Requirements Comprehension Using BPMN: An Empirical Study

Currently, the number of patients with neurological diseases is increasing, especially those older than 65 suffering from Alzheimer’s disease. This development increases the emphasis on understanding and mapping treatment and care processes, not only for the elderly. Service providers (of both treatment and care) are under general pressure to decrease charges and maintain or improve existing levels of care. This situation is significantly influenced by a comprehensive knowledge of the whole process and its values. This publication therefore aims to describe the fundamental procedural aspects of caring for patients with Alzheimer’s disease, using Business Process Model and Notation (BPMN). It also aims to show the possibilities of using BPMN in the description of treatment and care. Modeling of the business process is more frequently being applied not only by businesses but also by scientists involved in process models. It is used to model medical topics, with approximately 10% of its publications only, and most of these publications deal only with clinical pathways, not with overall treatment and care processes. However, the BPMN model allows the whole process of medical and nonmedical care for patients with Alzheimer’s disease to be described, including the decomposition of partial activities into individual threads and sub-processes or atomic tasks. This paper presents the BPMN modeling and mapping of the specific care path for neurodegenerative patients. The text provides a new perspective on the BPMN modeling of Alzheimer’s disease. The presented model offers the option of expanding treatment cost calculation to simulate the process using graphical tools and languages. The overall view of this system creates a much more complex concept of the system and its surroundings.

Large Language Models (LLMs) are emerging as powerful tools for automating business process modeling, promising to streamline the translation of textual process descriptions into Business Process Model and Notation (BPMN) diagrams. However, the extent to which these Al systems can produce high-quality BPMN models has not yet been rigorously evaluated. This paper presents an early evaluation of five LLM-powered BPMN generation tools that automatically convert textual process descriptions into BPMN models. To assess the external quality of these Al-generated models, we introduce a novel structured evaluation framework that scores each BPMN diagram across three key process model quality dimensions: clarity, correctness, and completeness, covering both accuracy and diagram understandability. Using this framework, we conducted experiments where each tool was tasked with modeling the same set of textual process scenarios, and the resulting diagrams were systematically scored based on the criteria. This approach provides a consistent and repeatable evaluation procedure and offers a new lens for comparing LLM-based modeling capabilities. Given the focused scope of the study, the results should be interpreted as an exploratory benchmark that surfaces initial observations about tool performance rather than definitive conclusions. Our findings reveal that while current LLM-based tools can produce BPMN diagrams that capture the main elements of a process description, they often exhibit errors such as missing steps, inconsistent logic, or modeling rule violations, highlighting limitations in achieving fully correct and complete models. The clarity and readability of the generated diagrams also vary, indicating that these Al models are still maturing in generating easily interpretable process flows. We conclude that although LLMs show promise in automating BPMN modeling, significant improvements are needed for them to consistently generate both syntactically and semantically valid process models.


 
 
 The most important result to standardize the notation for graphical representation of Business Processes (BPs) is the Business Process Model and Notation (BPMN). Despite the BPs modeled with BPMN being able to support business designers, BPMN models are not appropriate to support the analysis phase. BPMN models have no formal seman- tics to conduct qualitative analysis (validation and verification). In this work is presented how Model Checking (MC) verification technique for software and Timed Automata (TA) formal language are integrated within a formal verification approach to check BPs mod- eled with BPMN. Also, are introduced a set of guideline to transform BPMN models into TA. The use of our approach allow to business analysts and designers to perform evaluation (i.e., qualitative analysis) of BPs, based on the formal specification of BP–task model with TA. The application of the approach is aimed to evaluate the behavior of the BP–task model with respect to business performance indicators (for instance, service time, waiting time or queue size) derived from business needs, as is shown in an instance of an enterprise–project related to Customer Relationship Management.
 
 

Business Process Model and Notation (BPMN) is a standard for formally modeling complex business processes. Manual creation of BPMN models can be time-consuming and error-prone, prompting a need for automation. Existing approaches, such as rule-based methods, machine learning, and machine translation, have progressed but face accuracy and real-world applicability challenges. In this research paper, we propose a novel method for automated extraction of BPMN models from textual descriptions using natural language processing (NLP) tools and deep learning models, including the spaCy library for text processing, a fine-tuned BERT model, and state-of-the-art large language models like GPT-3.5-Turbo and GPT-4. We utilize Graphviz, an open-source graph visualization software, to visualize the extracted processes. Our method supports representing tasks, exclusive gateways, parallel gateways, and start and end events in the generated BPMN models. The evaluation of 31 textual descriptions shows that our method generates process models with 96% accuracy using GPT-4 and 80% accuracy using GPT-3.5-Turbo large language models. Although subject to certain limitations, such as occasional inaccuracies in model outputs and reliance on well-formed input text, our approach offers a valuable contribution to the growing body of research on automating BPMN model generation.

The recipe and grocery shopping application is designed to increase the ease of users in accessing cooking recipes and shopping for groceries in an integrated manner. This research aims to analyze the business processes involved in the application using the Business Process Model and Notation (BPMN) approach. Analysis is carried out to understand the workflow structure, identify potential efficiencies, and produce performance improvements before and after system implementation. The research results show that the application of this application is able to direct processes that were previously manual and fragmented, such as searching for recipes, making shopping lists, and ordering groceries. The BPMN model is used to visualize changes in business process flow, enabling comprehensive evaluation of aspects of time efficiency, error reduction, and increased user satisfaction. This research concludes that the BPMN approach is effective in analyzing and optimizing business processes in recipe and grocery shopping applications. These findings contribute to the development of similar applications, especially in designing systems that are more efficient and responsive to user needs.

Research on blockchains addresses multiple issues, with one being automating creation of smart contracts. One of the promising approaches, targeted for industry adoption, uses Business Process Model and Notation (BPMN) modeling to describe an application that is transformed into a smart contract. Although systems using this approach have been developed as a proof of concept, they do not address a major issue arising in the trade vertical that has applications involving collaboration of many actors over a long period of time. There are two difficulties: (i) BPMN's concept of a transaction is limited to actions of one actor, while trade transactions involve collaborations by many actors, and (ii) trade applications are long term and thus require support of transactional activities that span invocations of several methods of a smart contract. The difficulty is that a blockchain infrastructure supports the transactional properties only within the scope of one execution of one method of a smart contract. We extend the BPMN model by introducing the concept of a collaborative transaction to represent activities of many actors. In our previous research, we developed an approach, and a tool as a proof of concept, to transform a BPMN model into methods of a smart contract. In this paper we describe how collaborative transactions are supported by our approach even though the transaction activities span multiple calls to methods of a smart contract produced by transformation from a BPMN model to methods of a smart contract.

A metamodel to integrate business processes time perspective in BPMN 2.0

Reliability of BPMN Business Processes

During software maintenance, software models play a fundamental role, as they assist in understanding the software that will be corrected or evolved. Business process models, generally produced using Business Process Modeling and Notation (BPMN), help understanding the business processes supported by software. Therefore, these models are instrumental in maintaining corporate software. However, it is common that during software maintenance, models related to the software are not updated. Thus, the models are inconsistent with the version in use of the software, becoming increasingly obsolete over time. Given that, we developed the e-VOL BPMN technique to support BPMN models' evolution during software maintenance. The objective of e-VOL BPMN is to assist software engineers and developers, with little experience in BPMN notation, to carry out the necessary evolutions in the processes supported by the software. We evaluated the e-VOL BPMN in two experiments, and the results showed evidence that e-VOL BPMN helps the evolution of BPMN models in an appropriate manner. With this technique, such professionals can keep BPMN models up to date. This makes it easier to understand the processes supported by the software, which reduces the time and cost of future maintenance. Furthermore, understanding these processes correctly can reduce undue modifications to the software, contributing to the quality of the maintained software.

The Business Process Model and Notation (BPMN) is a widely used standard for business process modelling and maintained by the Object Management Group (OMG). However, the BPMN 2.0 specification is quite comprehensive and spans more than 500 pages. The definition of an element is distributed across different sections and sometimes conflicting. In addition, the structure of the elements and their relationships are described within the metamodel, however, further syntactical rules are defined within the natural text. Therefore, this paper defines an ontology that formally represents the BPMN specification. This ontology is called the BPMN 2.0 Ontology and can be used as a knowledge base. The description of an element is combined within the corresponding class and further explanations are provided in annotations. This allows a much faster understanding of BPMN. In addition, the ontology is used as a syntax checker to validate concrete BPMN models.

Canceling ongoing process instances is a natural phenomenon in practice. As such, modeling cancellation behavior is supported in the Business Process Model and Notation (BPMN) via exception events. Event-data-driven analysis techniques using such process models, e.g., conformance checking, require converting the BPMN model into a formal process modeling representation, i.e., Petri nets. However, the existing transformation of BPMN models with exception events renders a classical Petri net, with various additional modeling constructs to mimic the exception behavior. Using such a model in a subsequent analysis renders an infeasible computational complexity. Hence, this paper presents a novel conversion of BPMN models with exception events into reset nets, significantly reducing the number of required invisible transitions in the corresponding transformation. Our results show that the enhanced conversion reduces the computational effort of using the converted models for conformance checking.KeywordsBusiness process managementBusiness process modelingBPMNPetri nets

Business process modeling and management approaches are increasingly used and disclosed between organizations as a means of optimizing and streamlining the business activities. Among the various existing modeling languages, we stress the Business Process Model and Notation (BPMN), currently in version 2.0. BPMN is a widespread OMG standard that is actually used either in academia and in organizations. BPMN enables business process modeling, but does not facilitate the modeling of the information infrastructure involved in the process. However, interest in the data and its preservation has increased in BPMN's most recent version. The aim of this paper is to study BPMN 2.0, particularly on the usage and persistence of data, and present an approach for obtaining an early data model from the business process modeling, which may then be used as a starting data model in the software development process.

PurposeBusiness process modelling integrates and visualizes relevant information essential for managing day-to-day business operations. It plays a critical role in the design and execution of business transformations. Recognizing the role of process modelling, a large number of modelling languages, methods and techniques have been developed, each offering diverse advantages and having inherent limitations. Traditional and popular process modelling approaches focus on the exact specification of the control flow of business processes, whereas more recent approaches like Subject-oriented Business Process Management (S-BPM) are focused on the communication between process participants. This study aims to provide comparative insights about these two approaches through their experimental application. This study does so by comparing Business Process Model and Notation (BPMN); a control flow approach, with S-BPM; a communication approach, with a specific focus on their suitability for novice modellers.Design/methodology/approachThis paper reports on an exploratory experiment that compares BPMN to S-BPM. Applying cognitive load theory, this study compares the experiences and outcomes of novice process modellers, assessing perceived ease of use, model quality (syntactic and semantic) and modelling efficiency (time to model) across the two approaches.FindingsStudy results show that S-BPM (a communication approach) led to significantly better user performances for process modelling than BPMN (a control flow approach). This study points to how a different modelling approach such as S-BPM could be either considered as an alternative or to complement the more popular control flow approach BPMN. This observation was especially relevant for modelling contexts where domain experts are novice process modellers.Originality/valueThis study provides the first empirical evidence that communication approaches like S-BPM could outperform modelling approaches which are control-flow based (i.e. BPMN), especially when being used by novice process modellers who hold the domain and process knowledge. This study uses this as a springboard to present important considerations for practice and guide future process modelling research.

In 2014, the Object Management Group (OMG) published the Case Management Modeling and Notation (CMMN) version 1.0 specification, which is a new process modeling specification to complement its Business Process Modeling and Notation (BPMN) specification. The declarative nature of CMMN is intended to supplement the procedural perspective of BPMN. CMMN takes a data-centric view to process modeling based on business artifacts to provide flexibility for knowledge workers, while retaining the advantages of business process management (BPM) systems.

Business Process Model and Notation—BPMN