Modern automatic transmission (AT) application software validation is predominantly dependent on Hardware-in-the-Loop (HiL) systems, resulting in significant time, cost, and scalability limitations within continuous integration/continuous deployment (CI/CD) workflows. These constraints intensify with increasing software complexity and the need to support multiple transmission variants across evolving hardware configurations. Existing validation approaches lack a scalable, software-centric surrogate capable of representing both control logic and physical system behaviour under real-time operating conditions.
This study proposes a digital twin-based validation framework for AT application software, integrated within CI/CD pipelines. The framework restructures application software into a platform-based architecture enabling reuse across multiple transmission programs. A coupled plant model is developed to simulate drivetrain dynamics, ensuring system-level fidelity. The architecture also permits future data-driven augmentation for transient response correction and extended scenario coverage; however, the present study focuses on the baseline digital twin, hardware abstraction, and CI/CD validation workflow.
A key contribution is the introduction of a hardware-adaptive abstraction layer, enabling the digital twin to accommodate variations in sensors, actuators, and control interfaces without full HiL dependency. Across 15 scenario hardware validation combinations, the reduced-order digital twin achieved a runtime gain of 1.20×–1.68× over the HiL proxy execution, with an average gain of approximately 1.46× and an average execution-time reduction of approximately 31% per validation case.
The study demonstrates that integrating application software, plant models, and hardware-aware abstraction within a digital twin enables a transition from hardware-constrained validation to software-scalable paradigms. The framework provides a pathway toward efficient and adaptable validation in software-defined vehicle ecosystems. The proposed framework is aligned with ZF’s platform-based transmission software strategy, enabling reuse across multiple automatic transmission programs and supporting scalable validation of ZF powertrain control systems under evolving hardware configurations.