Arndt, S.; Schnieder, E.; Stechert, C.; Vietor, T.:
SDPS 2012 - The Society for Design and Process Science, Berlin, Deutschland, Juni 2012.


Natural language is the most widely-spread means of communication and documentation in the world in all domains of verbal exchange. It thus also is the most readily available form for process and object description in Requirements Engineering (RE). Natural language, however, has several properties like vagueness or ambiguity, which will negatively affect processes of language reception in those contexts that need to rely on specificity and precision. Natural Language Requirements (NLR) are thus often transformed by use of (semi-)formal models, for example written in SysML, which are not commonly shared by all stakeholders and thus might lack transparency to some participants (especially customers). The Institute for Traffic Safety and Automation Engineering (iVA) at TU Braunschweig is therefore engaging an interdisciplinary perspective to strengthen NLRs and the use of natural language in RE processes in its project iglos req – Semantic Requirements Engineering with integrative terminology management. The project’s main focus is on the development of a Plug-In for requirements documentation which will check for syntactic and semantic insufficiencies of such texts and assists knowledge management in form of terminologies. This Plug-In will be based on the already existing web application iglos. The Institute for Engineering Design (IK) at TU Braunschweig aims to support enterprises in developing and offering the right product at the right time. Therefore it develops methods and tools to model complex requirements and their relations to different partial models, such as goal-system, product-architecture, test-cases. A smart analysis of relations help to find conflicts already in early design phases and define solution strategies that avoid iterations and costs of change. In several research projects, such as SFB562, ELVA or KOMSOLV, these methods were used to realize reconfiguration scenarios for parallel robots, risk scenarios for electric vehicles or fast offering of centrifugal pumps. The institutes will closely work together to integrate both approaches to reach their common goals: shorten the development process, writing specifications free of redundancy and conflicts free, re-use proofed knowledge, develop better and safe products.