Production Technology method can be used to assign different indexes to each part of the reservoir based on their relative reservoir quality (RRQ). Figures 2A and 2B show the distribution results of fuzzy pattern recognition analysis for six months and three years of cumulative oil production, respectively. The darker colors represent areas with higher cumulative production (excellent RRQ) during the specified time interval. It should be noted that RRQ is a good indicator of anticipated cumulative production (during a time interval) for a newly drilled well in a specific area of the TABLE 1 Inputs for Top-Down Model Development Static data (reservoir properties) Static data (well design parameters) Dynamic data Pay thickness Porosity Water saturation Porosity (offset well) Neutron porosity (log) Formation top Well location Well drainage area TVD Perforation thickness Volume of total injected fluid Weight of injected proppant Distance to offset well Well drainage area (offset) Time Days of production q-oil (t-1) q-oil (t-2) Date of stimulation FIGURE 3 KPI Analysis Results for Relative Influence of Different Inputs on Monthly Oil Production Rank Feature % Degree of Influence 1 2 3 4 5 6 7 8 9 Stimulation Date q(t-1)-Oil q(t-2)-Oil Perf Thickness Weight of Inj Proppant Porosity (%)(1P) X/Longitude Porh(1-sw)(1P) Porh(1-sw)-Ave.(1P) Water Saturation (%) Elevation Vol of Total Inj Fluid Porh(1-sw)-Ave. Time Resistivity Porosity (%) Porosity (%)-Ave. Top (ft) Distance (1P) Days of Production (t) Top (ft)(1P) Y/Latitude 46 42 42 40 40 38 37 37 37 36 33 33 32 Time-Based Reservoir Model The most important aspect of data driven analysis for the oil production of this asset was developing a time-based reservoir model. This model used pattern recognition techniques (artificial intelligence) and honored reservoir engineering practices to regenerate production histories and predict the future performance of existing and new wells. In order to make an artificial intelligence-based reservoir FIGURE 4A TDM Production History Estimates/Predicted Production (Entire Field) 100 83 79 60 55 49 48 46 46 10 11 12 13 14 15 16 17 18 19 20 21 22 reservoir. In addition, it is possible to determine a production indicator (PI) for each well and compare it with the RRQ of the area in which the well is located. If the PI is less than the average RRQ, the well can be considered as underperforming, making it a good candidate for restimulation. The right-hand panel in Figure 2C shows the locations of underperforming wells, based on three years of cumulative production (the wells are denoted by the blue circles). FIGURE 4B TDM Production History Estimates/Predicted Production (Well 0512321700) JANUARY 2015 163