Energy efficiency in manufacturing plays a crucial role in decreasing manufacturing costs and reducing environmental footprint. This is particularly important for producing battery cells with novel processes due to their cost-sensitivity and high potential impact on the environment. Therefore, design and operation of these processes are critical and require a high level of process and machine specific understanding. A methodology based on machine learning is presented, which has the capability of identifying improvement potentials using machine and process specific influencing factors. A battery production case is used to demonstrate the accuracy, transferability and validity of the methodology.