Event-Driven Inconsistency Detection Between UML Class and Sequence Diagrams

Luan Lazzari, Kleinner Farias · 2025

Read Event-Driven Inconsistency Detection Between UML Class and Sequence Diagrams by Luan Lazzari, Kleinner Farias (2025) in the Tower of Babel, a free online library of public-domain works preserved in clean, searchable text. No sign-up. The full text is served as a plain-text file at a predictable URL, where text is present: https://antarctic-labs.com/tower-of-babel/arxiv-2511-07742.txt.

Introduction

Luan Lazzari and Kleinner Farias built a checker that catches clashes between software diagrams as modelers draw them. UML class diagrams and sequence diagrams must agree, yet students and engineers routinely let them drift apart. Teaching consistency is notoriously hard when feedback arrives late.

Harmony Validator plugs into the Papyrus modeling environment and watches editing events live. Inconsistencies are flagged and reported at the moment they appear. Event-driven checking turns consistency from an exam topic into a habit formed while drawing. Modeling courses and professional teams both gain earlier, cheaper correction of design mistakes.

Catalog record

CREATORLuan Lazzari, Kleinner Farias
YEAR2025
COLLECTIONPAPERS
FORMATTXT
FILE SIZE18 KB
VERSION1.0
LICENSEhttp://creativecommons.org/licenses/by/4.0/
RIGHTSCC BY
SOURCEarXiv

Source: https://arxiv.org/abs/2511-07742

Browse the full library · About the Tower of Babel

Full text

The opening of the work, from the library's hosted text:

Event-Driven Inconsistency Detection Between UML Class and Sequence Diagrams Luan Lazzari, Kleinner Farias Abstract . Modeling is a central and demanding activity in software engineering that requires skills such as abstraction, consistency maintenance, and precise communication. These skills are difficult to master and even harder to teach effectively. Educators and students often struggle to understand and manage inconsistencies that arise during the modeling process. To address this challenge, we present Harmony Validator , a tool integrated as a plugin for the Papyrus modeling environment, designed to automatically detect and report inconsistencies in UML models, including class and sequence diagrams. The tool adopts an event-driven architecture that continuously monitors modeling actions and notifies users of emerging inconsistencies in real time. This approach enhances awareness of model integrity and supports the iterative refinement of design artifacts. The paper describes the architecture, detection mechanisms, and usage scenarios of Harmony Validator. It also includes a case study conducted with students in a software engineering course to evaluate the perceived usefulness and benefits of UML modeling in teaching and learning. Our results indicate that Harmony Validator fosters a better understanding of model consistency and promotes reflective learning practices in software modeling education. 1. Introduction High-quality documentation significantly increases the success of software projects ( 4 ; 14 ; 24 ; 19 ) . This is particularly evident for visual documentation based on the Unified Modeling Language (UML) ( 18 ) , which remains a central artifact in both academia and industry ( 10 ; 11 ) . UML diagrams foster shared understanding, encourage design discussions, and support reasoning about system structure and behavior ( 19 ; 8 ) . Recent evidence shows that UML is regaining attention as a standard medium for documenting and communicating software design decisions ( 19 ) . In educational settings, UML enhances the teaching of abstraction, decomposition, and communication skills. In industry, consistent UML documentation ( 6 ; 25 ; 5 ) improves knowledge transfer across teams and contributes to software maintainability and quality ( 9 ; 1 ) . Empirical findings demonstrate that developers achieve higher functional correctness and introduce fewer defects during maintenance when UML models are up-to-date and consistent ( 4 ; 1 ; 8 ; 17 ) . These studies reinforce the importance of inconsistency detection to sustain the reliability of UML models and to preserve their benefits for software comprehension and evolution. The Harmony Validator tool (Table 1 ) was conceived to address an ongoing challenge in model-driven engineering, i.e., the automatic detection of inconsistencies across UML diagrams ( 22 ; 23 ) . As software systems evolve, disagreements often arise between structural (class) and behavioral (sequence) diagrams, leading to conceptual mismatches that compromise model integrity. Manual validation of these relationships is error-prone and time-consuming, particularly in collaborative modeling environments. Existing tools typically detect inconsistencies through post-hoc or batch verification, providing limited real-time support ( 16 ; 26 ) . Harmony Validator fills this gap through an event-driven architecture that continuously monitors modeling actions and reports inconsistencies as they occur. This design supports both students and practitioners in identifying, interpreting, and resolving inconsistencies early in the modeling process, reducing cognitive load ( 2 ; 20 ) and enhancing design correctness. Code metadata item Description Current code version V1.0 Permanent link to code/repository used for this code https://github.com/luanlazz/uml-harmony-validator-service version https://github.com/luanlazz/uml-harmony-validator-plugin Legal Code License MIT License Code versioning system used git Software code languages, tools, and services used Java, Kafka, Redis Compilation requirements, operating environments and dependencies Docker, Java 17, Apache Maven 3.5+ Link to developer documentation/manual https://github.com/luanlazz/uml-harmony-validator-service/blob/master/readme.md https://github.com/luanlazz/uml-harmony-validator-plugin/blob/main/README.md Support email for questions [email protected] Table 1. Description of the source code metadata. Users employ the Harmony Validator within the Papyrus 1 1 1 Papyrus: https://eclipse.dev/papyrus/index.html modeling environment to create and refine UML class and sequence diagrams while the tool continuously monitors model changes and automatically detects inconsistencies such as undefined operations, broken associations, or mismatched messages. Real-time notifications highlight the affected elements, prompting immediate user attention and fostering iterative correction and design coherence. In educational settings, Harmony Validator has been used in laboratory sessions to train students in identifying and resolving inconsistencies, improving modeling fluency and awareness of inter-diagram dependencies. Built on robust foundations in model validation and event-driven systems, the tool extends Papyrus—based on the Eclipse Modeling Framework 2 2 2 Eclipse Modeling Framework: https://eclipse.dev/emf/ (EMF) and integrates Apache Kafka 3 3 3 Apache Kafka: https://kafka.apache.org/ for efficient event handling. Implemented in Java language and deployed via Docker containers, Harmony Validator ensures interoperability, reproducibility, and scalability, enabling its seamless integration into both academic experiments and industrial modeling workflows. The Harmony Validator distinguishes itself from existing approaches ( 21 ; 15 ; 16 ) in three main ways: (i) It introduces an event-driven, continuous-monitoring paradigm , enabling real-time inconsistency detection rather than relying on static, batch-oriented validation; (ii) It performs bi-directional consistency checking between structural and behavioral UML diagrams, ensuring semantic alignment between class definitions and interaction flows; and (iii) It was designed as both a pedagogical and research-oriented tool , supporting empirical investigations into modeling practices while also assisting practitioners in maintaining model integrity. This study is structured as follows: Section 2 describe the proposed tool, outlining its architecture, functionalities and practical use case. Section 3 presents an practical usage example of the proposed tool. Section 4 outlines the impact of the proposed tool. Finally, Section 5 introduces some concluding remarks and future plans. 2. Software description 2.1. Software architecture Figure 1 presents an overview of the Harmony Validator architecture, which follows an event-driven architectural style ( 3 ; 13 ; 12 ) . The system is divided into two main modules: the Client App and the Docker Host environment. This separation ensures modularity, scalability, and ease of deployment across different platforms. In Figure 1 (A), the Client App corresponds to a plugin integrated into the Papyrus modeling environment, where users perform UML modeling activities. Within Papyrus, the Harmony Validator adds a custom panel that continuously monitors and reports inconsistencies detected in the UML model. The Harmony Validator interface presents feedback to the user through visual indicators that identify affected elements and diagrams. This integration enables users to maintain focus on modeling tasks while receiving real-time feedback on inconsistencies. In Figure 1 (B), the Docker Host contains a set of services deployed as independent containers, connected through an asynchronous Event Bus . The event-driven layer is responsible for receiving and processing model modification events. The Model Reader Service retrieves the updated model data from Papyrus and publishes the modification events to the system. The Inconsistency Service subscribes to these events, analyzes the model state, and identifies inconsistencies across UML diagrams. Detected inconsistencies are then cached using Redis for quick retrieval and aggregated notifications. All communication between the client and backend follows an HTTP pattern implemented through an API Gateway/BFF (Backend for Frontend) module, which mediates requests from the desktop client to the microservices running in Docker. This architecture promotes decoupling between components, supports parallel modeling scenarios, and allows for seamless scalability. Figure 1. Event-driven architecture of the Harmony Validator. (A) Client App integrated with Papyrus modeling environment. (B) Docker Host containing model reader, inconsistency detection, and event bus services. 2.2. Software functionalities The Harmony Validator provides

Read the complete text