This research explores the limitations of traditional excavatability assessment methods which predominantly focus on muck pile characteristics and offer limited insights for pre-excavation equipment selection. To address this gap, a novel empirical model was developed using key rock properties and shovel performance indicators, with Total Loading Time (TLT) as a critical metric. Regression analysis revealed a strong linear relationship between TLT and rock properties, with an R-squared value of 0.99 (p < 0.001), highlighting the predictive power of the model. Cohesion and Weight Bulk Density were identified as statistically significant predictors, validated within a 95% confidence interval. Additionally, a new excavatability classification system was proposed, integrating shovel performance and rock properties. Using a K-means clustering algorithm, three distinct excavatability classes were identified, ranging from easily excavated to difficult-to-excavate materials. This research provides a more comprehensive approach for excavation planning, facilitating more accurate equipment selection and operational efficiency in mining practices.