Predictive End-to-End Enterprise Process Network Monitoring

2023 | Business & Information Systems Engineering | Citations: 0

Authors: Oberdorf, Felix; Schaschek, Myriam; Weinzierl, Sven; Stein, Nikolai; Matzner, Martin; Flath, Christoph M.

Abstract: Ever-growing data availability combined with rapid progress in analytics has lai ... Expand

Abstract: Ever-growing data availability combined with rapid progress in analytics has laid the foundation for the emergence of business process analytics. Organizations strive to leverage predictive process analytics to obtain insights. However, current implementations are designed to deal with homogeneous data. Consequently, there is limited practical use in an organization with heterogeneous data sources. The paper proposes a method for predictive end-to-end enterprise process network monitoring leveraging multi-headed deep neural networks to overcome this limitation. A case study performed with a medium-sized German manufacturing company highlights the method’s utility for organizations. Collapse

Topics: logistics management computer hardware workstation organizational context supply chain management

Methods: event log machine learning case study longitudinal research deep learning

Why Work Around the Process? Analyzing Workarounds Through the Lens of the Theory of Planned Behavior

2023 | Business & Information Systems Engineering | Citations: 1

Authors: Soffer, Pnina; Outmazgin, Nesi; Hadar, Irit; Tzafrir, Shay

Abstract: Workarounds are a common practice in a broad range of organizational and technol ... Expand

Abstract: Workarounds are a common practice in a broad range of organizational and technological contexts, which has received much attention in information systems research. They are sometimes considered negatively, associated with business risks and noncompliance, and sometimes positively, as a source of innovation and improvement. Although workaround-based process improvements typically adopt the workaround if its consequences are not harmful, this is not always the best option. The paper aims to pave the way for leveraging workarounds for process improvement by understanding problems that motivate them, so alternative solutions can be developed for these problems. Despite many proposed explanations, an in-depth understanding of the reasons underlying employee decisions that lead to workarounds is still needed. To this end, following a qualitative research approach, data were collected in six case study organizations, investigating workaround decisions made by employees through the lens of the theory of planned behavior. The principal unique theme running throughout the findings is that workaround motivation stems from perceived misalignment between organizational, local unit, or personal goals and their realization in business processes. Workarounds are enabled by various factors, including perceived workaround opportunities in the organization. These findings are abstracted to form an explanatory model as a contextual adaptation of the theory of planned behavior to explain workaround intentions. The proposed model explains workaround decisions in a way that can leverage this understanding to promote concrete process improvements. Collapse

Topics: business process improvement process mining business process management organizational context data security

Methods: qualitative interview case study modeling language business process model and notation triangulation

Theories: theory of workarounds theory of planned behavior technological frames of reference institutional theory activity theory

Creating business value with process mining

2022 | Journal of Strategic Information Systems | Citations: 0

Authors: Badakhshan, Peyman; Wurm, Bastian; Grisold, Thomas; Geyer-Klingeberg, Jerome; Mendling, Jan; vom Brocke, Jan

Abstract: Information systems research has a long-standing interest in how organizations g ... Expand

Abstract: Information systems research has a long-standing interest in how organizations gain value through information technology. In this article, we investigate a business process intelligence (BPI) technology that is receiving increasing interest in research and practice: process mining. Process mining uses digital trace data to visualize and measure the performance of business processes in order to inform managerial actions. While process mining has received tremendous uptake in practice, it is unknown how organizations use it to generate business value. We present the results of a multiple case study with key stakeholders from eight internationally operating companies. We identify key features of process mining – data & connectivity, process visualiza­ tion, and process analytics – and show how they translate into a set of affordances that enable value creation. Specifically, process mining affords (1) perceiving end-to-end process visualiza­ tions and performance indicators, (2) sense-making of process-related information, (3) datadriven decision making, and (4) implementing interventions. Value is realized, in turn, in the form of process efficiency, monetary gains, and non-monetary gains, such as customer satisfac­ tion. Our findings have implications for the discourse on IT value creation as we show how process mining constitutes a new class of business intelligence & analytics (BI&A) technology, that enables behavioral visibility and allows organizations to make evidence-based decisions about their business processes. Collapse

Topics: process mining value creation organizational value analytical information system business process management

Methods: qualitative interview multiple case study qualitative coding personal interview grounded theory

Theories: theory of affordance

Semi-Supervised Discovery of DNN-Based Outcome Predictors from Scarcely-Labeled Process Logs

2022 | Business & Information Systems Engineering | Citations: 0

Authors: Folino, Francesco; Folino, Gianluigi; Guarascio, Massimo; Pontieri, Luigi

Abstract: Predicting the final outcome of an ongoing process instance is a key problem in ... Expand

Abstract: Predicting the final outcome of an ongoing process instance is a key problem in many real-life contexts. This problem has been addressed mainly by discovering a prediction model by using traditional machine learning methods and, more recently, deep learning methods, exploiting the supervision coming from outcome-class labels associated with historical log traces. However, a supervised learning strategy is unsuitable for important application scenarios where the outcome labels are known only for a small fraction of log traces. In order to address these challenging scenarios, a semi-supervised learning approach is proposed here, which leverages a multi-target DNN model supporting both outcome prediction and the additional auxiliary task of next-activity prediction. The latter task helps the DNN model avoid spurious trace embeddings and overfitting behaviors. In extensive experimentation, this approach is shown to outperform both fully-supervised and semi-supervised discovery methods using similar DNN architectures across different real-life datasets and label-scarce settings. Collapse

Topics: application programming interface Python business process management data breach database system

Methods: machine learning semi-supervised learning experiment computational algorithm statistical hypothesis test

Multi-level Team Assignment in Social Business Processes: An Algorithm and Simulation Study

2022 | Information Systems Frontiers | Citations: 0

Authors: Liu, Rong; Kumar, Akhil; Lee, Juhnyoung

Abstract: Business Process Management (BPM) systems usually neglect the human and social a ... Expand

Abstract: Business Process Management (BPM) systems usually neglect the human and social aspects (or team effects) involved in business process execution. Our work fills a large gap in literature by addressing multi-level teams that arise in business processes where teams are formed at both the task and process levels. In this paper, we develop a methodology called BPMTeams based on social network analysis for building an execution model for a social BPM. This model is used to make resource assignments to form dynamic teams that perform various team-based activities in a process. We further develop various resource assignment strategies and evaluate them using parameters estimated from a real data set in the IT incident management domain to understand how team effects play out in social business processes. The overall team effect in a process is analyzed at two levels: as a task team effect where the synergistic role of a team in a specific task is realized; and a process team effect that arises from inter-team synergies across the individual task teams in a process. The results offer some balanced insights for the interplay of these effects by highlighting the benefits and disadvantages of teams selected by a purely data-driven approach. Collapse

Topics: social network business process management resource allocation Python task complexity

Methods: simulation computational algorithm experiment design science process log

OrgMiner: A Framework for Discovering User-Related Process Intelligence from Event Logs

2021 | Information Systems Frontiers | Citations: 0

Authors: Deokar, Amit V.; Tao, Jie

Abstract: Process Intelligence refers to the extraction and analysis of valuable knowledge ... Expand

Abstract: Process Intelligence refers to the extraction and analysis of valuable knowledge nuggets embedded in business process instances/event logs or enterprise applications, for the purpose of supporting various decision-making processes. Researchers and practitioners mine such event logs using Process Mining and Analytics (PMA) techniques that help analyze business processes across three perspectives: control flow, organization, and data. While previous PMA studies have made advances toward the control flow and data flow perspectives, there is limited research toward the organizational perspective of process intelligence. In this study, we propose an organizational mining framework, OrgMiner, that supports constructing organizational models from event logs. The framework utilizes the notion of behavioral patterns, which rely on the weak order relations appearing in event logs. The various modules and knowledge elements in the framework are described in detail. The components of the framework support identifying, selecting, and applying behavioral patterns using different metrics for organizational mining purposes. The derived organizational models can be used to support decision making in scenarios such as task assignment, resource allocation, as well as role-based access control. Compared to extant studies, the proposed approach does not assume prior availability of explicit process models. Additionally, the process patterns presented in this study can be used as building blocks, so that researchers and practitioners can use them directly or extend them further to identify complex organizational processes. A case study is presented to evaluate the feasibility and effectiveness of the OrgMiner framework. Collapse

Topics: resource allocation business process management organization structure mobile system internet of things

Methods: event log computational algorithm case study business process modeling cluster analysis

Neural Process Mining: Multi-Headed Predictive Process Analytics in Practice

2021 | European Conference On Information Systems | Citations: 0

Authors: Oberdorf, Felix; Schaschek, Myriam; Stein, Nikolai; Flath, Christoph

Abstract: Ever growing data availability combined with rapid progress in the field of ana ... Expand

Abstract: Ever growing data availability combined with rapid progress in the field of analytics has laid the foundation for the emergence of Business Process Management in general and Business Process Analytics in particular. Going beyond descriptive process log analysis, manufacturing companies strive to leverage predictive process analytics to generate process-related predictions. However, current implementations are limited in their practical use as they are not able to combine multiple heterogeneous data sources without loss of information. To overcome this limitation, we propose a deep learning based approach leveraging multi-headed neural networks. We develop and evaluate the proposed approach in the context of a real-world disruption type classification showcase. To this end, we cooperate with a medium-sized German manufacturing company and observe promising results in an initial evaluation. Collapse

Topics: process mining business process management

Methods: artificial neural network process log computational algorithm deep learning machine learning

DeepRan: Attention-based BiLSTM and CRF for Ransomware Early Detection and Classification

2021 | Information Systems Frontiers | Citations: 0

Authors: Roy, Krishna Chandra; Chen, Qian

Abstract: Ransomware is a self-propagating malware encrypting file systems of the compromi ... Expand

Abstract: Ransomware is a self-propagating malware encrypting file systems of the compromised computers to extort victims for financial gains. Hundreds of schools, hospitals, and local government municipalities have been disrupted by ransomware that already caused 12.1 days of system downtime on average (Siegel 2019). This study aims at developing a deep learning-based detector DeepRan for ransomware early detection and classification to prevent network-wide data encryption. DeepRan applies an attention-based bi-directional Long Short Term Memory (BiLSTM) with a fully connected (FC) layer to model normalcy of hosts in an operational enterprise system and detects abnormal activity from a large volume of ambient host logging data collected from bare metal servers. DeepRan also classifies abnormal activity as one of the candidate ransomware attacks by extending attention-based BiLSTM with a Conditional Random Fields (CRF) model. The Term Frequency-Inverse Document Frequency (TF-IDF) method is applied to extract semantic information from high dimensional host logging data. An incremental learning technique is used to extend the model’s existing knowledge to prevent DeepRan quality degradation over time. We develop a testbed of bare metal servers and collect normal host logs of two users for 63 days (IRB-approved). 17 ransomware attacks are executed on the victim hosts, and the infected host logging data is used for validating DeepRan. Experimental results present that DeepRan produces 99.87% detection accuracy (F1-score of 99.02%) for ransomware early detection. The detector also achieves 96.5% accuracy to classify abnormal events as one of 17 candidate ransomware families. The application of incremental learning is validated as an efficient technique to enhance model quality over time. Collapse

Topics: ransomware malware data encryption Linux Microsoft Windows

Methods: machine learning deep learning term frequency–inverse document frequency experiment anomaly detection

A text analytics framework for automated communication pattern analysis

2020 | Information & Management | Citations: 4

Authors: Fan, Shaokun; Ilk, Noyan

Abstract: Human-coding reliant conversation analysis methods are ineffective when analyzin ... Expand

Abstract: Human-coding reliant conversation analysis methods are ineffective when analyzing large volumes of data. In this paper, we propose a text analytics framework for automated conversation pattern analysis. This framework first extracts speech acts (i.e., activities) from conversation logs, and then analyzes their flow through frequent pattern mining algorithms to reveal insightful communication patterns. Using a real-world data set collected from a customer service center, we demonstrate the usefulness of the framework for identifying patterns that are associated with service quality outcomes. Our work has implications for the design of communication policies and systems for customer service management. Collapse

Topics: customer satisfaction customer service process mining knowledge sharing service quality

Methods: computational algorithm case study qualitative content analysis ensemble learning machine learning

Theories: speech act theory linguistics theory

A Novel Business Process Prediction Model Using a Deep Learning Method

2020 | Business & Information Systems Engineering | Citations: 6

Authors: Mehdiyev, Nijat; Evermann, Joerg; Fettke, Peter

Abstract: The ability to proactively monitor business processes is a main competitive diff ... Expand

Abstract: The ability to proactively monitor business processes is a main competitive differentiator for firms. Process execution logs generated by process aware information systems help to make process specific predictions for enabling a proactive situational awareness. The goal of the proposed approach is to predict the next process event from the completed activities of the running process instance, based on the execution log data from previously completed process instances. By predicting process events, companies can initiate timely interventions to address undesired deviations from the desired workflow. The paper proposes a multi-stage deep learning approach that formulates the next event prediction problem as a classification problem. Following a feature pre-processing stage with n-grams and feature hashing, a deep learning model consisting of an unsupervised pre-training component with stacked autoencoders and a supervised fine-tuning component is applied. Experiments on a variety of business process log datasets show that the multi-stage deep learning approach provides promising results. The study also compared the results to existing deep recurrent neural networks and conventional classification approaches. Furthermore, the paper addresses the identification of suitable hyperparameters for the proposed approach, and the handling of the imbalanced nature of business process event datasets. Collapse

Topics: business process management open source Microsoft Azure Java information retrieval

Methods: deep learning experiment machine learning autoencoder classification method