建立了应用反向传播学习算法的多层前馈神经网络模型，对混合结构住宅建筑的主要工料消耗进行神经网络计算．以木门木窗和木门钢窗的砖墙住宅建筑为实例的计算结果表明，用神经网络估算住宅建筑工料消耗是非常有效的，借助该模型可实现主要工料消耗的快速估算． A multi-layer feedforward neural network model was established by applying back-propagation learning algorithm. Neural network was used to calculate the main material consumption of mixed-structure residential buildings. The calculation results of the brick-wall residential building with wooden doors and wooden doors and steel windows as examples show that it is very effective to use neural network to estimate the consumption of residential building materials. With this model, it is possible to quickly estimate the consumption of main materials.