Thoughts on the Evolution Towards Model-Integrating Software

Developing software that can be modified and evolved easily is a challenging task. Yet, the fastpaced market requires quick adaptation of products in reaction to emerging requirements. As a basis for flexible software, we proposed to develop software based on Model-Integrating Components (MoCos) in previous work [1]. A MoCo is a non-redundant, reusable and executable combination of logically related models and code in an integrated form where both parts are stored together in one component. The outcome of two comprehensive feasibility studies on building software with MoCos let us conclude that combining the strengths of modeling languages (e.g., abstraction, separation of concerns) and programming languages (e.g., performance) within components yields flexible and wellperforming software, indeed. Even though we provided the MoCo concept as a basis for designing a flexible target architecture, we neglected to introduce process descriptions for engineering Model-Integrating Software (MIS), so far. As a first step towards closing this gap, we introduce the current state of an engineering process for the development of MIS. In addition, we describe a set of evolution scenarios to evolve existing software towards MIS and discuss related evolution processes.