Requirements Engineering for Ethical Systems: An Ontology-based Approach
DR. RENATA GUIZZARDI SILVA SOUZA
ASSISTANT PROFESSOR, UNIVERSITY OF TWENTE.
The advent of socio-technical, cyber-physical and Artificial Intelligence (AI) systems has broadened the scope of Requirements Engineering which must now deal with new classes of requirements, concerning ethics, privacy and trust. Unfortunately, requirements engineers cannot be expected to understand the qualities behind these new classes of systems so that they can conduct elicitation, analysis and operationalization. To address this issue, we propose a methodology for conducting requirements engineering which starts with the adoption of an ontology for a quality domain, such as ethicality, privacy or trustworthiness, populates the ontology for the system-to-be and conducts requirements analysis grounded on the populated ontology. We illustrate our proposal within the ethical domain, using the case of a driverless car.
Renata is currently an Assistant Professor at the Behavioral, Management and Social Sciences Faculty, (IEBIS group) of the University of Twente, in the Netherlands. Moreover, she is a founding member of the Ontology & Conceptual Modeling Research Group (NEMO) and of the Laboratory of Supporting Technologies for Collaborative Networks (LabTAR), at UFES, Brazil, where she held an Associate Professor position from 2009-2016. In 2006, she received her PhD in Computer Science from the University of Twente in the Netherlands with a thesis entitled Agent-Oriented Constructivist Knowledge Management. She has also been a visiting scholar at the Bruno Kessler Foundation (FBK) in 2006 and 2018, and at the University of Trento in 2013-2014, Italy. In 2018, she was an Senior Associate Researcher at the University of Bristol, UK and in 2019, she was a lecturer at the University of Trento, Italy. Her main research interests include Knowledge Management, Ontologies and Goal-oriented Modeling.
A Multiperiod Drayage Problem with Customer-dependent Service Periods
DR. ALI GHEZELSOFLU
POSTDOCTORAL RESEARCHER, UNIVERSITY OF TWENTE
We investigate a multiperiod drayage problem in which customers request transportation services over several days, possibly leaving the carrier some flexibility to change service periods. We compare three approaches for the problem: a path-based model with all feasible routes, a “Column Generation” algorithm in which the pricing is formulated as a collection of shortest path problems in a cunningly constructed acyclic network, and a compact arc-flow formulation based on this network. The experiments show that the latter formulation is the most efficient, and can solve optimality instances of real-world size (and beyond) in time compatible with typical operational constraints. Also, the models allow us to assess that limited amounts of flexibility from customers can significantly improve routing costs for the carrier while decreasing customers’ costs as well.
Ali Ghezelsoflu is a new postdoctoral researcher within the Department of Industrial Engineering and Business Information Systems at the University of Twente (Enschede). He is working on a project "entitled Reinforcement Learning platform for Small and Medium-sized Enterprises (ReAL)". He holds a bachelor's degree in Applied Mathematics(2009) and hi did his master’s degree in Operations Research (2012) at the K.N.Toosi University of Technology, Tehran, Iran. He did his Ph.D. at the Department of Mathematics and Computer Science, University of Cagliari (UNICA), Italy (2018). His Ph.D. was on the Vehicle Routing Problems; their different models, algorithms, and solution methods. After finishing his Ph.D., Ali remained one year more as a Postdoc researcher at UNICA to complete his studies on the VRPs. From the beginning of 2019, he has been started his postdoc at the University of Pisa (UNIPI), Department of Informatics, where he researched Energy Optimizations under the European horizon 2020 project entitled as Plan4Res. During the three years of Postdoc at UNIPI, he has been working as a developer on the SMS++ which is a set of C++ classes intended to provide a system for modeling complex, block-structured mathematical models for Energy Systems, and solving them via sophisticated, structure exploiting algorithms such as decomposition approaches and structured Interior-Point methods. His research interests also contain Energy Optimization, Network Optimizations, Linear and Non-Linear Programming, Mixed Integer Programming, Stochastic Programming, Robust Optimization, Transportation and Logistics Management, Vehicle Routing Problems(VRPs), Exact and Heuristic approaches for IP Models, Decomposition methods for IP Models.