Abstract: Many offshore oil fields in China have entered the middle or late stage of exploitation, the equipment and facilities are generally aging, and the breakdown probability is increasing. Due to the high cost of production and maintenance of offshore oil and gas wells, it is very important to identify the potential breakdown of oil and gas wells accurately. By optimizing the standard SPC (statistical process control) chart, we established an appropriate early warning model with production parameter for offshore oil and gas wells, by which to the abnormal production parameters of offshore oil and gas wells can be judged dynamically. At the same time, combined with the business experts experience, we proposed a fault diagnosis model to predict the breakdown types based on weighted decision tree and multi-parameter combination. The early warning model improved the timeliness and accuracy of breakdown identification of offshore oil and gas wells. Application of the model in oil and gas wells in the western South China Sea shortened the response time to the countermeasure by at least 8.4 days, and the success rate of early warning reached 91.18%. Therefore, the model can realize the intelligent identification of oil and gas well production abnormalities and breakdownof oil-and-gas wells, and ensure the high and stable production of offshore oil and gas wells.