Industry Track

Have you alredy planned your professional qualification activities for 2020? REFSQ 2020 is an excellent qualification option if you are working as a professional Requirements Engineer, Requirements Manager, Requirements Engineering Coach, or Business Analyst.
Compared to previous years we will expand the industry activities.

Industry Keynote

Thursday, 25th of June 2020

Alistair Mavin
Independent Requirements Specialist

S is for System

I have always been slightly concerned by the name of this conference series: “Requirements Engineering: Foundation for Software Quality”. Most attendees would probably agree that requirements engineering is indeed a foundation of software quality. However, I am uncomfortable that the S in REFSQ stands for software. I would argue that requirements engineering is a foundation of system quality. In my view, requirements are always for a system. Of course, that system will often include software. However, even the most software-intensive system is still a system and includes some non-software elements. So-called “software systems” must have some affect in the real physical world. Since they must impact non-software elements in some way, they are part of a wider system. It follows that these non-software system requirements must also be considered as part of a requirements engineering effort.
I have heard members of the requirements community suggest that requirements are for software because software is inherently more complex than non-software system elements. I can only assume that these individuals have not worked on safety-critical systems. Next time you are on an aeroplane, look out of the window at the engines. How would you feel if you were told that the engine control software had been developed against a set of requirements, whilst the physical components had not? Personally, I would prefer to know that the whole engine control system had been developed against a robust set of requirements.

This talk will expand on this viewpoint and discuss the issue of requirements engineering for systems.


Alistair Mavin (Mav) is an independent requirements specialist based in the UK. Mav worked as a requirements specialist at Rolls-Royce PLC for 14 years. He has carried out requirements engineering projects in a range of industries including defence, aerospace, rail, automotive, industrial plant design and software systems. He is the lead author of EARS and EARS+ and has experience in the development and delivery of requirements engineering training and in innovation and creativity support. Mav has published many papers on requirements and systems engineering. He was Industry Chair for RE13 and Industry Laboratory Chair for RE14 and is a member of the IEEE “RE” conference series Industry Committee. Mav is a member of IEEE, INCOSE, the British Computer Society (BSC) and the BCS Requirements Engineering Specialist Group committee and is a chartered engineer.

Industry Talks

Thursday, 25th of June 2019

Like every year the industry track will feature a set of high-quality invited presentations. This year presentations will mainly focus on domain-specific RE and cyber-physical systems.

Fabio Falcini
National Research Council of Italy (CNR)

Giuseppe Lami
National Research Council of Italy (CNR)

Thursday, 25th of June 2020

State-of-the-Practice of Requirements Management in Automotive

In the automotive industry, the development of software-intensive components is mainly demanded to specialized suppliers that are required by car makers to assess and improve their process by applying process models such as Automotive SPICE. Automotive SPICE is a widely-applied reference framework providing a set of requested practices in the development life-cycle of software-intensive automotive components, including requirements management practices. Requirements management is a key issue in such a context because the high volatility of requirements during projects, the need of tailoring platform requirements into several releases for specific vehicles and the need of interactions among different stakeholders. This presentation aims at contributing in answering the following question: what are the most common weaknesses in requirements management in automotive? The authors present the results of an empirical study aimed at characterizing and analysing recurrent requirement management weaknesses in automotive industry. The authors, as Automotive SPICE assessors, have directly recorded and identified specific software process improvement opportunities on the basis of the evidences gathered from real software development projects during a significant number of assessments performed at several organizations worldwide. This presentation aims at deriving a picture of the state of the practice of requirements management in automotive and to provide researchers and practitioners with a reference for improvement initiatives aimed at solving those weaknesses.


Fabio Falcini is an independent professional affiliated with Italian Research Council with an extensive automotive software experience. His main research and activity areas include software engineering, functional safety, deep learning and software quality. Falcini received a Master in Software Engineering from TecnoPadova. He is a Principal Automotive SPICE Assessor and actively contributes to the ISO 26262 community.

Giuseppe Lami, a researcher at the Information Science and Technology Institute (ISTI) in Pisa (Italy) where he has been working for 20 years. His is the head of the System & Software Evaluation Center at ISTI. His research interests are mainly related to software quality evaluation and software process assessment and improvement. He is an Automotive SPICE Principal assessor and the president of Automotive SPIN (Software Process Improvement Network) Italia. He previously (2003-2006) worked as Resident Affiliate at the Software Engineering Institute (Pittsburgh, PA).

Paolo Busetta
Delta Informatica

Thursday, 25th of June 2020

Real requirements for virtual environments

Over the years, Delta Informatica has accumulated a significant experience in the development of virtual teams, i.e. of software agents acting in coordination in the real world (e.g. to control drones), in virtual environments (e.g. as Non-Player Characters), in decision support systems (e.g. to filter and fuse data and suggest plans of action). The dialogue between software specialists and domain experts, which is often difficult in any type of software projects, is complicated by unrealistic expectations of what automatic systems can achieve and on the state of technology in general whenever anything related to AI or virtual reality is involved. The talk will present a few lessons learned on how to handle this interaction in order to achieve user satisfaction.


Paolo Busetta is a software consultant, principal researcher, software architect as well as full stack software developer and analyst. He has contributed to various research and industrial projects involving multi agent systems, symbolic AI, simulation and virtual reality, in Italy and in Australia. He has been active in the ICT field since 1981.

Filippo Chiarello
Co-founder and CTO at TEXTY s.r.l.

Thursday, 25th of June 2020

Exploiting Patent Documents in Novel Ways Thanks to Named Entity Recognition

Patents are a huge and public source of valuable design knowledge, that is hidden behind legal and technical jargon. Designers and Engineers who want to use patents in their job, have to invest considerable amount of effort in the document retrieval and interpretation. In the presentation I show how Natural Language Processing tools can be used to solve this problem, presenting a system that automatically detects and extracts three information that are relevant for the process of new product design: the users, the advantages and drawbacks described in patents. The system is applied on a real cases of four patent sets of different technologies. The talk will end with reflection on the possible applications of the method for Requirements Engineering.


Filippo Chiarello received a Ph.D. in Management Engineering ("summa cum laude") at the University of Pisa in 2018. The doctoral thesis won the Italian prize “Bernardo Nobile”. Filippo Chiarello made use of data science and in particular text mining tools to build systems able to automatically extract knowledgeable information from technical texts. The ability to mix ontologies and technical dictionaries with natural language processing techniques brought his methods to be applied both in research and business (Whirlpool, Continental, Adecco, and many large enterprises and SMEs). His interests started from patents, have been immediately extended to papers, tenders, technical specifications but finally reached also Twitter and Facebook always used to investigate or assess technical contents.

Jordan Kyriakidis
QRA Corp.

Thursday, 25th of June 2020

An approach to modern tool development for Requirements Engineering

This talk will focus on the approach developed by QRA (likely non-uniquely) on developing products specifically in aid of requirements engineering.
We have developed a hierarchy-of-needs model for requirements analysis, starting from the base layer of the individual requirement, up to the integrated collection of the system requirements (almost always a combination of software, hardware, and people), and the integration with or hand-off to subsequent engineering phases.
I will describe how we approach, classify, and utilize AI technologies based on its practicality, its predictive power and, importantly, it’s explanatory capabilities. Above all we focus on the practicality of any solution, both on the type of problems on which we ought to focus, but equally on how any particular solution is to be deployed and used. A too-large barrier to adoption is a great way to build a solution nobody will use.
I will end the presentation describing how we see the large-scale macro-trend of systems engineering and the increasingly integrated role of requirements engineering. In particular, we see the focus appropriately shifting 1) leftwards, earlier in the development cycle, 2) upwards, towards the full integrated system, and 3) inwards, towards algorithms and data.


Dr. Jordan Kyriakidis is cofounder and CEO of QRA Corp. A provider of solutions intended to ensure the products our customers build have the behaviour they require. QRA's technology incorporates both traditional and non-traditional forms of AI, and is used by Fortune 500 companies in the automotive, aerospace, energy, and medical device industries. Jordan is also a founding member of the Canadian Federal Government Advisory Council on AI. Prior to QRA, and since 2002, Jordan was a Professor and head of the Quantum Theory Group at Dalhousie University. He holds a PhD, summa cum laude, in Theoretical Physics from the University of Basel, Switzerland.

Giampaolo Armellin

Thursday, 25th of June 2020

Software Quality Release Control on Critical Applications

CBA with its more than 1000 customers over Italy gained core experience in developing critical software in healthcare and administrative fields (e.g. EHR, HR, economic management) where early detection of bugs is crucial to minimize the impact on customers that may potentially be catastrophic on health of people and management of companies.
These challenges required our company to organize the entire development cycle with teams specialized in software analysis, testing, customer relationships in order to optimize: requirements elicitation, early detection of defects by (automatic) testing, customer's feedback collection (e.g. bugs reports, new requirements). These activities do not require only technical skills and knowledge of formalism from state of the art software development life-cycle techniques, but they are also a matter of human interaction especially when customers are directly involved.
In this talk I will present our experience in developing, managing and continuously evolve software to comply with regulations (e.g. GDPR, ISO, financial and employment law, ...), customers' expectations and new market opportunities. I will show how we adapted software development approaches to the new needs of our market, the internal skills of our development team and customers' expectations.


Giampaolo Armellin is the head of the Research and Innovation Unit at CBA. Directly reporting to the CEO, he devises plans and budgets for research and innovation projects concerning CBA’s products and solutions. Moreover, he constantly monitors resources and activities, evaluating technical options and scientific results. He takes care of CBA’s relationships and collaborations with Research Institutions. He has a Master’s degree in Computer Science at the University of Milan, he started working in healthcare and wellbeing fields in 2005. From then and on he has been involved in many research and industrial projects for local and national agencies in Italy and participated on EU FP7 projects.