Streamlining Acceptance Test Generation for Mobile Applications Through Large Language Models: An Industrial Case Study

TL;DR

This paper tackles the bottleneck of acceptance test generation for mobile apps, especially in cross-platform Flutter environments. It presents AToMIC, a modular framework that uses locally deployed, task-specialized LLMs to automatically generate Gherkin scenarios, Page Objects, and executable UI test scripts from Jira requirements and code changes, integrated into CI/CD workflows. The industrial case study on BMW's MyBMW app demonstrates high artifact quality and substantial time savings (e.g., ~259 seconds per issue locally, with cloud-based speeds potentially 10x faster), along with strong practitioner confidence in adoption. Key contributions include end-to-end pipeline design, formal Page Object generation methods, navigation modeling, tailored prompt strategies for privacy-conscious deployment, and a validated implementation across 13 real-world issues. The work has practical significance for scaling acceptance testing in large mobile projects and lays groundwork for broader platform applicability and further automation enhancements.

Abstract

Mobile acceptance testing remains a bottleneck in modern software development, particularly for cross-platform mobile development using frameworks like Flutter. While developers increasingly rely on automated testing tools, creating and maintaining acceptance test artifacts still demands significant manual effort. To help tackle this issue, we introduce AToMIC, an automated framework leveraging specialized Large Language Models to generate Gherkin scenarios, Page Objects, and executable UI test scripts directly from requirements (JIRA tickets) and recent code changes. Applied to BMW's MyBMW app, covering 13 real-world issues in a 170+ screen codebase, AToMIC produced executable test artifacts in under five minutes per feature on standard hardware. The generated artifacts were of high quality: 93.3% of Gherkin scenarios were syntactically correct upon generation, 78.8% of PageObjects ran without manual edits, and 100% of generated UI tests executed successfully. In a survey, all practitioners reported time savings (often a full developer-day per feature) and strong confidence in adopting the approach. These results confirm AToMIC as a scalable, practical solution for streamlining acceptance test creation and maintenance in industrial mobile projects.

Figures (8)

• Figure 1: MyBMW's Three-Tiered Architecture
• Figure 2: UI Test Development Workflow in the MyBMW Project
• Figure 3: High-Level Activity Diagram of the AToMIC System
• Figure 4: Activity Diagram for Page Objects Generation
• Figure 5: Activity Diagram for Gherkin Scenario Generation