BUBBLE POINT PRESSURE (PB) IS ONE OF THE MOST IMPORTANT PROPERTIES OF CRUDE OIL. SUBSTANTIALLY, PB IS DETERMINED LABORATORY PVT TESTS. HOWEVER, IN MANY CASES, LABORATORY DETERMINATION OF PB IS IMPOSSIBLE FOR SEVERAL REASONS. IN ADDITION, LABORATORY METHODS ARE VERY EXPENSIVE AND TIME CONSUMING. THEREFORE, IN SUCH CONDITION, A FAST AND CHEAP METHOD COULD BE USEFUL FOR PB PREDICTION. THE ARTIFICIAL INTELLIGENCE COULD BE A SUITABLE CANDIDATE METHOD FOR THIS PURPOSE. IN THIS STUDY, Adaptive NEURO- FUZZY Inference System (ANFIS), WHICH IS ONE OF THE ARTIFICIAL INTELLIGENCE TECHNIQUES, WAS APPLIED FOR PB PREDICTION. A TOTAL OF 429 DATA SETS OF DIFFERENT CRUDE OILS MIDDLE EAST RESERVOIRS WERE USED. DATA SETS INCLUDE PB AND CONVENTIONAL PVT PROPERTIES. AMONG THE DATA SETS, 286 DATA SETS WERE SELECTED RANDOMLY FOR CONSTRUCTING THE GENETIC ALGORITHM, AND THE OTHER INCLUDED 143 DATA SETS WERE USED FOR MODEL TESTING. THE CORRELATION FACTOR (R2) BETWEEN PREDICTED PB BY THE ANFIS MODEL AND THE EXPERIMENTAL PB IN THE TEST DATA WERE 0.87 WHICH SHOWS A GOODISH AGREEMENT BETWEEN THE PREDICTED VALUES AND EXPERIMENTAL VALUES.