Introduction: Wheat (Triticum aestivum L.) as the most strategic crop for human nutrition is cultivated in many countries under rainfed conditions in semiarid regions. To be vital importance to predict rainfed wheat yield and determine the important factors which affect this crop. Modeling is one of the approaches to predict the response of land to land use. Artificial neural networks (ANN) are considered as one of the modeling approaches to yield prediction and determination of the most important parameters in crop productions. In rainfed wheat hilly land of central Zagros of Iran, there are various parameters that influence this crop production. Therefore, the objective of this study was to identify these important factors.Materials and Methods: This study was conducted for two years and at two sites under rainfed conditions in Koohrang and Ardal districts in Chaharmahal and Bakhtiari provinces, central Zagros of Iran. At both sites, the study was made on farmer–operated winter wheat fields. At the Koohrang and Ardal sites, 102 and 100 sampling points were selected, respectively.202 sampling points were chosen on the landscape covering summit, shoulder, backslope, footslope, and toeslope at two sites with varying climatic conditions. Four parameter groups including terrain attributes, soil physical and chemical properties, precipitation, and weed biomass, including 54 factors were used as the inputs, and wheat grain and biomass yield as the targets for ANN models. A feed forward back-propagating ANN structure was used to develop yield prediction models. The data set was randomly shuffled; 60%, 20% and 20% of them were used for the learning network, testing and verification, respectively. After determination of the best structure of ANN model, crop yields were predicted by the ANN models. By the Hill sensitivity analysis method (Hill, 1998), response of each factor was studied and determined the most effective parameters on grain and biomass yield. This method calculates relative sensitivity coefficient by dividing the sensitivity coefficient of every variable when the variable is reduced 10% by the maximum sensitivity coefficient, therefore the maximum relative sensitivity coefficient is 1.Results and Discussion: The descriptive statistics for various soil characteristics showed that, soil chemical and physical parameters can be classified into three orders. Sand, TN, Kava., Pava., SOM, CCE, and gravel showed high variability (CV>35); clay, silt, and CEC had moderate variability (35<CV>15); and SP and pH indicated low variability (CV<15). The lowest variability was attributed for pH, and the highest is for Pava. Summary statistics of terrain attributes show that the lowest and the highest skewness ascribed to plan curvature and relative stream power, respectively. In this study, the effect of management practices was evaluated by weed biomass. The data on weed biomass percentage ranged from 2.69 to 105.88 Kg ha-1 and this high variability most likely related to farmers’ management practices. Coefficient of variation of grain yield and biomass yield were 49.71 and 42.97%, respectively. It seems that this variability describes to landscape position, various management practices at the two different study sites.The best structure of the ANN models was ascertained for each component yield. Each of the trained structures had 54 input nodes in 4 groups and one output node. The hidden-layer nodes were determined 90 and 50, and the optimum iteration learning rates based on trial and error 9000 and 10000 for grain and biomass yield, respectively. The ANN models for grain and biomass yield resulted in R values of 0.92 and 0.87, respectively, and explained 84% and 76% of the variability in grain and biomass yield, respectively.For grain yield, the sensitivity analysis results showed that, in general, all of the considered parameters were important, but weekly precipitation group was more important for grain yield than the other groups, and precipitation of the 30th, 29th, 10th, 12th, 25th and 13th weeks of the growing season were identified as the mostimportant weekly precipitations. The second important group was management group (weed wt), the third one was soil characteristics group (soil total nitrogen), and the forth one was terrain attributes (plan curvature and sediment transport capacity index). For biomass yield, all parameters were important, but the plan curvature was identified as the most important variable. After plan curvature, weekly precipitation during the 1st, 4th and 9th weeks, catchment area, 25th week, available potassium, profile curvature, 14th and 21th weeks precipitation were the ten top important parameters, respectively.