Benjamin Nagel

Adapt cases: extending use cases for adaptive systems

Adaptivity is prevalent in todays software. Mobile devices self-adapt to available network connections, washing machines adapt to the amount of laundry, etc. Current approaches for engineering such systems facilitate the specification of adaptivity in the analysis and the technical design. However, the modeling of platform independent models for adaptivity in the logical design phase remains rather neglected causing a gap between the analysis and the technical design phase. To overcome this situation, we propose an approach called Adapt Cases. Adapt Cases allow the explicit modeling of adaptivity with domain-specific means, enabling adaptivity to gather attention early in the software engineering process. Since our approach is based on the concept of use cases it is easy adoptable in new and even running projects that use the UML as a specification language, and additionally, can be easily incorporated into model-based development environments.

Model-driven specification of adaptive cloud-based systems

In cloud-based systems, business process models compose loosely coupled services to fulfill a certain goal. In order to ensure a high quality of service (QoS) of these systems, irregularities and violations need to be detected and handled at runtime by adapting the system. As a consequence, engineering such systems requires an explicit modeling of adaptation rules, which define how the system is adapted under certain conditions. These rules are defined based on models that specify QoS agreements as well as service bindings and business processes that are adapted. In this paper, we present a modeling approach for adaptive, cloud-based systems which supports the explicit specification of adaptation rules. We are introducing a metamodel for specifying bindings between business processes and cloud services considering Service Level Agreements that define the required QoS. In order to express adaptations of business processes on different levels of granularity, we are using an extended metamodel including fragments that group multiple process elements.

Ensuring Consistency among Business Goals and Business Process Models

The paradigm of service-oriented architectures has emerged as an architectural style for designing enterprise applications. Requirements engineering for such applications comprises the specification of business goal models representing stakeholder objectives and the operationalization to business process models that specify the required composition of services. Inconsistencies between business goals and derived business processes can lead to service compositions that are not in line with the actual stakeholder objectives. For preserving consistency it is required to consider logical and temporal dependencies among goals (e.g. the order in which they need to be achieved) in the derivation of business processes. In previous work, we provided a technique for the elicitation and specification of dependencies between business goals. Extending this approach, we aim at validating the consistency between business goal models and business process models regarding these dependencies. In this paper, we present a pattern-based approach for the automated generation of verifiable business process quality constraints from business goal models. We describe how these constraints can be used to check the consistency between business goals and business processes and demonstrate the applicability of our approach in a case study by using the implemented tool support.

Linguistic Consistency of Goal Models

Goal models are used for the elicitation and specification of strategic requirements in early phases of the software engineering lifecycle. By explicitly modeling requirements on a strategic level, these goals provide input for the derivation of operational software specifications. An unambiguous and consistent definition of the goals is the prerequisite for this derivation. Addressing this challenge, this paper presents an analytic approach for the automatic detection of linguistic inconsistencies in goal models. By providing syntactical and semantic consistency conditions, we support requirements engineers by improving the overall quality of goal-oriented requirements specifications. To demonstrate the applicability of our approach, we apply it to three case studies taken from literature using the implemented tool support.

Goal-Driven Composition of Business Process Models

Goal-driven requirements engineering is a well-known approach for the systematic elicitation and specification of strategic business goals in early phases of software engineering processes. From these goals concrete operations can be derived that are composed in terms of a business process model. Lacking consistency between goal models and derived business processes especially with respect to the dependencies between goals can result in an implementation that is not in line with the actual business objectives. Hence, constraints indicated from these dependencies need to be considered in the derivation of business process models. In previous work, we introduced the extended goal modeling language Kaos4SOA that provides comprehensive modeling capabilities for temporal and logical dependencies among goals. Further, we presented an approach to validate the consistency between goal models and business process models regarding these dependencies. Extending the previous work, this paper presents a constructive approach for the derivation of consistent business processes from goal models. We introduce an algorithm that calculates logically encapsulated business process fragments from a given goal model and describe how these fragments can be composed to a business process model that fulfills the given temporal constraints.