Predicting time to kidney allograft failure (TGF) before transplantation remains a critical unmet need. Previous research has estimated the probability of allograft survival for a specified number of years post-transplant, using both pre- and post-transplant data [1]. In this work, we propose a technology that enables the prediction of graft lifespan pre-transplant, which can inform more effective organ allocation policies. In this work, we introduce a novel artificial intelligence (AI) tool that leverages pre-transplant data to forecast deceased donor allograft survival. Leveraging the open database provided by the Organ Procurement and Transplantation Network (OPTN) [2], we retrieved data on all adult deceased donor, kidney-only transplants performed in the United States. The analysis is limited to data from the years 2000 to 2016 to account for era effect, with 14,090 entries containing complete datasets for the selected date range. The cohort was split into a training set, used to develop the artificial neural network (ANN), and a testing set to evaluate the model's performance. We selected a total of 41 input features, including donor, recipient, and transplant variables that are readily available to clinicians on the day of transplant. The output is the predicted allograft survival in days. The developed machine learning algorithm demonstrated excellent predictive performance on both the training and test sets. To compare the results, the allograft survival was also predicted using additional machine learning techniques. Furthermore, the most relevant transplant variables were identified using the random permutation feature importance method.