Business Process Monitoring Research Articles

Evaluating early predictive performance of machine learning approaches for engineering change schedule – A case study using predictive process monitoring techniques

By applying machine learning algorithms, predictive business process monitoring (PBPM) techniques provide an opportunity to counteract undesired outcomes of processes. An especially complex variation of business processes is the engineering change (EC) process. Here, failing to adhere to planned implementation dates can have severe impacts on assembly lines, and it is paramount that potential negative cases are identified as early as possible. Current PBPM research, however, has seldomly investigated the predictive performance of machine learning approaches and their applicability at early process steps, let alone for the EC process. In our research, we show that given adequate feature encoding, shallow learners can accurately predict schedule adherence after process initialisation. Based on EC data from an automotive manufacturer, we provide a case sensitive performance overview on algorithm-encoding combinations. For that, three algorithms (XGBoost, Random Forest, LSTM) were combined with four encoding techniques. The encoding techniques used were the two common aggregation-based and index-based last state encoding, and two new combinations of these, which we term advanced aggregation-based and complex aggregation-based encoding. The study indicates that XGBoost-index-encoded approaches outclass regarding average predictive performance, whereas Random-Forest-aggregation-encoded approaches perform better regarding temporal stability due to reduced influence by dynamic features. Our research provides a case-based reasoning approach for deciding on which algorithm-encoding combination and evaluation metrics to apply. In doing so, we provide a blueprint for an early warning and monitoring method within the EC process and other similarly complex processes.

Causal Learning: Monitoring Business Processes Based on Causal Structures.

Conventional methods for process monitoring often fail to capture the causal relationships that drive outcomes, making hard to distinguish causal anomalies from mere correlations in activity flows. Hence, there is a need for approaches that allow causal interpretation of atypical scenarios (anomalies), allowing to identify the influence of operational variables on these anomalies. This article introduces (CaProM), an innovative technique based on causality techniques, applied during the planning phase in business process environments. The technique combines two causal perspectives: anomaly attribution and distribution change attribution. It has three stages: (1) process events are collected and recorded, identifying flow instances; (2) causal learning of process activities, building a directed acyclic graphs (DAGs) represent dependencies among variables; and (3) use of DAGs to monitor the process, detecting anomalies and critical nodes. The technique was validated with a industry dataset from the banking sector, comprising 562 activity flow plans. The study monitored causal structures during the planning and execution stages, and allowed to identify the main factor behind a major deviation from planned values. This work contributes to business process monitoring by introducing a causal approach that enhances both the interpretability and explainability of anomalies. The technique allows to understand which specific variables have caused an atypical scenario, providing a clear view of the causal relationships within processes and ensuring greater accuracy in decision-making. This causal analysis employs cross-sectional data, avoiding the need to average multiple time instances and reducing potential biases, and unlike time series methods, it preserves the relationships among variables.

LS-ICE: A Load State Intercase Encoding framework for improved predictive monitoring of business processes

Research on developing techniques for predictive process monitoring has generally relied on feature encoding schemes that extract intra-case features from events to make predictions. In doing so, the processing of cases is assumed to be solely influenced by the attributes of the cases themselves. However, cases are not processed in isolation and can be influenced by the processing of other cases or, more generally, the state of the process under investigation. In this work, we propose the LS-ICE (load state intercase encoding) framework for encoding intercase features that enriches events with a depiction of the state of relevant load points in a business process. To assess the benefits of the intercase features generated using the LS-ICE framework, we compare the performance of predictive process monitoring models constructed using the encoded features against baseline models without these features. The models are evaluated for remaining trace and runtime prediction using five real-life event logs. Across the board, a consistent improvement in performance is noted for models that integrate intercase features encoded through the proposed framework, as opposed to baseline models that lack these encoded features.

Methods of constructing integral indicators for assessing entrepreneurial potential of the rural population

Purpose - methods of analyzing quantitative and qualitative indicators of the research results on entrepreneurial potential in rural areas, formation of methodological tools for integral assessment of the state of business activity in the regions of Kazakhstan are considered. Methods - statistical, ranking, rating system - reveal the main factors affecting the business initiative of entrepreneurs, contribute to the determination of indices of entrepreneurial activity, which makes it possible to achieve specific objective decisions. Results - it is shown that these regulatory materials comprehensively assess the state of small and medium-sized agribusiness in the republic and can be used in the development and adjustment of government programs, national projects, and further scientific research, taking into account trends in innovation and environmental friendliness of agro-industrial production. Conclusions - development of entrepreneurship in agricultural sector determines economic and social situation, creates favorable conditions for production of competitive agricultural products and food. Small and medium-sized businesses are able to raise the status of social sustainability of the village and support diversification. Effective measures are needed to effectively manage business risks. The necessity of forming optimal parameters when monitoring business processes at enterprises is substantiated as a basis for solving such problem as identifying their potential resource capabilities and using them in the future. The purpose of this approach is to timely identify “critical points” and apply modern methods to improve the efficiency of small agricultural businesses. The proposed methodology is especially important in the field of strategic planning and forecasting, for the implementation of results-oriented management procedures, program-target budgeting when implementing large innovative projects in the regions.

Optimizing Business Processes through Information System Integration A Case Study in Banda Aceh

This research aims to optimize business processes through information system integration using a case study in Banda Aceh. This research uses a qualitative approach with a single case study research method. Data was collected through in-depth interviews, direct observation and documentation. Research respondents consist of managers and employees involved in the business processes to be integrated. Data analysis was carried out using content analysis. The research results show that information system integration significantly increases the efficiency and effectiveness of business processes in Banda Aceh. Information system integration enables faster data processing, better monitoring of business processes, and more accurate decision making. Additionally, information system integration also enables better collaboration between different departments and stakeholders. Based on these findings, it can be concluded that information system integration plays an important role in optimizing business processes in Banda Aceh. By integrating information systems, companies can increase operational efficiency, reduce costs, and increase customer satisfaction. However, the implementation of information systems integration also faces challenges, such as technical difficulties and changes in organizational culture. Therefore, strong commitment and support from company management is needed to achieve optimal results.

Confidential Transaction Balance Verification by the Net Using Non-Interactive Zero-Knowledge Proofs

One of the main trends for the monitoring and control of business processes is to implement these processes via private blockchain systems. These systems must ensure data privacy and verifiability for the entire network here denoted by ‘Net’. In addition, every business activity should be declared to a trusted third party (TTP), such as an Audit Authority (AA), for tax declaration and collection purposes. We present a solution for a confidential and verifiable realization of transactions based on the Unspent Transaction Output (UTxO) paradigm. This means that the total sum of transaction inputs (incomes) $In$ must be equal to the total sum of transaction outputs (expenses) $Ex$, satisfying the balance equation $In=Ex$. Privacy in a private blockchain must be achieved through the encryption of actual transaction values. However, it is crucial that all participants in the network be able to verify the validity of the transaction balance equation. This poses a challenge with probabilistically encrypted data. Moreover, the inputs and outputs are encrypted with different public keys. With the introduction of the AA, the number of different public keys for encryption can be reduced to two. Incomes are encrypted with the Receiver’s public key and expenses with the AA’s public key. The novelty of our realization lies in taking additively-multiplicative, homomorphic ElGamal encryption and integrating it with a proposed paradigm of modified Schnorr identification providing a non-interactive zero-knowledge proof (NIZKP) using a cryptographically secure h-function. Introducing the AA as a structural element in a blockchain system based on the UTxO enables effective verification of encrypted transaction data for the Net. This is possible because the proposed NIZKP is able to prove the equivalency of two ciphertexts encrypted with two different public keys and different actors. This integration allows all users on the Net to check the UTxO-based transaction balance equation on encrypted data. The security considerations of the proposed solution are presented.

Improving Comprehensibility of IoT-Driven Business Processes through BPMN 2.0 Extension: A User Study

Enhancing business processes with IoT capabilities enables the collection of data, which can be leveraged to optimize and automate business processes. As a prerequisite for exploiting these benefits in the context of BPM, IoT aspects needs to be explicitly captured in business models. Modeling IoT-driven business processes with BPMN 2.0, however, might be hindered by the ambiguous use of the modeling elements. To tackle this challenge, we have proposed a BPMN 2.0 extension for IoT (BPMNE4IoT) that incorporates specific IoT modeling elements. BPMNE4IoT enables the modeling, execution, monitoring, and recording of IoT-driven business processes. In general, the effectiveness of any BPMN 2.0 extensions depend on its comprehensibility by BPM practitioners. Consequently, user studies are required, which investigate the effects of IoT-specific elements on process model readers. This paper presents the results of such a user study (N=249), which aims to determine whether IoT involvement becomes visually better apparent in our language extension BPMNE4IoT compared to standard BPMN 2.0. In particular, the user study confirms that the involvement of IoT aspects in BPMNE4IoT is more discernible for both experts and non-experts compared to the BPMN 2.0 standard.

Embedding Graph Convolutional Networks in Recurrent Neural Networks for Predictive Monitoring

Predictive monitoring of business processes is a subfield of process mining that aims to predict, among other things, the characteristics of the next event or the sequence of the next events. Although multiple approaches based on deep learning have been proposed, mainly recurrent neural networks and convolutional neural networks, none of them really exploit the structural information available in process models. This paper proposes an approach that simultaneously learns spatio-temporal information from both the event log and the process model by combining recurrent neural networks with graph convolutional networks. Thus, common patterns from process models, such as loops or parallels, can be learned while avoiding overwriting information during the encoding phase. An experimental evaluation of real-life event logs shows that our approach is more consistent and outperforms the current state-of-the-art approaches.

Design and Build a Web Service for Financial Technology Dashboard at PT. Mitra Kasih Perkasa with React JS

PT Mitra Kasih Perkasa or MKP is a financial technology or fintech company, as well as a system integrator. MKP processes thousands, even millions, of financial transaction data every day, requiring efficient data management through the use of a dashboard. Information from the dashboard is used to monitor business processes and improve company performance. This research focuses on the development of efficient and effective web services for financial technology dashboards, integrating attractive and user-friendly user interfaces (UI/UX). Additionally, this research will discuss how to access web services using React JS, a popular JavaScript framework for responsive user interface development. The research will also outline the steps in building a financial technology dashboard that can help PT. Mitra Kasih Perkasa efficiently and effectively manage financial information using modern and innovative React JS technology.

From process mining to augmented process execution

Business process management (BPM) is a well-established discipline comprising a set of principles, methods, techniques, and tools to continuously improve the performance of business processes. Traditionally, most BPM decisions and activities are undertaken by business stakeholders based on manual data collection and analysis techniques. This is time-consuming and potentially leads to suboptimal decisions, as only a restricted subset of data and options are considered. Over the past decades, a rich set of data-driven techniques has emerged to support and automate various activities and decisions across the BPM lifecycle, particularly within the process mining field. More recently, the uptake of artificial intelligence (AI) methods for BPM has led to a range of approaches for proactive business process monitoring. Given their common data requirements and overlapping goals, process mining and AI-driven approaches to business process optimization are converging. This convergence is leading to a promising emerging concept, which we call (AI-)augmented process execution: a collection of data analytics and artificial intelligence methods for continuous and automated improvement and adaptation of business processes. This article gives an outline of research at the intersection between process mining and AI-driven process optimization, classifies the researched techniques based on their scope and objectives, and positions augmented process execution as an additional layer on top of this stack.

Using blockchain technology for monitoring business processes in ERP systems

Database protection is one of the most difficult tasks facing the departments responsible for ensuring information security. On the one hand, to work with the database, it is necessary to provide access to data to all employees who, on duty, must collect, process, store and transfer confidential data. On the other hand, the expansion of databases does not always have a centralized architecture, and therefore the actions of violators are becoming more sophisticated. Thus, there is no clear and precise methodology for a comprehensive solution to the problem of protecting databases that could be applied in all cases; in each specific situation, an individual approach has to be found. Blockchain technology can solve some of these problems. The purpose of the study was to show the possibilities of blockchain technology for tracking possible external and internal attacks on Control Systems and their databases. It is proposed to combine and blockchain-based hashing methods in alphanumeric format and explore its application in tracking and monitoring the sequence of business process transactions in ERP systems. A cyber attack on transaction data in ERP systems will immediately affect the root of the Merkle tree, which can serve as a signal that the system has been hacked. We show that blockchain, considered one of the most disruptive technologies in various industries, certainly has the potential to apply and ensure the security of enterprise management systems in the ERP format.

A framework for modeling, executing, and monitoring hybrid multi-process specifications with bounded global–local memory

So far, approaches for business process modeling, enactment and monitoring have mainly been based on process specifications consisting of a single process model. This setting aptly captures monolithic scenarios from domains in which all possible behaviors can be folded into a single model. However, the same strategy cannot be applied to domains where multiple interacting (procedural) processes simultaneously work over the same objects, in the presence of additional (declarative) constraints relating activities from the same or different processes. A relevant example for this setting is that of healthcare, where co-morbid patients may be subject to multiple clinical pathways at once, in the presence of additional, general constraints capturing basic medical knowledge. To fill this gap, we have previously presented the M3 Framework and an accompanying monitoring technique, which allows for a hybrid representation of a process using both procedural and declarative models, and supports the modular creation of multi-process specifications where domain experts can focus on specific procedures and domain constraints without being forced to merge them into one single specification. In this paper, we make significant extensions to this framework, allowing us to go from simple toy examples towards addressing practical real-life scenarios. We achieve this by introducing a richer form of integration between the interacting process components, in particular supporting asynchronous and synchronous activities that may operate over local and global (shared) data variables. This is framed by a discussion of the business meaning of these concepts, the introduction of the corresponding modeling patterns, and the application of our approach to real-life business processes, the latter being the driving-force behind this paper.

Modelling and Predictive Monitoring of Business Processes under Uncertainty with Reinforcement Learning.

The analysis of business processes based on their observed behavior recorded in event logs can be performed with process mining. This method can discover, monitor, and improve processes in various application domains. However, the process models produced by typical process discovery methods are difficult for humans to understand due to their high complexity (the so-called "spaghetti-like" process models). Moreover, these methods cannot handle uncertainty or perform predictions because of their deterministic nature. Recently, researchers have been developing predictive approaches for running business cases of processes. This paper focuses on developing a predictive business process monitoring approach using reinforcement learning (RL), which has been successful in other contexts but not yet explored in this area. The proposed approach is evaluated in the banking sector through a use case.

Automatically reconciling the trade-off between prediction accuracy and earliness in prescriptive business process monitoring

Prescriptive business process monitoring provides decision support to process managers on when and how to adapt an ongoing business process to prevent or mitigate an undesired process outcome. We focus on the problem of automatically reconciling the trade-off between prediction accuracy and prediction earliness in determining when to adapt. Adaptations should happen sufficiently early to provide enough lead time for the adaptation to become effective. However, earlier predictions are typically less accurate than later predictions. This means that acting on less accurate predictions may lead to unnecessary adaptations or missed adaptations.Different approaches were presented in the literature to reconcile the trade-off between prediction accuracy and earliness. So far, these approaches were compared with different baselines, and evaluated using different data sets or even confidential data sets. This limits the comparability and replicability of the approaches and makes it difficult to choose a concrete approach in practice.We perform a comparative evaluation of the main alternative approaches for reconciling the trade-off between prediction accuracy and earliness. Using four public real-world event log data sets and two types of prediction models, we assess and compare the cost savings of these approaches. The experimental results indicate which criteria affect the effectiveness of an approach and help us state initial recommendations for the selection of a concrete approach in practice.

When to Invoke a Prediction Service for Business Process Monitoring?

Predictive monitoring of business processes aims at forecasting the future information of a business process and has gained increasing attention in recent years. With the development of cloud computing, prediction models, including remaining time prediction models, can be provided as cloud services. Reducing the number of invocation times of a remaining time prediction service is necessary since a large quantity of business process instances may be initiated and monitored every day. However, most of the current research focuses on designing new algorithms to improve prediction accuracy but there is no approach available to decide when to invoke a prediction service for each business process instance. In this paper, we propose a deep reinforcement learning based strategy that can learn the policies of selecting prediction points for the remaining time prediction. Specifically, the learned policies can dynamically decide the next prediction point at which the remaining time prediction service will be invoked for a business process instance. We performed extensive experiments on five real-world datasets. The experiment results show this strategy can reduce the number of prediction points significantly and still maintain the high prediction accuracy.

Artificial intelligence to counteract “KPI overload” in business process monitoring: the case of anti-corruption in public organizations

PurposeThe nature and amount of data that public organizations have to monitor to counteract corruption lead to a phenomenon called “KPI overload”, consisting of the business analyst feeling overwhelmed by the amount of information and resulting in the absence of appropriate control. The purpose of this study is to develop a solution based on Artificial Intelligence technology to avoid data overloading and, at the same time, under-controlling in business process monitoring.Design/methodology/approachThe authors adopted a design science research approach. The authors started by observing a specific problem in a real context (a healthcare organization); then conceptualized, designed and implemented a solution to the problem with the goal to develop knowledge that can be used to design solutions for similar problems. The proposed solution for business process monitoring integrates databases and self-service business intelligence for outlier detection and artificial intelligence for classification analysis.FindingsThe authors found the solution powerful to solve problems related to KPI overload in process monitoring. In the specific case study, the authors found that the combination of Business Intelligence and Artificial Intelligence can provide a significant contribution to the detection of fraud, corruption and/or policy misalignment in public organizations.Originality/valueThe authors provide a big-data-based solution to the problem of data overload in business process monitoring that does not sacrifice any monitored Key Performance Indicators and that also reduces the workload of the business analyst. The authors also developed and implemented this automated solution in a context where data sensitivity and privacy are critical issues.

Predictive Business Process Monitoring Approach Based on Hierarchical Transformer

Predictive business process monitoring is predicting the next stage of the business process based on the sequence of events that have occurred in the business process instance, which is positive for promoting the rational allocation of resources and the improvement of execution efficiency. There are drawbacks in modeling business process instances, such as conceptual drift phenomenon and long event sequences. Therefore, we propose a hierarchical Transformer-based business process prediction model to improve the performance of the Transformer-based predictive business process monitoring model. We encode the event features using two different encoding methods to obtain the relationship between activities and attributes. A drift detection algorithm is proposed to segment the business process and calculate the correlation between activities and segments by using cross-attention. Furthermore, learnable position encoding is designed to capture the relative position information of subsequences. Finally, the information of different granularity, such as event attributes, event subsequences, and complete instances, is fused by different weights. Experiments were run on seven real event logs for the next activity prediction and remaining time prediction, and the next activity prediction accuracy improved by 6.32% on average, and the mean absolute error of remaining time prediction reduces by 21% on average.

BPMNE4IoT: A Framework for Modeling, Executing and Monitoring IoT-Driven Processes

The Internet of Things (IoT) enables a variety of smart applications, including smart home, smart manufacturing, and smart city. By enhancing Business Process Management Systems with IoT capabilities, the execution and monitoring of business processes can be significantly improved. Providing a holistic support for modeling, executing and monitoring IoT-driven processes, however, constitutes a challenge. Existing process modeling and process execution languages, such as BPMN 2.0, are unable to fully meet the IoT characteristics (e.g., asynchronicity and parallelism) of IoT-driven processes. In this article, we present BPMNE4IoT—A holistic framework for modeling, executing and monitoring IoT-driven processes. We introduce various artifacts and events based on the BPMN 2.0 metamodel that allow realizing the desired IoT awareness of business processes. The framework is evaluated along two real-world scenarios from two different domains. Moreover, we present a user study for comparing BPMNE4IoT and BPMN 2.0. In particular, this study has confirmed that the BPMNE4IoT framework facilitates the support of IoT-driven processes.

Lifecycle Support for IoT-Driven Business Rules

The Internet of Things (IoT) is prevalent in various areas and enables a variety of smart applications such as smart factory, smart home, and smart health. The IoT corresponds to a network of physical objects, fostering the collection and exchange of data over the Internet. Each physical object is equipped with sensors, actuators, and other technologies, and is uniquely identifiable through its embedded computing system. Business Process Management (BPM), in turn, is concerned with the modeling, execution, discovery, analysis, and monitoring of business processes. By enriching business processes with real-world IoT data and IoT-driven business rules, process automation as well as real-time process monitoring and decision making can be enhanced. Despite the increasing interest in IoT-enhanced business process support, IoT-driven business rules have not been well integrated with business process execution so far. In this context, providing integrated lifecycle support for IoT-driven business rules in BPM systems constitutes a challenge. This paper discusses how to integrate lifecycle support of IoT-driven business rules with BPM systems covering the modeling, execution, monitoring, analysis, and evolution of IoT-driven business rules. Furthermore, all lifecycle phases need to be covered by holistic approaches to IoT-driven business rules. For each lifecycle phase, we elaborate on the requirements emerging in this context and discuss existing approaches.

Entropy Model of Business Processes

The article proposes an approach to solve the problem of monitoring business processes based on the entropy characteristics of the information space of an existing business system, which is considered from the point of view of the behavior and modeling of open systems. The time component of the process is taken into account. In addition, the paper describes a method for assessing the stability of the dynamic structure of a business process. An entropy approach to the implementation of the task of entropy monitoring of a business system and its processes is proposed. The article points to the expediency of using system theories that consider the totality of interacting business processes as an integral large system. At the same time, the whole is characterized by a triple interdependence, namely, the interdependence of each of its most important parts, then the dependence of each of its parts on the entire system (in this case, the boundary conditions of the system in the selected phase of space) and, finally, the dependence of the entire system on its specific part or parts from combinations. It should be noted that the synthesis of all these relationships leads to the need to take into account the complex nonlinearity of processes. Keywords: entropy, methodology, business administration system.