Proceedings of the 11th International Conference on Requirements Engineering: Foundations for Software Quality (REFSQ’05). A Goal-Based Round-Trip Method for System Development Gemma Grau 1 , Xavier Franch 1 , Neil A.M. Maiden 2 1 Universitat Politècnica de Catalunya (UPC). c/ Jordi Girona 1-3, Barcelona E-08034, Spain. {ggrau, franch}@lsi.upc.edu 2 Centre for HCI Design, City University. Northampton Square, London EC1V OHB, UK. n.a.m.maiden@city.ac.uk Abstract. In most cases information system development can be seen as an exercise of business process reengineering, either because it automates some human-based processes or because a legacy system is going to be replaced. From this point of view, we can say that the specification of the system-to-be goes from the observation and analysis of the current system to the specification of the system-to-be, going through the construction and evaluation of alternatives. Goal-oriented models are a valuable formalism to support the strategic analysis of the current process. In this paper, we propose a method supporting that round-trip engineering process, focusing in the prescriptive construction of strategic i* models and the systematic generation of alternatives. Several requirements engineering techniques are used in order to model the existing process, which allow a reliable generation and evaluation of alternatives as well as the reuse of strategic knowledge for information system development. 1 Introduction Development of information systems is an activity that seldom takes place from scratch. A new information system may automate some tasks that are undertaken by humans in an organization, or may substitute a system that is becoming obsolete from the organizational point of view. Therefore, most of the times we can say that information systems development and business process reengineering are two views of the same activity and therefore we can reconcile them. From the business process reengineering perspective, the specification of the system-to-be starts from the observation of the current system and the synthesis of its model, the understanding of its rationale, the formulation of new processes or possible ways to enhance the existing ones, the generation and evaluation of alternatives and finally the construction of the detailed target specification itself. We have thus a round-trip from current to ongoing system prescriptive specification, and during this trip we need some support for the intermediate stages: for supporting the strategic analysis of the current system, its weaknesses and strengths, and its alternatives.