No abstract

Denver Public Schools (DPS) serves roughly 90,000 K–12 students using a mixed bus fleet. Developing and reviewing bus-route assignments manually has been challenging and time consuming for DPS. During 2017–2018, DPS analysts reviewed and adjusted over 700 routes assigned to approximately 200 buses, considering time and capacity feasibility. We developed a decision support tool (DST) to generate feasible bus-route assignments and help inform DPS’s decisions. The DST employs optimization models to solve the bus-route assignment problem using distance data from Google Maps Application Programming Interface and various interroute reposition-time scenarios to account for the impact of potential traffic delays. The model incorporates multiple objectives related to minimizing cost, meeting demand, and maximizing “consistency”—that is, the difference between a newly created and previously implemented solution The solutions generated by the DST for the 2017–2018 school year utilized significantly fewer buses and lower reposition mileage compared with the DPS solution. Considering the convenience, efficiency, and flexibility of generating high-quality bus-route assignments using the DST, the DPS transportation team has used the DST in the route planning process since 2018.