Weather Driven Predictive Scheduling for Intelligent Planner and Scheduler

The dynamic and volatile characteristics of weather greatly impact the scheduling decisions in personal planners and schedulers. It often causes inefficiencies and disruptions to our activities, including financial losses. Many researchers have identified that the machine learning algorithms and predictive modeling forecast climatic trends with significant accuracy. This research implements an integrated system that uses real-time API weather data sources and machine learning algorithms for intelligent planners and schedulers. The proposed system provides real-time, adaptable planner recommendations based on environment variables. This paper proposes the Support Vector Machine, XGBoost, Random Forest, Decision Tree, and Logistic Regression algorithm implementations for weather prediction and the potential impact of utilizing these algorithms in an intelligent scheduling assistant. Compared to all models, Random Forest shows the best performance by resulting in the highest accuracy of 0.99 and the lowest error of 0.19 in R² and RMSE metrics.