Nominee for BPM 2018 best paper award

The paper ‘On the synergies between Business Process Management and Digital Innovation’ by Amy Van Looy is nominated for the BPM 2018 best paper award. BPM is the premier conference on Business Process Management and is held this year in Sidney, Australia.

UGentMIS congratulates Amy and her Master of Science in Business Administration students taking the course Research Methods in Management and Informatics, who helped with collecting the data for this paper.

Enhancing Agile Development through Modelling of User Stories (Emma Willard & Axelle Pysson)

Context

In Agile Software Engineering, working software is valued over comprehensive documentation (http://agilemanifesto.org). Currently, practitioners of agile development practices often misinterpret this as ‘no documentation’. Consequently, user stories are almost the only kind of requirements documentation maintained in the application of agile development methods.

A user story describes a feature required to be provided by a software system in terms of a standard template structure. User stories are always formulated from the perspective of the person that desires the feature. A typical template for formulating user stories is:

As a type of user, I want some goal so that some reason

For example, related to requirements for a car navigation system, a user story could be:

As a driver, I want to be notified of traffic jams on my route, so that I avoid losing time.

Past research has shown that the documentation as a separate user story of each system feature desired by some system stakeholder might cause problems when at a later stage the agile development team loses the high-level view of the various system functions and their interaction. The problem gets magnified when the composition of the team changes, e.g., when new members join the team and need to comprehend the high-level architecture of the system under development. Even for moderately complex software, the number of user stories easily exceeds human capacity of overview and understanding. Furthermore, the more user stories are created, the more details are required in the acceptance criteria which are used for testing the system. Acceptance criteria are metrics that define when the system’s functionality is acceptable in response to a user story. If user stories are not integrated and aligned with each other, they create ambiguity, making it harder to decide when a system meets its requirements.

To alleviate the problems mentioned above, the use of conceptual models is suggested. Conceptual models are visual representations that are commonly used for understanding, analyzing and communicating system requirements. In ‘traditional’ (i.e., not agile) software development methods, a conceptual model is a graphical model with well-defined syntax and semantics that describes the domain to be supported by a software system (e.g., the domain of car navigation). This description is independent of the manner in which the software system is implemented. In other words, a conceptual model captures the function of a domain (e.g., how to use car navigation to achieve a purpose) and its behavior (e.g., how car navigation is used during a car trip), as well as its construction (e.g., what components does car navigation consist of and how are they related -car, GPS satellite network, GPS receiver, etc.) and its operation (e.g., GPS satellites broadcast signals about their current time and position), but does not prescribe decisions that should be left to the discretion of system developers (e.g., which track algorithm the GPS receiver runs to calculate the location, speed and direction of the car).

What is currently missing in the state-of-the-art is the automated generation of conceptual models that show both the functional (de)composition of the domain (i.e., function and behavior)) and the constructional (de)composition of the domain (i.e., construction and operation). Further, there is no empirical study that investigates the costs (e.g., effort to understand the conceptual model) and benefits (e.g., increased understanding of dependencies and other relationships between user stories) of using conceptual models generated from user stories. In other words, showing feasibility of automated generation of conceptual models from user stories is one thing; demonstrating the economic value of such approach in terms of increased quality and productivity of agile software development is something else, at least equally important for the valorization of the research.

Thesis description

At UGentMIS we initiated a line of research in which we investigate, together with researchers from Saint Louis University (USA) and a high-tech USA company, whether Natural Language Processing (NLP) techniques could be used to automatically generate conceptual models from user stories. The NLP-based generation of models has proven to be a challenging area, giving rise to many research questions, some of which we wish to address via master dissertation research. The current thesis topic results from this initiative.

There have been past attempts at text-to-model transformations of user stories. Lin et al. (2014) present an approach were an analysis of user stories is used to refine the top-level goals in a given goal model, i.e., a conceptual model that describes a domain in terms of its stakeholders (which typically includes the software system itself), their goals, and how these stakeholders depend on each other for goal achievement. Mesquita et al. (2015) succeeded in the automatic generation of skeletons of goal models (articulated in the i* language) from user stories. Robeer et al. (2016) built a tool that automatically generates an ontological model of the domain (articulated as an OWL ontology) from user stories. The ontological model shows domain concepts and relationships extracted from the user stories through the use of state-of-the-art Natural Language Processing (NLP) techniques.

The purpose of the thesis is to assess the application of text-to-model transformation of user stories in Agile Software Engineering. How can Natural Language Processing of user stories enhance the process of Agile development? To investigate this research question, the thesis will look into past and current techniques (including those developed in the project with SLU/Testing Algorithms), apply them to example sets of user stories and evaluate their application potential in enhancing requirements engineering. If results are promising, a scientific publication will be extracted from the thesis work, co-authored by the thesis student.

References

Lin, J., Yu, H., Shen, Z., and Miao, C. 2014. Using goal net to model user stories in agile software development. In: Proceedings of IEEE/ACIS International Conference on Software Engineering, Artificial Intelligence, Networking and Parallel/Distributed Computing. Las Vegas, IEEE, pp. 1-6.

Mesquita, R., Jacqueira, A., Agra, C., Lucena, M., and F. Alencar. 2015. US2StarTool: generation i* models from user stories. In: Proceedings of the International i* Workshop (iStar). pp. 103-108.

Robeer, M., Lucassen, G., van der Werf, J.M.E., Dalpiaz, F., and S. Brinkkemper. 2016. Automated extraction of conceptual models from user stories via NLP. In: Proceedings of the 24th International Conference on Requirements Engineering. IEEE. pp. 196-205.

dr. Michael Verdonck

On the 24th of May Michael Verdonck successfully defended his phd “Ontology-driven conceptual modeling: Model Comprehension, Ontology Selection and Method Complexity.”

The Phd of Michael was supervised by Frederik Gailly and the members of the examination board were:

  • Prof. dr. Patrick Van Kenhove (Chairman, University Ghent)
  • Prof. dr. Frederik Gailly (Supervisor, University Ghent)
  • Prof. dr. Geert Poels (University Ghent)
  • Prof. dr. Amy Van Looy (University Ghent)
  • Prof. dr. Sergio de Cesare (University of Westminster, UK)
  • Prof. dr. Ben Roelens (Open Universiteit, Netherlands)
  • Prof. dr. Giancarlo Guizzardi (Free University of Bolzano, Italy)

Congratulations Michael !!!!

PhdMichael

Hurray – UGent Business Informatics has a new doctor!

DOCTOR OF BUSINESS ECONOMICS

Unknown

Laleh Rafati

Capability-Actor-Resource-Service:
A conceptual modelling approach for value-driven
strategic sourcing

Supervisor:
Prof. dr. Geert Poels

Thursday, 25 January 2018

EXAMINATION BOARD

Prof. dr. Patrick Van Kenhove
Dean of the Faculty of Economics and Business Administration
Ghent University

Prof. dr. Geert Poels
Supervisor
Ghent University

Prof. dr. Frederik Gailly
Ghent University

Prof. dr. Ben Roelens
Open University Netherlands

Prof. dr. Jan Verelst
Antwerpen University

Prof. dr. Francesco Polese
Salerno University

Prof. Monica Dragoicea
University Politehnica of Bucharest

DSC_0283

DSC_0325

DSC_0313

DSC_0314 2

DSC_0320

DSC_0326

IMG_1813

Goodbye and congratulations to Dr. Ben Roelens

AAIA_wDGAAAAAQAAAAAAAAyNAAAAJDZjZGQ4Nzg4LTYxYzAtNDZiZi1hYTdkLTQ5YzUxMmEyNDE2MA

UGentMIS wishes to thank Ben for his enthusiastic engagement since October 1, 2011. Ben has accepted a position of Assistant Professor (Universitair Docent) at the Open Universiteit, Heerlen, The Netherlands, and will be affiliated with our team as a parttime voluntary postdoctoral researcher. We wish him all the best with his new career in far-away Holland!

PhD position Business & IT at UGentMIS, Dept. Business Informatics and Operations Management

The UGentMIS research team of the Dept. of Business Informatics and Operations Management will open a PhD position (junior assistant) for research and (limited) teaching in Business & IT (teaching requires mastering of the Dutch language). The candidate is expected to complete a PhD research within 4 – 6 years.

The UGentMIS research team is headed by Prof. Geert Poels, assisted by Prof. Frederik Gaillly (associate professor) and Prof. Amy Van Looy (assistant professor). We are active in the domains of Enterprise Modelling/Architecture and Business Process Management (both from managerial and technological perspectives). The current themes in our research are threefold:

How to ensure that IT delivers business value (with particular emphasis on business (model) innovation through application of emergent technologies)?

How to employ IT to better manage business processes (with particular emphasis on administrative and clinical processes in the healthcare sector)?

How mature are organisations in managing digital innovation in business processes (with particular emphasis on citizen-facing processes in the government sector)?

The UGentMIS professors will help the PhD student to define a PhD research topic related to one of these themes.

Our research approach is pragmatic, encompassing both design-based research and mixed methods research. Reference disciplines include Business/Management/Organisational studies, Computer Science, and Information Systems. The PhD student will be trained (as part of the UGent doctoral training programme) in research methodology and domain-specific topics, and the UGentMIS professors will help the candidate to develop an appropriate research plan.

UGentMIS research is often performed in close collaboration with organisations, like for instance Inno.com, VDMbee, PwC, ISACA, UZGent, Maria Middelares, Collibra, City of Ghent.

Given the multidisciplinary focus and applied nature of our research bridging business and IT, UGentMIS is actively searching for 2nd Master Business Engineering students (all fields of study) or Master students of Complementary Studies in Business Economics, with a background in engineering or computer science. Expressions of interest by other students (e.g., Computer Science, Informatics, Applied Informatics, Business Informatics, Engineering) of Ghent University or other universities (Flanders and abroad) are welcome.

The PhD position will be opened once one or more suitable candidates have expressed their interest.

If interested, contact me at geert.poels@ugent.be
or just talk to me when you see me at the Faculty (open door policy).

Kind regards
Prof. dr. Geert Poels

UGentMIS-ResearchClusters

UGentMIS congratulates Nadejda Alkhaldi and supervisor Frederik Gailly for obtaining the joint UGent-VUB doctoral degree in Business Economics

A fine accomplishment of the UGent-VUB Alliance Research Group in Information Systems (UVIS)

DOCTOR OF BUSINESS ECONOMICS
Nadejda Alkhaldi

On the Symbioses between Conceptual Modeling and Ontology Engineering: Recommendation-Based Conceptual Modeling

Advisors:
Prof. dr. Wouter Verbeke (Vrije Universiteit Brussel)
Prof. dr. Frederik Gailly (Ghent University)
Prof. dr. Sven Casteleyn (Universitat Jaume I de Castello)

Friday, 17 November 2017 at 14h30
in Room D.2.18, Pleinlaan 2, 1050 Brussels