Michael Eisenbarth

Requirements management for product lines: extending professional tools

The key idea of software product lines is the integrated development of a set of products, exploiting commonalities and variabilities among the products to achieve high levels of reuse. The commercial potential of this approach has already been demonstrated in numerous case studies. However, while requirements management tools are already widespread, the range of professional tool support for product line development is still very poor. In this paper we analyze the question whether and how existing requirement management tools can be seamlessly extended to product line development. We present a general approach, which has been prototyped based on the DOORS requirements management tool and leads to the REMAP-tool extension.

Using Task-oriented Requirements Engineering in Different Domains Experiences with Application in Research and Industry

Early orientation towards tasks of the application domain to be supported by a software system has been proposed as a fruitful means for achieving more appropriate and usable systems as well as for focusing the requirements engineering process. Besides goal orientation, task orientation has therefore been recognized as a promising concept for assuring more completeness and correctness of requirements specifications, and better integration with usability engineering issues. In this paper, we present experiences made with the task-oriented requirements engineering framework “TORE” in four different case studies. These case studies were selected based on their contextual specifics, like service orientation or ambient intelligence, that might impose certain challenges on task-oriented RE. As a lesson learned, we experienced TORE to be highly beneficial even in systems that do not seem to be “traditional” information systems at first glance. However, we also identified limitations that call for necessary adaptations.

Using Ontology-Based Reference Models in Digital Production Engineering Integration

Abstract In the digital planning process of a manufacturing plant several partners like OEM, prime contractor and further service providers are participated usually. Since the partners have partially overlapping views (electricity, mechanical structure, plant controlling) on the same plant to be produced, they have to exchange data during their collaboration. However the syntactical, structural and even semantical differences of output and expected data of partners makes a data integration necessary. Whilst dealing with syntactical and structural differences is rather easy, overcoming semantical differences has several obstacles. In this paper we introduce our ontology based reference model approach to solve this problem.

Effect-oriented requirements elicitation and specification

To support the early evaluation of the effects of a novel system on its environment, it is important to understand and analyze current deficits and problems within the application domain and to derive hypotheses on the potential impact of the system. An emerging application domain of such cause-effect oriented requirements engineering methods is the automotive domain. New upcoming functionalities in automobiles and traffic systems do not only require accurate development and consideration of safety-related aspects, but are also more and more constrained to proving their positive effect on existing traffic systems, ecological and sociological aspects, as well as, e.g., future reduction of the number of traffic deaths. Based on a systematic elicitation and refinement of system requirements, explicit measures can be derived to evaluate the achievement and coverage of envisioned system goals and to identify further optimization goals. In this paper, we will introduce a requirements engineering method and specification guideline that supports the deductive derivation and specification of measures from a causal chain of goals, deficits, hypotheses and solutions for evaluating a systems effect on its application environment. We already applied this approach in a research project in the automotive domain and will exemplarily describe the realization of a concrete vehicle function that interacts with the environment. At the end, we will provide a general conclusion regarding our first experiences with the described method and with the specification guideline.