Context. Cloud-based systems offer unique elasticity in the number of computational resources they provide. When modeling these systems, architects face the challenge of engineering elasticity to determine how to manage resource allocation. Too few resources lead to unreliable service, while too many result in unnecessary expenses. To address this, autoscaling policies can be employed to automatically adjust resource availability based on predefined conditions. Problem. Engineering elasticity is a complex task. When designing autoscaling policies, many antipatterns [SEK+23] can emerge, which are difficult to detect at design time. A system is needed that integrates simulation data to guide architects toward problematic components in their model that may lead to antipatterns. Objective. To facilitate this process, previous work [Hah23] proposed the development of a graphical editor that provides visual feedback from simulation data. Building on this foundation, we enhanced the system with textual explanations generated by LLMs to guide developers toward a deeper understanding of the SPD-model and aid in identifying antipatterns. Method. We extended existing implementations of a graphical editor for SPD-models that provide visual feedback on the design. A new feature was introduced that leverages large language models to generate textual explanations as feedback. We conceptualized different types of explanations a user might seek when working with SPD-models. Additionally, we designed templates for generating prompts that request these explanations from the LLM. Finally, we evaluated the quality of the generated explanations to assess their potential in helping developers identify and resolve antipatterns at design time while gaining a deeper understanding of the SPD-model. Result. We delivered a functional prototype that enhances previouswork with our newly implemented explanation feature. Additionally, we evaluated the quality of the generated explanations, justifying their integration into the editor. We demonstrated that the explanations are of sufficient quality to assist architects in understanding SPD-models and identifying antipatterns at design time, although there is still room for improvement. Conclusion. Experts rated most kinds of explanations with a promising rate of usefulness. This implies that architects could benefit from the explanations in identifying antipatterns, creating fixes, and gain an overall deeper understanding of the policies. However, the explanations also have problems with imprecisions and lack of conciseness in some cases. Fine-tuning the prompts, contriving new types of explanations, and advancements in LLM-technology could help reduce these problems.