Abstract: This study aims to investigate the impact of gravity material density on wall pressure and flow patterns during silo discharge using indoor tests and numerical simulations. A self-developed semi-cylindrical flat-bottomed circular silo was utilized in the indoor tests. The custom pressure sensors were embedded in the silo walls to measure the wall pressure. The cameras were used to record the evolution of flow patterns. Two types of gravity materials were selected with different densities and an average particle size of 5.5 mm: ceramic balls (referred to as ceramic particles) and soybeans. Five discharge tests were then conducted on each material. Subsequently, discrete element method (DEM) numerical simulations were performed to supplement the test observations. The experimental and numerical simulations were combined to explore the effects of gravity material density on the flow pattern and wall pressure. The results reveal that there was a similar flow evolution of ceramic particles and soybeans, indicating the transitions from the mass, funnel, and mixed flow to the tubular flow. There was a dynamic variation in the boundary between the flowing and stationary zones during discharge. The boundary of the stationary zone remained stable at the early and middle stages of discharge, with an angle of 54.03° and a height of 0.310 m. In the late stage of discharge, the flowing boundary on the surface of the gravity material was intersected with the stationary zone boundary at an angle of 28.5°, ultimately forming a "detention zone" near the bottom of the silo with an angle of 28.5° and a height of 0.120 m. The density of the gravity material then dominated the fluctuation of wall pressure. The materials with the higher gravity density caused the more intense variations in the pressure on the silo wall. Among them, the ceramic particles shared a greater pressure amplitude than the soybeans. There was no influence of gravity material density on the location of the peak lateral pressure. The peak points of lateral pressure for both ceramic particles and soybeans were located at approximately 3/10 of the silo wall height from the bottom. During the initial discharge (the first 5 s, accounting for 1.43% of the total discharge duration), there was a sharp increase in the wall pressure of ceramic particles and soybeans, with peak increments of 263% for ceramic particles and 257% for soybeans. At the very beginning of the discharge, the wall pressure increased rapidly in a very short time, inevitably causing an impact on the silo wall. This test verified that the silo failures frequently occurred at the beginning of discharge in the practical scenarios. According to the evolution of the mesoscopic force chain during discharge, the ceramic particles with the higher gravity density exerted greater pressure on the silo wall via the force chain network. The fluctuation of wall pressure was directly related to the breaking, reformation, and arching of the force chains. Furthermore, there was a direct connection between the transmission of the mesoscopic force chain network, force chain rupture, and reformation, as well as the macroscopic distribution of wall pressure. These findings can provide theoretical support and test evidence for the optimal design and safe operation of silo structures. The insights can be gained to mitigate the risks for the high efficiency in the structural integrity of silos during material handling.