During crises like the COVID-19 pandemic, efficient resource allocation is critical for governments. Procuring goods and services through government contracts is vital for effective disaster management. This study analyzes trends in government spending and develops a framework to assist businesses in making strategic decisions when bidding for these contracts. We apply advanced predictive models, including Decision Trees, Random Forests, K-Nearest Neighbors (KNN), Regression, and XGBoost. We then developed a user interface using the Python library Streamlit to provide businesses with an accessible decision support tool based on the proposed framework. The goal of this research is to improve business response times and resource allocation during crises by providing insights into government spending patterns and predicting contract duration. The framework enables businesses to identify successful bidding strategies, facilitate informed decision-making, and contribute to effective disaster management.