Extraboard-Driver Workforce Planning for Bus Transit Operations
Metro Transit, an agency of the Metropolitan Council, is the dominant public-transit provider in the Twin Cities metro area, providing approximately 95% of all transit service. Budgetary pressures on public financing of transit operations have increased in recent years, leading to greater scrutiny of operational efficiency. Because labor costs are the single largest line item in transit agencies’ budgets, there is strong interest in improving transit-workforce efficiency. However, optimizing efficiency is difficult because the demand for bus service varies by time of day, day of the week, and time of year, and can be affected by unanticipated events. A common practice for coping with this supply-and-demand variability is to have a certain fraction of the transit workforce serve as extraboard drivers—that is, drivers who are not assigned beforehand to regular routes or schedules. Although extraboard drivers are essential for delivering reliable bus service, this flexibility comes at a price. Extraboard drivers are guaranteed full pay for their shift, even though a significant portion of their time may be spent “standing by” while they await an assignment. In addition to scheduling extraboard-driver assignments, transit authorities also match supply and demand by making use of overtime when needed. Thus, a key challenge for Metro Transit management is to find the right balance between the size of the extraboard-driver workforce and reliance on assigning drivers to work overtime, while maintaining their ability to provide all of the needed transit trips for a given day. The research reported in this article, performed in collaboration with Metro Transit, had three broad objectives: to develop performance metrics that help monitor bus transit workforce efficiency and support better transit workforce planning; to analyze the tradeoffs between cost and reliability in the choice of extraboard-driver workforce-management practices; and to develop mathematical models that could help improve operational efficiency in assigning and scheduling extraboard drivers.