Andrea Burattin

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.

PLG: A Framework for the Generation of Business Process Models and Their Execution Logs

Evaluating process mining algorithms would require the availability of a suite of real-world business processes and their execution logs, which hardly are available. In this paper we propose an approach for the random generation of business processes and their execution logs. The proposed approach is based on the generation of process descriptions via a stochastic context-free grammar whose definition is based on well-known process patterns. An algorithm for the generation of execution instances is also proposed. The implemented tools are publicly available.

Control-flow discovery from event streams

Process Mining represents an important research field that connects Business Process Modeling and Data Mining. One of the most prominent task of Process Mining is the discovery of a control-flow starting from event logs. This paper focuses on the important problem of control-flow discovery starting from a stream of event data. We propose to adapt Heuristics Miner, one of the most effective control-flow discovery algorithms, to the treatment of streams of event data. Two adaptations, based on Lossy Counting and Lossy Counting with Budget, as well as a sliding window based version of Heuristics Miner, are proposed and experimentally compared against both artificial and real streams. Experimental results show the effectiveness of control-flow discovery algorithms for streams on artificial and real datasets.More and more business activities are performed using information systems. These systems produce such huge amounts of event data that existing systems are unable to store and process them. Moreover, few processes are in steady-state and due to changing circumstances processes evolve and systems need to adapt continuously. Since conventional process discovery algorithms have been defined for batch processing, it is difficult to apply them in such evolving environments. Existing algorithms cannot cope with streaming event data and tend to generate unreliable and obsolete results. In this paper, we discuss the peculiarities of dealing with streaming event data in the context of process mining. Subsequently, we present a general framework for defining process mining algorithms in settings where it is impossible to store all events over an extended period or where processes evolve while being analyzed. We show how the Heuristics Miner, one of the most effective process discovery algorithms for practical applications, can be modified using this framework. Different stream-aware versions of the Heuristics Miner are defined and implemented in ProM. Moreover, experimental results on artificial and real logs are reported.

Techniques for a Posteriori Analysis of Declarative Processes

The increasing availability of event data recorded by information systems, electronic devices, web services and sensor networks provides detailed information about the actual processes in systems and organizations. Process mining techniques can use such event data to discover processes and check the conformance of process models. For conformance checking, we need to analyze whether the observed behavior matches the modeled behavior. In such settings, it is often desirable to specify the expected behavior in terms of a declarative process model rather than of a detailed procedural model. However, declarative models do not have an explicit notion of state, thus making it more difficult to pinpoint deviations and to explain and quantify discrepancies. This paper focuses on providing high-quality and understandable diagnostics. The notion of activation plays a key role in determining the effect of individual events on a given constraint. Using this notion, we are able to show cause-and-effect relations and measure the healthiness of the process.

Online Discovery of Declarative Process Models from Event Streams

Todays business processes are often controlled and supported by information systems. These systems record real-time information about business processes during their executions. This enables the analysis at runtime of the process behavior. However, many modern systems produce “big data”, i.e., collections of data sets so large and complex that it becomes impossible to store and process all of them. Moreover, few processes are in steady-state but, due to changing circumstances, they evolve and systems need to adapt continuously. In this paper, we present a novel framework for the discovery of LTL-based declarative process models from streaming event data in settings where it is impossible to store all events over an extended period of time or where processes evolve while being analyzed. The framework continuously updates a set of valid business constraints based on the events occurred in the event stream. In addition, our approach is able to provide meaningful information about the most significant concept drifts, i.e., changes occurring in a process during its execution. We report about experimental results obtained using synthetic logs and a real-life event log pertaining to the treatment of patients diagnosed with cancer in a large Dutch academic hospital.

Heuristics Miner for Time Intervals

Many control-flow discovery algorithms proposed up to now, assume that each activity is considered instantaneous. This is due to the fact that usually a single log for each performed activity is recorded, regardless of the duration of the activity.

Online Process Discovery to Detect Concept Drifts in LTL-Based Declarative Process Models

Today’s business processes are often controlled and supported by information systems. These systems record real-time information about business processes during their executions. This enables the analysis at runtime of the process behavior. However, many modern systems produce “big data”, i.e., collections of data sets so large and complex that it becomes impossible to store and process all of them. Moreover, few processes are in steady-state and due to changing circumstances processes evolve and systems need to adapt continuously. In this paper, we present a novel framework for the discovery of LTL-based declarative process models from streaming event data in settings where it is impossible to store all events over an extended period or where processes evolve while being analyzed. The framework continuously updates a set of valid business constraints based on the events occurred in the event stream. In addition, our approach is able to provide meaningful information about the most significant concept drifts, i.e., changes occurring in a process during its execution. We report about experimental results obtained using logs pertaining the health insurance claims handling in a travel agency.

Conformance checking based on multi-perspective declarative process models

We introduce a semantics for Multi Perspective Declare (MP-Declare).We introduce an abstract syntax for MP-Declare.We provide a set of algorithms for conformance checking based on MP-DeclareThe approach has been implemented in the process mining tool ProM.The approach has been demonstrated with real life data. Process mining is a family of techniques that aim at analyzing business process execution data recorded in event logs. Conformance checking is a branch of this discipline embracing approaches for verifying whether the behavior of a process, as recorded in a log, is in line with some expected behavior provided in the form of a process model. Recently, techniques for conformance checking based on declarative specifications have been developed. Such specifications are suitable to describe processes characterized by high variability. However, an open challenge in the context of conformance checking with declarative models is the capability of supporting multi-perspective specifications. This means that declarative models used for conformance checking should not only describe the process behavior from the control flow point of view, but also from other perspectives like data or time. In this paper, we close this gap by presenting an approach for conformance checking based on MP-Declare, a multi-perspective version of the declarative process modeling language Declare. The approach has been implemented in the process mining tool ProM and has been experimented using artificial and real-life event logs.

Time and activity sequence prediction of business process instances

The ability to know in advance the trend of running process instances, with respect to different features, such as the expected completion time, would allow business managers to timely counteract to undesired situations, in order to prevent losses. Therefore, the ability to accurately predict future features of running business process instances would be a very helpful aid when managing processes, especially under service level agreement constraints. However, making such accurate forecasts is not easy: many factors may influence the predicted features. Many approaches have been proposed to cope with this problem but, generally, they assume that the underlying process is stationary. However, in real cases this assumption is not always true. In this work we present new methods for predicting the remaining time of running cases. In particular we propose a method, assuming process stationarity, which achieves state-of-the-art performances and two other methods which are able to make predictions even with non-stationary processes. We also describe an approach able to predict the full sequence of activities that a running case is going to take. All these methods are extensively evaluated on different real case studies.

Business models enhancement through discovery of roles

Control flow discovery algorithms are able to reconstruct the workflow of a business process from a log of performed activities. These algorithms, however, do not pay attention to the reconstruction of roles, i.e. they do not group activities according to the skills required to perform them. Information about roles in business processes is commonly considered important and explicitly integrated into the process representation, e.g. as swimlanes in BPMN diagrams. This work proposes an approach to enhance a business process model with information on roles. Specifically, the identification of roles is based on the detection of handover of roles. On the basis of candidates for roles handover, the set of activities is first partitioned and then subsets of activities which are performed by the same originators are merged, so to obtain roles. All significant partitions of activities are automatically generated. Experimental results on several logs show that the set of generated roles is not too large and it always contains the correct definition of roles. We also propose an entropy based measure to rank the candidate roles which returns promising experimental results.