A requirement-driven method for process mining based on model-driven engineering

Background: Over the last 20 years, process mining has become a vibrant research area due to the advances in data management technologies and techniques and the advent of new process mining tools. Recently, the links between process mining and simulation modelling have become an area of interest. Objectives: The objective of the paper was to demonstrate and assess the role of process mining results as an input for discrete-event simulation modelling, using two different datasets, one of which is considered data-poor while the other one data-rich. Methods/Approach: Statistical calculations and process maps were prepared and presented based on the event log data from two case studies (smart mobility and higher education) using a process mining tool. Then, the implications of the results across the building blocks (entities, activities, control-flows, and resources) of simulation modelling are discussed. Results: Apart from providing a rationale and the framework for simulation that is more efficient modelling based on process mining results, the paper provides contributions in the two case studies by deliberating and identifying potential research topics that could be tackled and supported by the new combined approach. Conclusions: Event logs and process mining provide valuable information and techniques that could be a useful input for simulation modelling, especially in the first steps of building discreteevent models, but also for validation purposes.

Process Mining helps to analyze and interpret business processes using event logs of information systems as ground truth. It has been successfully applied to process-oriented enterprise systems but is less suited for communication- and document-oriented Enterprise Collaboration Systems (ECS). This is due to two core characteristics of ECS processes and process logging. On the one hand, ECS processes are highly unstructured as their usage is not related to predefined business processes. The usage of ECS is ad-hoc and flexible. On the other hand, ECS commonly provide rather low-level process logs, i. e., logged events are very fine-granular, and, thus hard to interpret. Both characteristics lead to so-called spaghetti process models, which are barely interpretable when applying “traditional” process mining to ECS. This work focuses on the second aspect, i. e., to solve the problem of too fine-granular process logs of ECS. A common solution and preprocessing step for this granularity challenge is event abstraction, i.e., converting low-level logs into more abstract high-level logs before applying further process mining analysis techniques. ECS logs come with special characteristics that have so far not been fully addressed by existing event abstraction approaches. We aim to close this gap with a tailored ECS event abstraction (ECSEA) approach that trains a model by comparing observed actual user activities (also named high-level traces) with the system-generated low-level traces. The model allows us to automatically convert historic and future low-level traces into abstracted high-level traces that can be used for Process Mining. This article describes the ECS event log characteristics, introduces the proposed ECSEA approach with the underlying algorithms, and shows a corresponding evaluation. Our two-fold evaluation shows that the algorithm produces accurate results. In the evaluation, we use both synthetic and real-world event data taken from a productive ECS with more than 4000 users currently running at a university. • Derivation of relevant low-level event log characteristics: Based on a comparison of low-level and related high-level event logs, we derive low-level event log characteristics. They serve as requirements for a novel event abstraction approach and are further used for a comparison of related approaches. • Conceptualization of a novel supervised event abstraction approach: The main contribution is a novel event abstraction algorithm, which is described and evaluated in this article.

Most business processes are today rooted into an information system recording operational events in log files. Process Mining algorithms exploit this information to discover and qualify differences between observed and modelled process. However, the output of these algorithms are not clearly connected with business properties. Our work faces these limitations by proposing an approach for calibrating Process Mining results based on the Business Rules adopted by an organisation. The general idea relates on applying Process Mining algorithms on subsequent refinements of the event log, filtering process executions based on Business Rules. This way we are able to associate these results with specific characterisations of the process, as entailed by the corresponding Business Rules. This approach is confronted to a real world scenario using data provided by an Italian manufacturing company.

Since IoT devices supporting business processes (BPs) in sectors like manufacturing, logistics or healthcare can collect data on the execution of the processes, there is a growing awareness of the opportunity to use IoT data for process mining (PM). However, several challenges need to be addressed to enable IoT-enhanced PM, among which the need for a standard IoT-enhanced event log model. In this paper, we identify four main challenges for the integration of IoT data in event logs, from which 10 requirements for an IoT-enhanced event log model are derived. We then analyse the two main current event log models, the extensible event stream (XES) and the object-centric event log (OCEL), and confront them with the requirements, showing that both present important limitations to storing IoT-enhanced event logs. Based on this analysis, we conclude that a comprehensive data model is required to develop more advanced IoT-enhanced PM techniques .KeywordsProcess miningIoTIoT-enhanced event log

Process Mining analyzes event logs from business processes to enhance understanding, verification, and improvement of these processes. Event logs typically originate from multiple sources and require extensive preprocessing before applying process mining techniques. This preprocessing includes filtering, composing, joining events into traces, and fixing or calculating attribute values. Existing process mining tools offer ad-hoc manual actions through graphical interfaces and simple domain-specific languages, but these methods have limitations, and there is no standardized approach for processing event logs. The primary contribution of this work is the Process Query Language (PQL), a domain-specific language (DSL) for processing event logs that surpasses current techniques. We begin by analyzing several existing DSLs concerning key success factors, followed by an evaluation of current business process DSLs to identify their limitations. Based on this analysis, we propose the PQL specification as a draft technical standard. PQL’s design decisions are validated through an interpreter implementation and its application to numerous use cases.

Process mining is an emerging discipline incorporating methods and tools for the analysis of system/process executions captured in the form of event logs [1]. Traditionally process mining can be divided into three research areas: discovery (construction of process models from event logs), conformance checking (finding log and model deviations), and enhancement of existing process models with additional event log data.BPMN (Business Process Model and Notation) 2.0 [2] is a widely used process modeling notation, supported by various process modeling and analysis tools, and is a de-facto process modeling standard. Using BPMN within process mining opens perspectives for applicability of the existing process mining techniques: for instance discovered process models can be analyzed or enacted using existing BPMN-based software, and vice versa, manually created models can be imported to a process mining tool, verified against event logs, and enhanced with additional data.In this work we bridge the gap between conventional process modeling formalisms used in the context of process mining (e.g., Petri nets, causal nets, process trees) and BPMN. For that purpose we developed a suite of conversion algorithms and provide formal guarantees relating the behavior of Petri nets (including non-free-choice nets) to the corresponding BPMN models (and vice versa). The derived relations are used to enhance the BPMN models with information learned from the event logs.The developed conversion techniques are described in detail in [3] and have been implemented [4] as a part of ProM (Process Mining Framework) [5] -- an open source tool for process mining and verified on real event log data. Moreover, cases for which conversion algorithms give more compact process models in comparison with the initial models are identified. Although the developed algorithms deal with basic control flow constructs only, they can be applied in the discovery of advanced BPMN modeling elements [2], including subprocesses [6-7], cancellations [8], conditional branching and data objects, swimlanes, message flows, and others.

Towards interactive event log forensics: Detecting and quantifying timestamp imperfections

Process mining aims at gaining insights into business processes by analysing event data recorded during process execution. The majority of existing process mining techniques works offline, i.e. using static, historical data stored in event logs. Recently, the notion of online process mining has emerged, whereby techniques are applied on live event streams, as process executions unfold. Analysing event streams allows us to gain instant insights into business processes. However, current techniques assume the input stream to be completely free of noise and other anomalous behaviours. Hence, applying these techniques to real data leads to results of inferior quality. In this paper, we propose an event processor that enables us to filter out spurious events from a live event stream. Our experiments show that we are able to effectively filter out spurious events from the input stream and, as such, enhance online process mining results.

The performance of business processes is a critical success factor for companies on highly competitive markets. The improvement of business processes requires a deep understanding of as-is processes and existing weaknesses that hamper their performance. Traditionally, process mapping and weakness detection are conducted manually based on workshops and interviews with employees involved in the process. Consequently, these practices are not merely time-consuming and costly, but at the same time liable to subjective influences of the interviewed participants. For process mapping, these challenges can be overcome with the data based technology of process mining, that can model as-is-processes based on event log data from companies' information systems. For weakness detection in business processes, however, this technology is not applicable yet since this step requires domain knowledge about process weaknesses, which is not available for processing with process mining. This paper presents an approach to model domain knowledge about weaknesses in business processes to enable their automated and databased detection in event logs with process mining. Thereby, the weakness detection within business process improvement can be conducted more objectively at lower efforts.

Process mining is a technique to model and analyze business process based on traces of activities performed and stored in the database of any information systems being operated by the company i.e. event logs. This paper aims to add literature on the implementation of process mining in warehouse management process. First, extraction of all related activities from LTAK and LTAP tables of SAP Warehouse Management module is conducted. The event log are processed with Heuristic Miner Algorithm in PROM. The output of the mining is the actual business process model. The analysis of the output discover deviation from the standard procedure set by the company with the presence of additional activities and nonstandard flow of materials. The bottleneck analysis shows that the material spent a long time in the high rack and transferred between high racks. The average lead time from material receipt until issued is slightly higher than the target set the company. Further analysis of the issues revealed root-cause of the problems which are insufficient warehouse related capacity such as forklift, racks and staffs to serve incoming materials. Finally, recommendations are provided to deal with the warehouse management issues facing the case companies.

Business processes change over time in response to changing circumstances, it is crucial for process managers to discover and understand such changes from event logs. While existing techniques in process mining cannot efficiently detect and locate concept drifts in uncompleted event logs. Hence, an appropriate method, which could prompt to unravel the process evolution, may greatly assist organizations to manage the flexibility and change of business processes in the context of business process management (BMP). In this paper, we first proposed an extensible feature and employed it with the sliding window technique and heuristic miner for detecting and locating concept drifts in uncompleted event logs. Then, Genetic Process Mining (GPM) was improved by the Weight Heuristic Miner (WHM) and Differential Evolution (DE) for mining the new process model of an evolving process. Finally, experimental results on four event logs which are created based on a real-world business process have validated the proposed method.

Many present-day companies carry out a huge amount of daily operations through the use of their information systems without ever having done their own enterprise modeling. Business process mining is a well-proven solution which is used to discover the underlying business process models that are supported by existing information systems. Business process discovery techniques employ event logs as input, which are recorded by process-aware information systems. However, a wide variety of traditional information systems do not have any in-built mechanisms with which to collect events (representing the execution of business activities). Various mechanisms with which to collect events from non-process-aware information systems have been proposed in order to enable the application of process mining techniques to traditional information systems. Unfortunately, since business processes supported by traditional information systems are implicitly defined, correlating events into the appropriate process instance is not trivial. This challenge is known as the event correlation problem. This paper presents an adaptation of an existing event correlation algorithm and incorporates it into a technique in order to collect event logs from the execution of traditional information systems. The technique first instruments the source code to collect events together with some candidate correlation attributes. Based on several well-known design patterns, the technique provides a set of guidelines to support experts when instrumenting the source code. The event correlation algorithm is subsequently applied to the data set of events to discover the best correlation conditions, which are then used to create event logs. The technique has been semi-automated to facilitate its validation through an industrial case study involving a writer management system and a healthcare evaluation system. The study demonstrates that the technique is able to discover an appropriate correlation set and obtain well-formed event logs, thus enabling business process mining techniques to be applied to traditional information systems.

The aim of process mining is to identify and extract process patterns from data logs to reconstruct an overall process flowchart. As business processes become more and more complex there is a need for managers to understand the processes they already have in place. To undertake such a task manually would be extremely time consuming so the practice of process mining attempts to automatically reconstruct the correct representation of a process based on a set of process execution traces. This paper outlines an alternative approach to business process mining utilising a Genetic Programming (GP) technique coupled with a graph based representation. The graph based representation allows greater flexibility in the analysis of process flowchart structure and offers the possibility of mining complex business processes from incomplete or problematic event logs. A number of event logs have been mined by the GP technique featured in this paper and the results of the experimentation point towards the potential of this novel process mining approach.

The current business environment has no room for inefficiency as it can cause companies to lose out to their competitors, to lose customer trust, and to experience cost overruns. Business processes within the company continue to grow and cause them to run more complex. The large scale and complexity of business processes pose a challenge in improving the quality of process model because the effectiveness of time and the efficiency of existing resources are the biggest challenges. In the context of optimizing business processes with a process mining approach, most current process models are optimized with a trace clustering approach to explore the model and to perform analysis on the resulting process model. Meanwhile, in the event log data, not only the activities but also the other resources, such as records of employee or staff working time, process service time, and processing costs, are recorded. This article proposes a mechanism alternative to optimize business processes by exploring the resources that occur in the process. The mechanism is carried out in three stages. The first stage is optimizing the job shop scheduling method from the generated event log. Scheduling the time becomes a problem in the job shop. Utilizing the right time can increase the effectiveness of performance in order to reduce costs. Scheduling can be defined as the allocation of multiple jobs in a series of machines, in which each machine only does one job at a time. In general, scheduling becomes a problem when sequencing the operations and allocating them into specific time slots without prolonging the technical and capacity constraints. The second stage is generating the resource value that is recorded in the event log from the results of analysis of the previous stage, namely, job shop scheduling. The resource values are multivariate and then clustered to determine homogeneous clusters. The last stage is optimizing the nonlinear multipolynomials in the homogeneous cluster formed by using the Hessian solution. The results obtained are analyzed to get recommendations on business processes that are appropriate for the company’s needs. The impact of long waiting times will increase service costs, but by improving workload, costs can be reduced. The process model and the value of service costs resulting from the mechanism in the research can be a reference for process owners in evaluating and improving ongoing processes.

Medical billing is a long chain of processes that cut across different sectors and institutions, evolving into a multimillion-dollar industry. Despite the emergence of institutional involvement in the practice, the parties involved face many challenges that hinder the successful management of the diverse process. Process mining uses computer-based tools and techniques to analyze business processes and ensure best practices for improving business operations as it has a track record of success in improving business processes and procedures. As research on process driving problem solving continues to grow, there is a need to start applying the tools and techniques to specific domain areas and bridge the gap between research, development and application. We demonstrated how process mining techniques can be applied to medical billing using the L * life cycle model from a structured perspective. We developed a handmade comprehensive medical billing process flow using domain knowledge reference and established some time perspective Key Performance Indicators (KPIs) of the billing process. We then explored the hospital billing event log generated from an enterprise resource planning (ERP) system to see how the "as-is" from the event log deviate from the generally known knowledge-based (handmade) model. Taking each trace of the event log as patient visit, we discover that about 63 % of all the visit were billed and also about 3 % billed got regretted. The event log used in this research does not capture the complete activity flow for a real-world medical billing system. Hence, we were not able to directly compare the process map with the model and give precise operational support