Controlled rocking walls are considered an excellent alternative seismic force resisting system for modern resilient cities. This is because of their ability to self-center following seismic events with minimal residual drifts compared to those of conventional walls (i.e. with fixed base). This significantly reduces costs due to service shutdown for structural repairs/replacement. In general, controlled rocking walls depend on unbonded post-tensioning tendons to re-center the wall after loading, thus minimizing or eliminating residual drifts. However, adding post-tensioned tendons in masonry walls create challenges during construction, and prestressing losses are relatively large compared to those in post-tensioned concrete walls. For these reasons, it was desired to investigate an alternative source of self-centering for controlled rocking masonry walls (CRMWs). Recent experimental studies conducted by the authors demonstrated the performance of a new system that relies solely on gravity loads to provide self-centering behavior, while also using supplemental energy dissipation. These walls are termed energy dissipation-controlled rocking masonry walls (ED-CRMWs), and were tested under quasi-static fully-reversed cyclic loading up to failure. The focus of this paper is on a numerical model using OpenSees, based on a multi-spring modelling approach, that simulates the seismic performance of such ED-CRMWs. The model is validated against the experimental results of three ED-CRMWs, which include different confining techniques at the wall compression toes and different energy dissipation locations. The validation results explore the ability of the model to capture the most relevant characteristics of the walls behavior such as load displacement response, lateral load capacity, yield and ultimate displacements, displacement ductility, hysteretic shape and pinching behavior, effective stiffness and energy dissipation at different drift levels for the three specimens.