Every country is in some sort of rivalry with its neighboring countries. Military forces safeguarded people's lives day in and out. The forces are being placed in many zones near the hotspots of the border areas to fight off enemies. These zones are critical and challenging in many ways, which put the militia's life at risk. The main goal of this research is to save the militia's life. This paper proposes an Environment Condition-based Resource Allocation in Military (ECRAM), which helps in decision-making and in providing a report to aid the militia in a war zone. This ECRAM Model analyzes many factors like climatic condition, resource availability, and health status of the militants at the base station. It was implemented by comparing different machine learning approaches to analyze the appropriate one for the proposed model. Results show that the accuracy of prediction and decision-making with the random forest algorithm has outperformed with an accuracy of 94.4 percent.