<![CDATA[This study analyzes bike-sharing demand patterns and translates the analytical results into an interactive Streamlit dashboard for interpretable decision support. Using the Bike Sharing Dataset for 2011–2012 (17,379 hourly records), the workflow combines data-quality checking, temporal and environmental exploratory data analysis, aggregate usage indicators, variable-level Pearson correlation analysis, and dashboard implementation. The contribution of the study is not a new forecasting algorithm; rather, it is a reproducible and lightweight analytical pipeline that connects interpretable statistical evidence directly to an interactive interface for nontechnical exploration. Results show that rental activity peaks in the late afternoon and under clear weather conditions. Temperature (r = 0.404772) and apparent temperature (r = 0.400929) have moderate positive associations with total rentals, whereas humidity (r = -0.322910) has a weak negative association and weather severity (r = -0.142430) has a very weak negative association. The dashboard passed all defined black-box functional test scenarios. These findings demonstrate how descriptive analytics and interactive visualization can be combined to support transparent exploration of bike-sharing demand patterns while avoiding causal claims.]]>
Copyrights © 2026
