The pursuit of sustainability in manufacturing often involves balancing conflicting goals, such as minimising production time or reducing energy consumption to meet environmental demands. Additive Manufacturing (AM) processes, as Wire Arc Additive Manufacturing (WAAM), face particular challenges in this regard because of their complexity and sensitivity to process parameters. Nevertheless, Artificial Intelligence (AI) provides valuable support for process optimisation planning, as it enables the application of advanced parameter search algorithms in conjunction with ML-based models to maximise given reward functions. In this regard, this study proposes a data-driven framework that integrates AI models into slicing software to predict layer geometry and specific energy consumption (SEC), thus enabling multi-objective optimisation of WAAM parameters and consequent generation of the path planning strategy. By incorporating both process parameters and scanning strategy variables into a geometry-specific optimisation framework around the slicer software, the user can achieve component-specific planning optimisation with minimal intervention. A prototype production of an Invar 36 alloy impeller showcases the potential of the framework to support sustainable and scalable production planning, aligned with modern green manufacturing paradigms, particularly when dealing with complex geometries produced via WAAM.