A School of Continuous Improvement poster sorts pedestrian crossing conduct into four quadrants and reports the Transport Coordination Unit locking each site's signal hold to the resulting quadrant share for the full term.
The Crossing Conduct Index sorts pedestrians at fourteen signalised sites into four quadrants, and the Transport Coordination Unit has locked each site's hold time to its own quadrant history for the term.
A signalised crossing’s timing plan is ordinarily the kind of thing that gets reviewed once every few years, on a warrant count nobody outside Facilities ever reads. At fourteen sites across the University’s affiliated interchanges, school gates and a shopping strip, that cycle has been replaced with something that updates every quarter.
Dr Osei Vandermeer, Research Fellow and Convenor of the Review Cadence Observatory, working with Dr Torun Ezeigwe and Dr Solveig Adeyemi, built the Crossing Conduct Index from eight months of kerbside camera and induction-loop data, scoring every logged crossing on signal compliance and the safety margin it left itself, then sorting the result into one of four named quadrants. The Transport Coordination Unit adopted the quadrant outright: a site’s dominant category now sets its hold duration for the coming term, with no manual override once the recompute runs.
For a School whose Review Cadence Observatory exists to notice exactly this kind of drift, an outside unit adopting one of its instruments unprompted counts as vindication of the underlying method, whatever it ends up doing to the sites that carry it.
A hold time should reward the crossing, not whoever happened to be crossing it the week the camera was watching.
— Dr Osei Vandermeer, Research Fellow and Convenor of the Review Cadence Observatory
Dr Ezeigwe, Senior Lecturer and Director of the Master of Applied Measurement, said the project “gives students an unusually clean case for what a threshold starts doing once an institution stops reviewing it and starts acting on it.”
The full poster is available from the University’s research repository under an open licence, doi:10.5555/slop.appnqy.