WDTE 2026 Abstracts


Area 1 - Digital Transformation in the Enterprise

Short Papers
Paper Nr: 7
Title:

LawAI: An Intelligent and Accessible Legal Assistance Agent for Portuguese Law

Authors:

Adriana F. Meira, Eduarda F. S. Gomes, Estrela Ferreira Cruz and A. M. Rosado da Cruz

Abstract: The principles of freedom and equality are the pillars of contemporary democratic societies. In these societies, all citizens possess equal social dignity and are equal before the law, reflecting the universal commitment to justice and civic equality. However, there is a justice gap that undermines the principle of legal equality, making only those who know how to navigate the system, and can afford legal services, able to access legal advice. In this paper, we propose an LLM-based legal agent, enhanced with Retrieval-augmented Generation, to provide legal advice and answer legal questions. Preliminary assessments based on Portuguese legislation on Higher Education show that the agent has the ability to act as a first-line service to help any citizen clarify doubts about legislation and help them navigate the legal landscape.
Download

Paper Nr: 8
Title:

Tuning Large Language Models for Building Specialized Agents for Legal Texts

Authors:

Inês Esteves, A. M. Rosado da Cruz and Estrela Ferreira Cruz

Abstract: Large Language Models (LLMs) have demonstrated remarkable performance across a wide range of general-purpose tasks. However, their effectiveness in specialized domains remains limited by outdated knowledge, insufficient domain-specific precision, and significant computational requirements. This project explores the application of fine-tuning and Retrieval-Augmented Generation (RAG) techniques to develop specialized agents tailored to the Portuguese legal domain. This work presents a design and implementation of a fine-tuning (FT) pipeline based on the Phi-1.5 model using Low-Rank Adaptation (LoRA), trained on curated legal texts extracted from Diário da República, the official gazette of Portuguese legislation. The development environment was stabilized through the version alignment of key machine learning libraries, including PyTorch, Transformers, PEFT, and Accelerate, ensuring reproducibility and compatibility across the training workflow. Preliminary training results show a consistent reduction in loss values throughout the fine-tuning process, supporting the technical feasibility of the proposed methodology. Although the RAG component remains part of future work, current results establish a reproducible foundation for fine-tuning LLMs in resource-constrained environments, particularly for Portuguese legal texts. Complementarily this work identifies and discusses key challenges and opportunities associated with adapting LLMs to high-stakes professional domains, such as the legal sector.
Download

Paper Nr: 13
Title:

Towards I5.0 Industrial System Architectures: A Systematic Literature Review

Authors:

André Araújo and Luis Ferreira

Abstract: The transition from Industry 4.0 to Industry 5.0 created a demand for industrial architectures capable of incorporating human-centricity, resilience, and sustainability, principles largely absent from existing architectures. This Systematic Literature Review aims to identify Industry 5.0 system architectures with a focus on effective Human-System Integration. Following the PRISMA guidelines, 17 articles were selected from 286 candidates and analyzed. The reviewed architectures demonstrate alignment with Industry 5.0 principles, with Human-System Integration emerging as an architectural concern through dedicated components and interaction mechanisms. Yet, gaps were identified: Human-System Integration is implemented as a one-directional interaction, positioning humans as passive recipients of system outputs rather than effective and active participants in the decision-making process; resilience and sustainability are rarely translated into verifiable mechanisms; and the lack of standardized evaluation methods limits the measurability of the architectures.
Download

Paper Nr: 9
Title:

Estimating Material Consumption per Production Order Using Machine Learning and a Fiware-Based Architecture

Authors:

Vitor Amorim, Sérgio Ivan Lopes and A. M. Rosado da Cruz

Abstract: The adoption of data-driven architectures has improved efficiency and accuracy in industrial production systems. However, in textile and clothing manufacturing, plant floor traceability remains a critical limitation. The tracking of raw materials and work-in-progress items, and their association with manufacturing orders, is often fragmented and dependent on manual processes, leading to reduced visibility and operational inefficiencies. This paper proposes a data-driven framework integrating a material tracking technology, such as RFID, with ML–based predictive analytics. Real-time automated identification and localization of materials throughout the production lifecycle ensures consistent and fine-grained traceability. The resulting data is consolidated into a unified digital layer supporting downstream analytics. Four ML models are compared to predict material requirements for production orders, leveraging historical consumption data and contextual production parameters. The models have been designed to support decision-making by optimizing material allocation, with the objective of minimizing both material shortages and excess inventory. The proposed approach enhances synchronization between physical and digital production systems, and demonstrates the potential to improve planning accuracy and resource utilization. Preliminary results also demonstrate the potential of using ML-based methods to enable more adaptive and efficient manufacturing systems aligned with Industry 4.0 principles.

Paper Nr: 12
Title:

An Integrated Framework for Social Unrest Analysis in Sri Lanka Using a Data Warehouse and Full Orchestration

Authors:

P. K. D. M. Premasinghe, N. P. K. Dewmini, H. L. N. T. de Silva, R. P. Abeynayake, K. B. A. Bhagyanie Chathurika and Kaushalya Dissanayake

Abstract: Sri Lanka’s COVID-19 pandemic period (2020–2021) and the subsequent economic crisis (2021–2022), characterized by foreign exchange shortages, inflation, and resource scarcity, highlight the need for systematic analysis of the economic drivers of social unrest. This paper presents a Sri Lanka-specific analytical framework to examine the relationship between key macroeconomic indicators personal consumption expenditure (PCE), gross domestic product (GDP), unemployment, government expenditure, wages, and agriculture-and the annual Reported Social Unrest Index (RSUI) from 2000 to 2022. A fully orchestrated data warehouse integrates heterogeneous data sources and produces analysis-ready datasets to ensure reproducibility. ARDL models with ECM representation and bounds testing are applied to estimate short-run and long-run relationships for each economic channel. The findings reveal heterogeneous effects across indicators, with GDP generally negatively associated with RSUI, while expenditure, unemployment, and agriculture vary across categories and groups. A dashboard and grounded chatbot provide accessible interpretation of results, supporting decision-making and further analysis.
Download