Pedestrian Clearance Blows Up Corridor Coordination
A six-intersection arterial runs a 90-s coordinated-actuated cycle in the PM peak. At intersection 4, phase 2 (arterial through) split is 45 s and phase 4 (minor street) split is 15 s. Over the last two weeks the corridor has lost progression at intersection 4: cycles intermittently run 97 to 99 s, then re-sync, and the pattern repeats whenever someone presses the pedestrian button on the phase 4 crossing. There are no detector failures in the log and no preemption events. The phase 4 crosswalk is 54 ft curb to curb with no refuge island. The controller is a modern NEMA TS-2 unit in coordinated operation with internal time base and a 7-s walk interval.
1. Diagnose why the cycle stretches to 97 to 99 s and progression breaks, using the numbers given.
Read the worked answer
The split is physically too short to serve a pedestrian, so the ring cannot close on time.
Compute the required pedestrian time:
- Walk interval = 7 s (the configured value).
- Pedestrian change interval = crossing distance / walking speed = 54 ft / 3.5 ft/s = 15.4 s.
- Total pedestrian time = 7 + 15.4 = 22.4 s, and the vehicle green for phase 4 must also accommodate minimum green and vehicle clearance.
The phase 4 split is only 15 s. It cannot hold a 22.4-s pedestrian movement. When the button is pressed, the controller does what it is built to do: it will not truncate a pedestrian clearance once the walk has started, so the ring runs long by roughly 7 to 9 s - exactly the observed 97 to 99 s cycles. The coordinated phase then either absorbs the overrun or the offset is lost, which is why progression breaks and then recovers once the overrun clears.
Do not chase the controller programming first. The log is clean on detectors and preemption, and the failures correlate with pedestrian calls - that correlation is the diagnosis. The controller is behaving correctly; the timing plan is wrong.
2. Propose a corrected split and describe how you would rebalance the corridor, including the trade-offs.
Read the worked answer
Set phase 4 to at least 23 s and take the time from somewhere deliberately.
Deficit: 23 s required − 15 s allocated = 8 s.
Options, in order of preference:
- Shorten the crossing. A refuge island or bulb-out reduces 54 ft to roughly 30 ft, dropping required pedestrian time to about 7 + 8.6 = 15.6 s. This is the only option that fixes the underlying geometry and pays for itself over the design life. It also allows a two-stage crossing, which is better for slower pedestrians.
- Reallocate split. Move 8 s from phase 2 (45 → 37 s) to phase 4 (15 → 23 s). Then re-run the time-space diagram: losing 8 s of arterial green reduces capacity and may thin the progression bandwidth to the point that the corridor no longer meets its LOS target. Verify with the peak-hour counts before adopting.
- Lengthen the cycle to about 100 s and re-optimize offsets. This preserves the arterial split but adds delay to every minor street and increases queues at the ramps.
Whichever path is chosen: recompute offsets from the new splits, update every controller in the corridor on the same day, and verify with a travel-time run in both directions during the peak. Document the pedestrian counts that justified the change.
3. A technician proposes reducing the walk interval to 4 s and setting the walking speed to 4.0 ft/s to recover the time instead of reallocating split. Evaluate this proposal.
Read the worked answer
Reject the 4.0 ft/s walking speed outright; treat the 4-s walk only as a narrow exception.
- The 3.5 ft/s walking speed used in the clearance calculation is not a design preference that can be tuned to make the arithmetic work. It is the standard value, and where pedestrians who walk more slowly than 3.5 ft/s use the crossing, the calculation should use an even lower value (3.0 ft/s). A 4.0 ft/s assumption systematically under-provides clearance for older pedestrians, children, and people with mobility disabilities, and it is indefensible in a crash investigation or an ADA review.
- The 7-s walk interval may be reduced, but only in the limited circumstances the MUTCD permits, and only with agency policy support - not as a schedule-driven expedient. Even at a 4-s walk, this crossing needs 4 + 15.4 = 19.4 s, which still does not fit a 15-s split.
- Cutting the walk interval to save time has a second-order effect: pedestrians who start late because the walk is short will still be in the crosswalk when the light changes, generating complaints and, worse, right-turn-on-green conflicts.
Correct response: the arithmetic must drive the decision, not the reverse. Fix the split, or fix the crossing geometry. Then verify the pedestrian clearance in the field with a stopwatch during a.m. and p.m. peaks to confirm the installed timing matches the plan.