Management of supplemental irrigation in humid regions is critical because both over and under estimation of the irrigation water requirement can cause cotton lint yield reduction. Spatial variation of the soil physical characteristics is significant in west Tennessee hence precision irrigation strategies should be applied to achieve the optimum lint yield. Despite the significant enhancement in instrumentation and measurements, there are several challenges that need to be addressed for precision irrigation scheduling and management. Soil hydraulic properties are essential inputs for irrigation scheduling but direct measurement of them is time-consuming and costly. Also, the mathematical relationship between yield and applied irrigation should be carefully investigated by means of a precise model. Overall, the individual roles of the irrigation, soil and climate and possible interaction among them should be accurately examined in both temporal and spatial manners to be able to identify the irrigation zones and the optimized irrigation quantity at each zone. The sophisticated crop models are able to precisely model the crop-soil-water relationships yet are mostly impractical in reality since they need numerous input variables. On the other hand, empirical regression models could be run with less input data but are not able to adequately project the complex nonlinear systems. This study aimed to address these challenges by introducing an integrated experimental-data mining approach which only makes use of easy-collected site specific data hence is practical. The proposed framework provides a step by step plan to answer critical questions such as: (i) how to generate reliable soil maps which contain soil physical and hydraulic properties information using available soil data in precision agriculture domain, (ii) how to identify the role of irrigation on yield at different parts of a field in practice, (iii) how to quantify the contribution of soil, water and climate on yield variation each year based on the generated yield map and finally (iv) how to analyze the above data to zone a field for precision irrigation. One top of introducing the details of our framework we would like to share our finding on the results that we have achieved so far by implementing the proposed framework in a case study.