A School of Continuous Improvement paper scores more than 238,000 freeway merges across eighteen work-zone sites, and finds that six cameras going dark for maintenance took the campaign's measured gains down with them.
An eighteen-site trial finds a zipper-merge campaign's gains may travel with the camera, not just the sign
A state road agency asked Slop University’s School of Continuous Improvement to find out whether its new zipper-merge campaign was changing how drivers behave at freeway lane closures. Eighteen sites and a purpose-built camera rig later, the School has an answer, complicated by an accident of scheduling nobody engineered.
The finding lands squarely inside the School’s ongoing brief: build the instrument, then ask what the instrument itself might be doing to the answer it returns.
A number that moves when a driver merges better is exactly what we set out to build. What we didn’t expect was how much it also moved when a driver could simply see the camera doing the counting.
— Dr Osei Vandermeer, Research Fellow and Convenor, Review Cadence Observatory, School of Continuous Improvement
Marit Osayande, Lecturer and Convenor of the Living Dashboard, said the rig’s future is already settled regardless of the finding: “We’re not switching anything off. If a lane closure is worth a sign, it’s worth a camera that outlasts the sign.”
The School’s researchers instrumented eighteen freeway lane-closure sites with the vision rig, staggering the campaign’s rollout so that half the network served, for eight weeks, as an untouched comparison for the other half. A maintenance event partway through the trial — a contractor’s routine fix to six of the rigs’ housings — became an unplanned second test of what the instrument was actually picking up.
The full paper is available from the University’s research repository under an open licence, doi:10.5555/slop.x7p1ah.