Optimizing staff scheduling for the Calgary 911 call center requires accurately predicting call volumes and workload to dynamically adjust staffing levels to meet the demand. To achieve this, we have turned to machine learning models that analyze historical call data to capture seasonal trends, fluctuations, and anomalies. The model also incorporates external factors such as major events, including Calgary Stamped and New Year’s Eve, which are known to drive spikes in emergency calls. The predicted call volumes for each half-hour interval are then processed using the Erlang C formula to determine the optimal number of staff needed to meet our target grade of service. The key improvement over the previous system is the dynamic nature of the model. Previously staffing levels were determined using a static model that did not account for variations in call volumes due to seasonality or special events. The new system updates weekly, allowing it to adapt to real-time changes in call volume patterns. This approach not only enhances the resource allocation but also provides schedulers with better insights into high-demand periods. The model allows schedulers to proactively adjust staffing in critical time slots. By improving coverage during peak hours while avoiding overstaffing during low-demand periods, the model enhances operational efficiency, reducing waiting and improves overall workforce management.