ML4PM 2020) Organization Workshop Chairs Program Committee Additional Reviewers Predicting Remaining Cycle Time from Ongoing Cases: A Survival Analysis-Based Approach 1 Introduction 2 Related Work 2.1 Baseline Approach 3 Background 3.1 Events, Trace, Logs 3.2 Survival Analysis and Censored-Learning 4 Learning from Incomplete Traces 4.1 Neural Network Setup 4.2 Optimization Function 5 Evaluation 6 Conclusion References Time Matters: Time-Aware LSTMs for Predictive Business Process Monitoring 1 Introduction 2 Related Work 3 Background 3.1 Preliminaries 3.2 Long Short-Term Memory Cells 4 Methodology 4.1 Time-Aware Long Short-Term Memory Cells 4.2 Network Architecture 5 Experiments 5.1 Datasets 5.2 Preprocessing 5.3 Training Setup 5.4 Evaluation 5.5 Implementation 6 Results 6.1 Next Activity Prediction 6.2 Next Timestamp Prediction 7 Discussion 8 Conclusion and Future Work References A Preliminary Study on the Application of Reinforcement Learning for Predictive Process Monitoring 1 Introduction 2 Related Work 3 Background 4 Methodology 4.1 Pre-processing 4.2 Learning Architecture 5 Evaluation 5.1 Experimental Setup 5.2 Metrics 5.3 Results 6 Conclusions and Future Works References An Alignment Cost-Based Classification of Log Traces Using Machine-Learning 1 Introduction 2 Related Work 3 Preliminaries 3.1 Log Traces, Process Model, Fitness and Alignments 3.2 Supervised Learning from Sequences 4 Classifying Traces and Bounding the Fitness of a Model 5 Experiments 5.1 Alignment Datasets 5.2 Learning Methods 5.3 Results and Interpretation 6 Conclusion and Opening References Improving the Extraction of Process Annotations from Text with Inter-sentence Analysis 1 Introduction 2 Related Work 3 Preliminaries 3.1 Natural Language Processing 3.2 Annotated Textual Descriptions of Processes (ATDP) 3.3 TRegex 4 Generalized Approach 4.1 Basic Approach: Intra-sentence Analysis 4.2 Inter-sentence Analysis 5 Tool Support and Experiments 6 Conclusions and Future Work References Case2vec: Advances in Representation Learning for Business Processes 1 Introduction 2 Related Work 3 Case2vec 4 Experimental Evaluation 4.1 Datasets 4.2 Real-Life Event Logs: Trace Clustering 4.3 Amended Real-Life Event Logs: Event Abstraction 4.4 Synthetic Paper Process: Vector Arithmetic Interpretability 5 Discussion 5.1 Trace Clustering 5.2 Event Abstraction 5.3 Interpretability Task 6 Conclusion References Supervised Conformance Checking Using Recurrent Neural Network Classifiers 1 Introduction 2 RNN-Based Conformance Checking 2.1 Overview 2.2 Model Log Generation 2.3 Antilog Generation 2.4 Recurrent Neural Network Classifier 3 Experimental Evaluation 4 Related Work 5 Conclusion and Future Work References 1st International Workshop on Streaming Analytics for Process Mining (SA4PM’20) 1st International Workshop on Streaming Analytics for Process Mining (SA4PM) Organization Workshop Chairs Program Committee Online Anomaly Detection Using Statistical Leverage for Streaming Business Process Events 1 Introduction 2 Related Work 3 Research Framework 3.1 Anomaly Score 3.2 Online Anomaly Detection 4 Evaluation 5 Conclusions References Concept Drift Detection on Streaming Data with Dynamic Outlier Aggregation 1 Introduction and Motivation 2 Related Work 3 Preliminaries 4 Dynamic Outlier Aggregation 5 Evaluation on a Synthetic Log 5.1 Execution Times 5.2 Impact of Inter-drift Distance and Sliding Window Size 6 Evaluation on the Event Log of BPIC 2015 7 Conclusion References OTOSO: Online Trace Ordering for Structural Overviews 1 Introduction 2 Related Work 3 Preliminaries 4 OTOSO 4.1 Monitoring Temporal Deviations 4.2 Structure Analysis 5 Evaluation 5.1 Datasets 5.2 Hash Table Size 5.3 Static Clustering vs. Dynamic Clustering 5.4 OTOSO on Event Stream with Concept Drifts 6 Conclusion References Performance Skyline: Inferring Process Performance Models from Interval Events 1 Introduction 2 Related Work 3 Preliminaries 3.1 Interval Events 3.2 Skyline Operator 3.3 Geometric Interval Representation 4 Performance Models for Interval Events 4.1 Geometrical Process Representation 4.2 Performance Skyline 5 Statistical Analysis Techniques 5.1 Average Trace Skyline 5.2 Average Skyline Trace 5.3 Expected Skyline Activity Set 6 Discussion 7 Conclusion and Future Work References 5th International Workshop on Process Querying, Manipulation, and Intelligence (PQMI 2020) 5th International Workshop on Process Querying, Manipulation, and Intelligence (PQMI 2020) Organization Workshop Organizers Program Committee Alignment Approximation for Process Trees 1 Introduction 2 Related Work 3 Preliminaries 3.1 Event Logs 3.2 Process Trees 3.3 Alignments 4 Formal Framework 5 Alignment Approximation Approach 5.1 Overview 5.2 Calculation of Process Tree Characteristics 5.3 Interpretation of Process Tree Characteristics 5.4 Approximating on Choice Operator 5.5 Approximating on Sequence Operator 5.6 Approximating on Parallel Operator 5.7 Approximating on Loop Operator 6 Evaluation 7 Conclusion References Stochastic Process Discovery by Weight Estimation 1 Introduction 2 Preliminaries 3 Stochastic Process Model Weight Estimation 3.1 A Framework for GSPN Discovery 3.2 Frequency Estimator 3.3 Activity-Pair Frequency Estimators 3.4 Mean-Scaled Activity-Pair Frequency Estimator 3.5 Fork Distribution Estimator 3.6 Alignment Estimator 4 Implementation and Evaluation 4.1 Evaluation Design 4.2 Results and Discussion 5 Related Work 6 Conclusion References Graph-Based Process Mining 1 Introduction 2 Background 2.1 Process Mining 2.2 Graph Database 3 Approach 3.1 Definitions 3.2 Example 4 Implementation 5 Evaluation 5.1 Experiment 1 5.2 Experiment 2 6 Related Work and Discussion 6.1 Scalability 6.2 Graph Database 7 Conclusion References 3rd International Workshop on Process-Oriented Data Science for Healthcare (PODS4H) Third International Workshop on Process-Oriented Data Science for Healthcare (PODS4H) Organization Workshop Chairs Program Committee A Process Mining Approach to Statistical Analysis: Application to a Real-World Advanced Melanoma Dataset 1 Introduction 2 Materials and Methods 2.1 Material 2.2 Methods 3 Results 3.1 Data Preprocessing 3.2 Descriptive Statistics 3.3 Inferential Statistics 4 Discussion and Conclusion References Process Mining of Disease Trajectories in MIMIC-III: A Case Study 1 Introduction 2 Background 3 Method 3.1 Data Source for the Case Study 3.2 PM2 for Disease Trajectory Mining 4 Results 5 Discussion 6 Conclusion References The Need for Interactive Data-Driven Process Simulation in Healthcare: A Case Study 1 Introduction 2 Related Work 3 Background: Capacity Management at the Radiology Department 3.1 General Context 3.2 Data Description 4 Scenario Analysis: The Impact of Data Quality Issues 4.1 Experimental Design 4.2 Results 4.3 Discussion 5 Interactive Data-Driven Process Simulation 6 Conclusion References Process Mining on the Extended Event Log to Analyse the System Usage During Healthcare Processes (Case Study: The GP Tab Usage During Chemotherapy Treatments) 1 Introduction 2 Patient Pathways Manager (PPM) EHR System 3 Methodology 4 Results and Discussion 4.1 The Extracted Data 4.2 Discovered Process Models and the Conformance 4.3 The Enhanced Process Model 4.4 Process Analytics 4.5 Statistical and Clinical Evaluation 5 Conclusion References Process Mining on FHIR - An Open Standards-Based Process Analytics Approach for Healthcare 1 Introduction 1.1 Problem Statement 1.2 Related Work 1.3 Proposed Solution 2 Background 2.1 HL7 FHIR 2.2 XES 3 Materials and Methods 3.1 Simulate 3.2 Store and Provide 3.3 Analyze 4 Results 4.1 FHIR Resources 4.2 XES Log 4.3 Process Model 5 Discussion 5.1 Impact on Standardization 5.2 AuditEvent vs. Provenance 5.3 Considering Different Perspectives References Deriving a Sophisticated Clinical Pathway Based on Patient Conditions from Electronic Health Record Data 1 Introduction 2 Proposed Framework 2.1 Step 1: Data Preparation 2.2 Step 2: Feature Engineering 2.3 Step 3: Statistical Analysis 2.4 Step 4: Post Hoc Analysis and CP Development 3 Case Study 3.1 Introduction 3.2 Data Preparation 3.3 Feature Engineering 3.4 Statistical Analysis 3.5 Post Hoc Analysis and CP Development 4 Discussion 5 Conclusion References Exploration with Process Mining on How Temperature Change Affects Hospital Emergency Departments 1 Introduction 2 Materials and Methods 2.1 Assigning Temperature to Cases and Discretization 3 Results 3.1 Temperature and Heat Strokes 3.2 Otitis Cases Related to Temperature 3.3 Harsh Changes in Temperature and ED 4 Conclusion and Discussion References 1st International Workshop on Trust and Privacy in Process Analytics (TPPA) 1st Workshop on Trust and Privacy in Process Analytics (TPPA) Organization Organizing Committee Program Committee Towards Quantifying Privacy in Process Mining 1 Introduction 2 Related Work 3 Preliminaries 4 Privacy Quantification 4.1 Disclosure Risk 4.2 Utility Loss 5 Experiments 5.1 Disclosure Risk Analysis 5.2 Utility Loss Analysis 6 Conclusion References Author Index