China
Africa
Explore chapters and articles related to this topic
Hybrid Modelling
Published in Andrew Greasley, Simulation Modelling, 2023
The use of process mining involves obtaining and extracting event data to produce an event log and transforming the event log into a process model termed process discovery. The process model can then be used to check the conformance of the system with the process design and to measure the performance of the process. In terms of event log construction, the data required to make an event log can come from a variety of sources, including collected data in spreadsheets, databases and data warehouses or directly from data streams. The minimum data required to construct an event log consists of a list of process instances (i.e. events), which are related to a case identification number and, for each event, a link to an activity label. Activities may reoccur in the event log, but each event is unique, and events within a case need to be presented in order of execution in the event log so that causal dependencies can be derived in the process model. It is also usual for there to be a timestamp associated with each event in the event log. Additional attributes associated with each event may also be included, such as the association of a resource required to undertake the event and the estimated cost of the event.
Process Mining – Prerequisites and Their Applicability for Small and Medium-sized Enterprises
Published in Pedro Novo Melo, Carolina Machado, Business Intelligence and Analytics in Small and Medium Enterprises, 2019
Alexander Zeisler, Christopher Bernhard, Julian Marius Müller
A process can be defined as a “sequence of activities performed in a specific order to achieve a specific goal” (Munoz-Gama, 2016). End-to-end processes – like order-to-cash, manufacturing processes, or service processes – are integral part of industries and professional business process management is crucial for companies who want to be competitive in an ever faster and complex economic environment. Process mining is a relatively young research discipline and is bridging the gap between process science and data science, aiming to discover, monitor, and improve real processes by using data from event logs (Van der Aalst, 2016). Today, business processes are being performed with support of IT systems to varying degrees and thus, process mining is possible due to the simple fact that data already exists (Rozinat and Günther, 2014). Information about business processes is being extracted from enterprise transaction systems and hence, information about real-life processes can be generated on the basis of data-driven facts (Davenport and Spanyi, 2019). Process mining offers an innovative approach to analyze the performance of a process. Commonly used manual tools – like spreadsheets in Excel, dashboards, or Power Point slides – are being replaced by dynamic tools. Process mining tools are visually reconstructing the actual flow of business processes, which helps to create a common understanding and process transparency among an organization. As a result, process analysis can be performed much more quickly and efficiently compared to the manual approach (Rozinat and Günther, 2014).
Towards level 3 BIM process maps with IFC & XES process mining
Published in Jan Karlshøj, Raimar Scherer, eWork and eBusiness in Architecture, Engineering and Construction, 2018
A process can be defined as a sequence of activities having a common objective (process objective). Execution of activities generate “events”; e.g. start and end of an activity are considered two events. Process Mining is the science of discovering the as-is, end-to-end processes from the analysis of recorded data from events collected and stored in “event logs”. An event log consists of “traces” (aka cases), each of which may contain one or more events for execution of activities from the beginning through the end. Event logs include instances (rows of data, i.e. events) and variables (columns of data, i.e. “features”). Variables (attributes) of executed events are also stored in event logs and could have numerical (e.g. timestamp) or categorical (ordinal or nominal) nature (e.g. resources associated with events) [3]. Fig. 1 provides a schematic hierarchical view of the event log and the associated concepts introduced. Process mining is, in fact, the analysis of event logs to learn the as happened process models (known as “process discovery”); detect where the as happened is different from as planned (known as “conformance checking”); and re-engineering and optimism business processes (known as “process enhancement”).
Event Log Privacy Based on Differential Petri Nets
Published in Applied Artificial Intelligence, 2023
Daoyu Kan, Xianwen Fang, Ziyou Gong
Table 1 shows the event log of a hospital diagnostic process. Each different kind of event trace corresponds to a different diagnostic process. In this article, we use a process mining algorithm to model this event log as a Petri net and obtain the Petri net model corresponding to the original event log. The process mining approach is a process-oriented modeling and analysis method, the idea of which is to obtain information from event logs to discover, improve or monitor actual business processes. In the past decades, many kinds of process mining algorithms have been proposed, such as -algorithm (van der Aalst, Weijters, and Maruster 2004), Inductive algorithm (Leemans, Fahland, and van der Aalst 2013), HPNs (Liu et al. 2022), MBPM (Liu 2022), etc. In this section, we use Inductive algorithm to model the event logs and apply process mining algorithms to each different kind of process variants in the event logs to obtain the corresponding Petri net models. Figure 3 shows the model for one of the classes of traces.
Using process mining to improve productivity in make-to-stock manufacturing
Published in International Journal of Production Research, 2021
Rafael Lorenz, Julian Senoner, Wilfried Sihn, Torbjørn Netland
This paper overcomes the limitations of existing methods for process mapping in manufacturing by proposing the use of process mining. Process mining is a recent development in information systems research that models process flows based on event log data (van der Aalst 2016). In contrast to manual process mapping, process mining allows analysing process flows dynamically and identifying non-value-adding activities in an automated manner (Schuh et al. 2020b). For this reason, process mining is particularly suitable for discovering process deviations and identifying factors that negatively affect productivity. Despite the promising opportunities process mining offers, there is scarce research addressing productivity improvement in manufacturing. This paper addresses this gap in the literature by proposing and validating a procedure for using process mining in make-to-stock manufacturing.
Towards data-driven approaches in manufacturing: an architecture to collect sequences of operations
Published in International Journal of Production Research, 2020
Ashfaq Farooqui, Kristofer Bengtsson, Petter Falkman, Martin Fabian
Process mining is a field of study that deals with process discovery, conformance checking and enhancement using logged data from the system (van der Aalst 2011). By looking at sufficiently large logs of labelled data, spanning over several cycles, it is possible to discover an accurate representation of the process – this is called process discovery. The obtained model can then be checked against other event logs from the same process to find deviations – this is called conformance checking. Furthermore, obtained models can also be updated or extended using new event logs, and this process is called enhancement. The output model can be in the form of a Petri net, a transition graph, or a Business Process Model and Notation graph. Process mining has shown significant benefit in understanding underlying task flows, bottlenecks, resource utilisation and many other factors within large corporations (Van Der Aalst et al. 2007; van der Aalst 2013), and also proved beneficial in healthcare (Mans et al. 2008; Partington et al. 2015; Rojas et al. 2016) to learn and improve the underlying process.