Abstract: With the development of technologies such as artificial intelligence and cloud computing, drilling analysis and optimization technologies gain new development opportunities. To fully utilize massive drilling time-series data and overcome technical challenges such as real-time monitoring of downhole working conditions, recommendation of rate of penetration optimization schemes, and drilling risk prevention and control, an intelligent drilling analysis and decision-making system based on cloud-edge collaboration was developed. Based on cloud-edge model collaborative computation and task scheduling technologies, dual closed-loop decision flows were established, including a risk prevention and control closed loop of “edge early warning–cloud analysis–instruction feedback–model update” and an optimization closed loop of “edge recommendation–cloud optimization–scheme correction–rig site execution”. A three-dimensional soft string model and an LSTM network were integrated to achieve real-time and high-precision prediction of friction torque while drilling (accuracy ≥ 91%); spatiotemporal features were mined based on a CNN−LSTM network to achieve advanced early warning of lost circulation (accuracy>80%); a three-tier model of “threshold−intelligence−dynamic optimization” was utilized to achieve precise identification of 27 working conditions and automatic analysis of time efficiency. The system has been applied on large scale in over 2 000 wells in Sinopec Northwest Oilfield, Zhongyuan Oilfield, Jianghan Oilfield, and other locations. By establishing a remote collaborative decision-making mechanism between rig sites and rear experts, the system successfully realizes comprehensive support for drilling acceleration and safety management.