Cutting force is one of the research hotspots in direct sand mould milling because the cutting force directly affects the machining quality and tool wear. Unlike metals, sand mould is a heterogeneous discrete deposition material. There is still a lack of theoretical research on the cutting force. In order to realize the prediction and control of the cutting force in the sand mould milling process, an analytical model of cutting force is proposed based on the unequal division shear zone model of orthogonal cutting. The deformation velocity relations of the chip within the orthogonal cutting shear zone are analyzed first. According to the flow behavior of granular, the unequal division shear zone model of sand mould is presented, in which the governing equations of shear strain rate, strain and velocity are established. The constitutive relationship of quasi-solid-liquid transition is introduced to build the 2D constitutive equation and deduce the cutting stress in the mould shear zone. According to the cutting geometric relations of up milling with straight cutting edge and the transformation relationship between cutting stress and cutting force, the dynamic cutting forces are predicted for different milling conditions. Compared with the experimental results, the predicted results show good agreement, indicating that the predictive model of cutting force in milling sand mould is validated. Therefore, the proposed model can provide the theoretical guidance for cutting force control in high efficiency milling sand mould. which permits unrestricted use, distribution, and reproduction in any medium, provided you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons license, and indicate if changes were made.the ability of the process, surface quality, and monitoring the tool wear [2]. Milling is one of the most complex machining processes. At present, the models for predicting cutting forces mainly include empirical model, numerical model, artificial intelligence model and analytical model. The empirical model is expressed as an exponential function of cutting force and process parameters, based on the statistical and regression analysis of orthogonal experimental results [3]. FEM is the common numerical method for building the model, which can be used to analyze the chip deformation, tool wear, stress and temperature distribution, and optimize the process parameters. Through artificial neural network (ANN) [4] and particle swarm optimization (PSO) and other modern artificial intelligence algorithms, more accurate milling force model is established. The