Abstract: As hydrocarbon reservoir development extends into deeper zones and unconventional resources, the importance of productivity forecast technology in development plans optimization and hazards control has become increasingly prominent. Currently, however, the absence of a systematic review on productivity forecast methods has hindered technical integration and progress in this field. Through a systematic review of the evolution of productivity forecast technologies, this study categorizes these methods into four methodological systems, i.e.: comprehensive qualitative analysis of reservoir parameters, statistical learning-based productivity forecast, deterministic reservoir productivity forecast, and digital-intelligent productivity forecast. The fundamental principles, application scenarios, and limitations of each method are analyzed, and a integrated evaluation index system for productivity forecast is proposed. Research shows that the error in reservoir productivity forecast has been reduced from 30% to 5%, and the field is advancing toward data-intelligent integration and intelligent decision-making. In the future, the efforts would focus on promoting interdisciplinary data integration, hybrid modeling, and standardized evaluation system, so as to enhance the forecast accuracy and applicability.