To quantify the producer development effect in high water-cut reservoirs，we constructed characteristic indexes from numerical simulation and machine learning. The backward flight time of the producer control area was obtained by numerical simulation. The flow heterogeneity of producers was evaluated based on the Lorentz coefficient. The potential index was proposed to describe the potential and displacement capacity of producers in the control area. The copious development performance data of oilfield were collected to establish the time series model. The vector autoregression（VAR）algorithm was used to fit the production history of producers；the production capacity of producers was evaluated through impulse response analysis. The comprehensive evaluation scores of producer were given with the entropy weight method. As a result，the two methods with different assumptions led to similar trends in producer scores. The final score could objectively reflect the development effect of producer with the consideration of the influence of numerical simulation and development performance data. The evaluation method was applied to Y Block of G Oilfield in China to score producers. Eventually，four efficiently developed producers were identified.