Jorge Munoz-Gama

Process Mining Manifesto

Process mining techniques are able to extract knowledge from event logs commonly available in today’s information systems. These techniques provide new means to discover, monitor, and improve processes in a variety of application domains. There are two main drivers for the growing interest in process mining. On the one hand, more and more events are being recorded, thus, providing detailed information about the history of processes. On the other hand, there is a need to improve and support business processes in competitive and rapidly changing environments. This manifesto is created by the IEEE Task Force on Process Mining and aims to promote the topic of process mining. Moreover, by defining a set of guiding principles and listing important challenges, this manifesto hopes to serve as a guide for software developers, scientists, consultants, business managers, and end-users. The goal is to increase the maturity of process mining as a new tool to improve the (re)design, control, and support of operational business processes.

A fresh look at precision in process conformance

Process Conformance is a crucial step in the area of Process Mining: the adequacy of a model derived from applying a discovery algorithm to a log must be certified before making further decisions that affect the system under consideration. Among the different conformance dimensions, in this paper we propose a novel measure for precision, based on the simple idea of counting these situations were the model deviates from the log. Moreover, a log-based traversal of the model that avoids inspecting its whole behavior is presented. Experimental results show a significant improvement when compared to current approaches for the same task. Finally, the detection of the shortest traces in the model that lead to discrepancies is presented.

Alignment based precision checking

Most organizations have process models describing how cases need to be handled. In fact, legislation and standardization (cf. the Sarbanes-Oxley Act, the Basel II Accord, and the ISO 9000 family of standards) are forcing organizations to document their processes. These processes are often not enforced by information systems. However, torrents of event data are recorded by today’s information systems. These recorded events reflect how processes are really executed. Often reality deviates from the modeled behavior. Therefore, measuring the extent process executions conform to a predefined process model is increasingly important. In this paper, we propose an approach to measure the precision of a process model with respect to an event log. Unlike earlier approaches, we first align model and log thus making our approach more robust, even in case of deviations. The approach has been implemented in the ProM 6 tool and evaluated using both artificial and real life cases.

Enhancing precision in Process Conformance: Stability, confidence and severity

Process Conformance is becoming a crucial area due to the changing nature of processes within an Information System. By confronting specifications against system executions (the main problem tackled in process conformance), both system bugs and obsolete/incorrect specifications can be revealed. This paper presents novel techniques to enrich the process conformance analysis for the precision dimension. The new features of the metric proposed in this paper provides a complete view of the precision between a log and a model. The techniques have been implemented as a plug-in in an open-source Process Mining platform and experimental results witnessing both the theory and the goals of this work are presented.

Single-Entry Single-Exit decomposed conformance checking

An exponential growth of event data can be witnessed across all industries. Devices connected to the internet (internet of things), social interaction, mobile computing, and cloud computing provide new sources of event data and this trend will continue. The omnipresence of large amounts of event data is an important enabler for process mining. Process mining techniques can be used to discover, monitor and improve real processes by extracting knowledge from observed behavior. However, unprecedented volumes of event data also provide new challenges and often state-of-the-art process mining techniques cannot cope. This paper focuses on conformance checking in the large and presents a novel decomposition technique that partitions larger process models and event logs into smaller parts that can be analyzed independently. The so-called Single-Entry Single-Exit (SESE) decomposition not only helps to speed up conformance checking, but also provides improved diagnostics. The analyst can zoom in on the problematic parts of the process. Importantly, the conditions under which the conformance of the whole can be assessed by verifying the conformance of the SESE parts are described, which enables the decomposition and distribution of large conformance checking problems. All the techniques have been implemented in ProM, and experimental results are provided.

Conformance checking in the large: partitioning and topology

The torrents of event data generated by todays systems are an important enabler for process mining. However, at the same time, the size and variability of the resulting event logs are challenging for todays process mining techniques. This paper focuses on conformance checking in the large and presents a novel decomposition technique that partitions larger processes into sets of subprocesses that can be analyzed more easily. The resulting topological representation of the partitioning can be used to localize conformance problems. Moreover, we provide techniques to refine the decomposition such that similar process fragments are not considered twice during conformance analysis. All the techniques have been implemented in ProM, and experimental results are provided.

Hierarchical conformance checking of process models based on event logs

Process mining techniques aim to extract knowledge from event logs. Conformance checking is one of the hard problems in process mining: it aims to diagnose and quantify the mismatch between observed and modeled behavior. Precise conformance checking implies solving complex optimization problems and is therefore computationally challenging for real-life event logs. In this paper a technique to apply hierarchical conformance checking is presented, based on a state-of-the-art algorithm for deriving the subprocesses structure underlying a process model. Hierarchical conformance checking allows us to decompose problems that would otherwise be intractable. Moreover, users can navigate through conformance results and zoom into parts of the model that have a poor conformance. The technique has been implemented as a ProM plugin and an experimental evaluation showing the significance of the approach is provided.

Event-Based Real-Time Decomposed Conformance Analysis

Process mining deals with the extraction of knowledge from event logs. One important task within this research field is denoted as conformance checking, which aims to diagnose deviations and discrepancies between modeled behavior and real-life, observed behavior. Conformance checking techniques still face some challenges, among which scalability, timeliness and traceability issues. In this paper, we propose a novel conformance analysis methodology to support the real-time monitoring of event-based data streams, which is shown to be more efficient than related approaches and able to localize deviations in a more fine-grained manner. Our developed approach can be directly applied in business process contexts where rapid reaction times are crucial; an exhaustive case example is provided to evidence the validity of the approach.

Decomposing Alignment-Based Conformance Checking of Data-Aware Process Models

Process mining techniques relate observed behavior to modeled behavior, e.g., the automatic discovery of a Petri net based on an event log. Process mining is not limited to process discovery and also includes conformance checking. Conformance checking techniques are used for evaluating the quality of discovered process models and to diagnose deviations from some normative model (e.g., to check compliance). Existing conformance checking approaches typically focus on the control-flow, thus being unable to diagnose deviations concerning data. This paper proposes a technique to check the conformance of data-aware process models. We use so-called Petri nets with Data to model data variables, guards, and read/write actions. Data-aware conformance checking problem may be very time consuming and sometimes even intractable when there are many transitions and data variables. Therefore, we propose a technique to decompose large data-aware conformance checking problems into smaller problems that can be solved more efficiently. We provide a general correctness result showing that decomposition does not influence the outcome of conformance checking. The approach is supported through ProM plug-ins and experimental results show significant performance improvements. Experiments have also been conducted with a real-life case study, thus showing that the approach is also relevant in real business settings.