One of the most critical problems in the construction sector is the inadequate bearing capacities of subsoils. To solve this problem, various soil improvement methods are employed. Soil improvement is defined as the improvement in soil properties to the desired level by using various methods when the soil is not suitable for superstructure loads. Various types of soil improvement methods exist, and their application depends on the construction site, soil properties, earthquake zone, application time, and cost. One of the most widely used methods recently is the deep soil mixing method. In this study, a laboratory-scale deep soil mixing device is first developed; subsequently, the effects of injection pressure, mixing time, and dosing parameters on application are investigated. Deep soil mixing columns are prepared using different injection pressures, mixing times, and dosages and then subjected to the unconfined compression test. Results show that the effects of injection pressure, cement dosage, and mixing time on the unconfined compressive strength of deep soil mixing samples vary based on the initial soil properties.