Effective bench planning is an essential parameter to achieve productivity, safety, and environmental compliance in the early stage of open pit mine planning. Every mine site establishes its own design criteria based on operational priorities. While safety is universally critical, some sites place more detailed focus on vibration control due to proximity with critical infrastructure, sites, or residential areas. While many engineering software packages now provide automation to generate a bench plan, they often lack integrated consideration for the potential vibration generated from the blast. Peak Particle Velocity (PPV) is a primary metric used to evaluate the impact of blasting activities on nearby receptor sites Key blast design parameters such as burden, spacing, hole depth, and diameter were used to train regression models capable of accurately estimating PPV levels. Building on this predictive model, a multi-objective optimisation algorithm is built to recommend blast designs that minimise PPV while considering spatial and operational constraints. A custom scoring function was constructed using weighted factors, including proximity to vibration receptors, ore and waste classification, and total blast tonnage. Polygon clusters were generated using hierarchical clustering with shape control, and parameter optimisation was performed using simulated annealing to explore high-performing blast configurations under real-world constraints. Results demonstrate that integrating optimised machine learning with spatial analytics and metaheuristic search algorithms enables more consistent compliance with vibration limits, while also improving fragmentation and diggability. The framework provides a scalable decision-support tool for engineers to evaluate bench plans that are both operationally effective and contextually aware of vibration impact in mine operations.