<p>To improve explosive energy efficiency in rock blasting, this study investigates the effect of different aggregate sizes used as stemming material on blasting outcomes at the Kevitsa open pit mine. Four types of aggregates with different size ranges were involved: 8–16&#xa0;mm, 18–25&#xa0;mm, 3–8&#xa0;mm, and 8–25&#xa0;mm. Three full-scale blast tests were conducted in the mine, with each blast divided into a test area and a reference area. Aggregate sized 8–16&#xa0;mm were used in all the reference areas, while the other three types of aggregates served as the stemming material in the respective test areas. In each test, the reference and test areas intentionally employ the same blast design parameters, except the stemming materials. Rock fragmentation and shovel loading rates were measured during muckpile extraction. Fragments in the shovel bucket were captured by the camera mounted on the shovel. By using image analysis, particle size distributions of the whole muckpile were determined. Each blast test was filmed to observe the gas ejection. Results show that the median size (<i>x</i>₅₀) in the test area with 18–25-mm aggregate is reduced by 2.73% and <i>x</i>₈₀ by 3.13%, compared to the reference area with 8–16-mm aggregate. Meanwhile, <i>x</i>₅₀ and <i>x</i>₈₀ in the test area with 3–8-mm aggregate are increased by 7.98% and by 6.44%, respectively, compared to 8–16-mm aggregate. In addition, the loading rate is increased by 1.25% for 18–25-mm aggregate and by 0.6% for 8–25-mm aggregate, compared to 8–16-mm aggregate. Conversely, the loading rate is decreased by 0.7% for 3–8-mm aggregate, compared to 8–16-mm aggregate. From blast videos of field tests, it can be found that areas using large-size aggregates as stemming have fewer blastholes with premature gas ejection than those using small-size aggregates. Therefore, large-size aggregates as stemming contribute to blasting outcomes in the open pit mine.</p>