Salta al contenuto principale

The research activities carried out at DICI in Management Engineering area concern the typical topics of the Degrees Courses (i.e., management and control of systems, processes, products, and services) and are developed in an interdisciplinary and multidisciplinary key to enable a broader vision of business management and the use of structured methods to observe innovation dynamics and their impact on human resources, to support business decisions with objective and timely data.
The research and teaching activity is also enriched by the two masters in Industry 4.0 Design and Scalability promoted by the Department, and by various post graduate specialization courses including Data Driven Project Management.

Natural Language Processing (NLP) is a field of research on automatic techniques to process unstructured data, i.e., texts. In the research group we explore the applications related to technical texts (tender specifications, technical specifications, scientific articles, and patents). Some examples are automatic information extraction from text and natural language generation. NLP include several approaches and methods, among those, Named Entity Recognition can identify information units in text to tag specific part using list of entities (gazzeteer-based approach) and/or syntactic rules and recurrent patterns (rule-based approach). The advance in Artificial Intelligent systems allows to exploit database of document with tags to train machine learning algorithms in detecting relevant part of texts or identify and measure the semantic similarity among different contents.
In the research activities carried out at DICI, NLP methodologies are used to extract relevant information from open resources such as scientific publications, patents, education-related content, policy documents, technical reports, and private documents that companies share for research purposes. The typical information we extract are users (for marketing reasons), their needs and requirements (for product development and industrial design), technologies (for research and development), but also failures (for engineering and maintenance). The transformation of text into formal elements of an engineering nature allows a faster elaboration of project documentation (QFD, FMECA, etc.), the analysis of the congruence of the specifications, the analysis of the completeness of document systems such as quality in the company.

Industry 4.0 (also referred as Fourth Industrial Revolution) refers to a novel approach to the industrial system, that is based on the real-time connection of people, machines, and objects for the intelligent management of logistic-production systems. The term Industry 4.0 has been initially formulated in Germany in 2011 by during the Hannover Trade Fair, to address the current phase of computerization of the manufacturing systems. The Fourth Industrial Revolution had changed the role of many actors and stakeholders in the market as happened previously with the first, second and third revolution. The new industrial paradigm sees technological complexity (the 9 enabling technologies), but also organizational and cultural complexity.
Research in this area is oriented towards the definition of new methods to assess the readiness of companies to face the era of digitalization, the design of new approaches for the extraction of Industry 4.0 technologies, studies on innovation dynamics, technology development and lifecycle, exploiting NLP and
Engineering Design approaches.


Paper & Patent Intelligence is the practice to gather information about the current trends in the scientific and technological world from publications and from patents. The recent improvement of Natural Language Processing techniques is making more and more valuable to perform such information extraction since extremely detailed insights can be identified from unstructured data (i.e., texts). The analysis concern technological changes and innovation dynamics, enabling companies’ decision making process with objective and timely data and insights.

HR 4.0 refers to the management of people in a company leveraging on digital technologies. It encompasses employee recruitment, training and development, performance evaluation and management of employee-benefits and reward. In this context, we explore new approaches and indicators for an intelligent integration of data in HR management.
We aim at mapping future professional roles in the digitalization era, considering on one hand the new skills and competences needs and, on the other hand, the challenges for the Education Systems to effectively prepare students for the job market. The necessary capabilities to face the current challenges of digital transformation include not only the technical knowledge and abilities to properly exploit the potential of digital technologies, but also the so-called soft skills. Those are transversal capabilities, social skills referred to personality traits, and behaviour in working environment. Our studies focus on tools to automatically extract skills from unstructured texts applying NLP approaches on a corpus of texts to fuel future quantitative research on this topic.

Nowadays, data related to education are varied and in huge quantity, including books, slides, papers, technical documents, MOOCs, courses descriptions, universities curricula, lessons transcriptions, job vacancies, job descriptions, videos (with transcripts), commented graphs and images and so far, and so further. The feature in common among these resources is the form: all those documents contain a significant amount of text. This feature enables the utilization of Natural Language Processing (NLP) techniques in the field of education design.
The applications in this field of research concern the comparison between the offer in educational institutions and the demand in labour market, and the identification of skills and learning outcomes from courses description. In addition, it is possible to detect career paths, develop tailored educational programs, evaluate the alignment with standards and qualification both for validate learning materials and prior knowledge of learners.

The Business Engineering for Data Science Lab (B4DS) is a joint research laboratory of the University of Pisa, with members from two departments of the Engineering Schools (DICI e DESTEC) working on cutting edge applications of data science in the context of business management. Collaborations with colleagues in the mechanical area occur frequently, in particular with technologists and plant engineers. The joint research focuses on technologies and their evolution, the development of quantitative approach for market or business analysis, the study of methods for improving the performance of the company (economic, management, financial, but also with a view to improving the environmental impact of production activities).