Yield forecasting is becoming increasingly important in the context of climate variability and change using approaches like remote sensing and crop modeling. Climate variability and change are affecting crops and efficiency of input resources. The situation is demanding efficient management of input resources. Changing climate is also affecting current production technology which needs modification using modern tools. Two field experiments were conducted at Water Management Research Center, University of Agriculture Faisalabad, Pakistan. Field experiments addressed above mentioned issues. The first experiment included full (100%) and three reduced levels (80%, 60% and 40%) of irrigation with four levels of nitrogen (160, 200, 240 and 280 kg ha-1) at different critical growth stages of maize. Second experiment involved four sowing dates (i.e.27 January, 16 February 8 March and 28 March) and three maize hybrids (i.e. P-1543, DK6103 and NK8711) during the years 2015 and 2016. Different system approaches were used to optimize the volume of irrigation, amount of nitrogen (N) and sowing date for maize hybrids. CERES-Maize model was calibrated and evaluated with second experiment. Calibrated model was used to explore the effects of climate variability and climate change (CC) at regional scale. Different adaption strategies were developed to mitigate the negative effects of climate change. Remote sensing framework was used for regional yield forecasting that can assess seasonal and interannual variability. In first experiment results from model and economic analysis showed that the N rates of 235, 229, 233, and 210 kg ha-1 were the most economical optimum N rates to achieve the economic yield of 9321, 8937, 5748 and 3493 kg ha-1 at 100%, 80%, 60% and 40% irrigation levels, respectively. The optimum level of irrigation was 250 mm. The results of second experiment revealed that grain yield was continuously decrease with delay in Planting date, among maize hybrids Poineer-1543 performed best in spring season. Model parametrization results showed a reasonably good result in prediction of biological and grain yield with RMSE values of 963 kg ha-1 and 451 kg ha-1. Different GCMs were used for understanding the CC impacts, which indicated that there would be increase of 3.4°C in maximum and 3.8°C in minimum temperature in hotdry GCM. The reduction in maize yield due to rise in temperature will be 27% under mid-century (2040-2069). Different adaptations options could be used with RAPs then maize yield would be increased by 15%.Landcover classification of maize were done by Machine Learning algorithms which estimate 14% less area reported by Reporting Service (CRS) of Punjab Pakistan for 2015 and 2016. For yield forecasting, seasonal multitemporal, a total of 8 LST and NDVI values for 64 farms were taken to develop model. Model was used to predict the yield of previous 10 year (2007-2016) which showed a high accuracy with mean % error of 1.25. Seasonal mean Tmax and Tmin of 10 years with predicted yield showed a negative relationship with Tmax. (R2= 0.76) and Tmin. (R2= 0.69). It can be concluded from the study that modern tools are very helpful to optimize input resources to ensure food security.