Software bugs are considered inevitable in the software life cycle. Regardless of their cause, software stakeholders could benefit from knowing the time to fix them to enhance software management processes and liaise with developers to handle software pipeline tasks better. In this paper, we use neural network machine learning approach with non-linear regression model to conduct a prediction of bug-fix time retrieved from actual bug fixed records taken from 9 bugs reports from two open-source projects. The paper evaluates the bug-fix time prediction results by using four regression metrics. The result of the nonlinear regression testing showed a high correlation between the adjacent distributions of the bug samples time by giving high prediction accuracy rates. Additionally, we found that the module gave better results for samples with fewer bug records with the condition of fair distribution with respect to the fix time values.