Allocating and scheduling resources for a mobile photo enforcement program

Abstract

We present a scheduling model for an urban mobile photo radar speed enforcement (MPE) program. An MPE program utilizes automated photo radar technology to capture speed limit violators, towards an aim to reduce speeding and thus, improve traffic safety. We propose a binary quadratic program that determines where (by location visit tasks) and when (in shifts) to send enforcement resources (operators and equipment) over a month-long schedule. The aim of this program is to minimize violations of enforcement time halos (and thus, efficiency losses) in attaining traffic-safety focused program goals. We solved this problem using a combined Dantzig-Wolfe and column generation optimization approach, after exploring several different types of commonly-used solution methods. The model is applied to a currently operational MPE program in the City of Edmonton, Canada. Using a five-day enforcement time halo, we find that our model results improve resource utilization by 24% over historical program deployment. The scheduling model is the final step of a larger unified MPE program framework that systematically and efficiently connects high-level urban traffic safety goals down shift-level allocations of highly limited enforcement resources. It provides a data-driven and transparent framework that both contributes to cities’ pursuits of Vision Zero and combatting negative public perceptions of MPE programs. Additionally, the framework can be easily adapted for, and transferred to, other automated traffic enforcement technologies across jurisdictions.