Abstract Manager

Seasonal staffing for Snow & Ice Control operations presents a complex planning challenge for large municipalities, requiring the reconciliation of operational demands, collective agreement rules, and diverse employee preferences. At the City of Edmonton, this process historically relied on a manual, opaque, and time-intensive approach, resulting in inefficiencies, limited transparency, and diminished employee trust.

This presentation will cover SmartShift, a data-driven optimization framework that formalizes seasonal workforce placement as a constrained matching problem. SmartShift integrates employee-reported preferences, skills, and seniority with operational requirements for over nine hundred staff across five districts and forty winter crews supporting citywide transportation network reliability. The system consists of four components: (1) a customized survey application capturing employee preferences and qualifications; (2) a data pipeline aggregating operational demand and staffing constraints; (3) a matching model; and (4) an output engine producing assignment lists, dashboards, and postable schedules.

The placement problem is modeled as a constrained variant of the Stable Marriage Problem and solved using a customized Gale-Shapley–based algorithm. Unlike classical formulations, the model incorporates hard constraints such as collective agreement compliance, skill certification, and minimum crew sizes, alongside soft constraints reflecting employee preferences for work location and shift configuration. Job positions act as proposers based on operational priority, while employees’ acceptances reflect ranked preferences, enabling transparent and auditable trade-offs between fairness and service needs.

Early implementation demonstrates improved operational efficiency, increased transparency in placement decisions, and greater alignment between staffing outcomes and employee preferences. By strengthening winter service readiness and reducing administrative friction, SmartShift illustrates how transportation agencies can apply data-driven decision frameworks to workforce planning challenges critical to safe and reliable network operations.