This research considers a two-stage planning problem where a fleet of snowplow trucks is first divided among a set of independent regions and then each region designs routes for efficient snow removal. In the first stage, we run routing heuristics to optimize the plowing routes with the goal of minimizing total travel time. Compared with the original routes operated by UDOT, the proposed routes reduce total travel time by 5.04% on average across all regions. In the second stage, we design a custom branch-and-bound algorithm to allocate trucks such that the maximum turnaround time across all regions is minimized. The resulting allocation reduces the turnaround time by more than 20% compared with the original allocation.