Offered here is a critique of a simple scheme recently proposed for metering freeway on-ramps. An earlier report of this scheme's potential for reducing commuter delay is shown to be exaggerated. The discussion makes clear that to reduce delay, metering should increase the rates at which commuters exit the freeway. The scheme critiqued here, as well as other well-known metering algorithms, are shown to have deficiencies in this, particularly when the freeway is plagued by a diverge bottleneck with a congested off-ramp. Other more effective schemes for reducing the delay caused by these types of bottlenecks are described. Suitable ways of field-testing freeway traffic management strategies are discussed as well.

Freeway traffic was observed over multiple days and was found to display certain regular features. Oscillations arose only in queues; they had periods of several minutes; and their amplitudes stabilized as they propagated upstream. They propagated at a nearly constant speed of about 22 to 24 kilometers per hour, independent of the location within the queues and the flow measured there; this was observed for a number of locations and for queued flows ranging from about 2,000 to 850 vehicles per hour per lane. The effects of the oscillations were not felt downstream of the bottleneck. Thus, the only effect on upstream traffic was that a queue's tail meandered over time by small amounts. (For the long queues studied here, the tails deviated by no more than about 16 vehicle spacings, as compared with predictions that ignored the oscillations). Notably, the character of queued traffic at fixed locations did not change with time, despite the oscillations; i.e., traffic did not decay.

There were changes over space, however. New oscillations formed in moderately dense queues near ramp interchanges and then grew to their full amplitudes while propagating upstream, even though the range of wave speeds was narrow. The formations of these new oscillations are strongly correlated with vehicle lane changing. But this pattern of formation and growth was less evident in a very dense queue (caused by an incident), although frequent lane changing occurred near the interchanges. It thus appears that the oscillations were triggered by random lane changing in moderately dense queues more than by car-following effects.

Finally, kinematic wave theory was found to describe the propagation of the oscillations to within small errors. For distances approaching one kilometer, and for 2-hour periods, the theory predicted the locations of vehicles to within about 5 vehicle spacings. Further analysis showed that some of these small discrepancies are explained by differences in car-following behavior across drivers.

This thesis unveils a reason for the time-varying patterns in discharge flow that are commonly observed at freeway bottlenecks. Observations reveal that four well-known effects in freeway traffic can interact upstream of a bottleneck in ways that trigger periodic bursts in its discharge flow. A hypothesis concerning the capacity-increasing mechanism is that following an initial "capacity drop", (i) the bottleneck's expanding queue can trigger a new and more restrictive bottleneck further upstream. As a result, (ii) the queue downstream recedes toward the original bottleneck, leaving free flow conditions in its wake. If the tail of this receding queue (iii) passes an on-ramp junction that resides upstream of the original bottleneck, then (iv) a higher bottleneck discharge flow can temporarily ensue. We also hypothesize that this higher discharge rate should diminish after some minutes due to the capacity drop phenomenon. The process would then start anew. Detailed data collected from a 3-km freeway stretch over multiple days support this hypothesis.

The above suggests that the flow through the bottleneck can be increased by metering the freeway's on-ramps in a coordinated but unconventional way. Experiments with this approach consistently produced long-run bottleneck discharge flows that were higher than those generated from a more traditional metering policy. The average difference over multiple days was nearly 3%.

Traffic data near the junction of a single-lane on-ramp (with a ramp meter) and a three-lane freeway were measured for six weekdays during the rush and studied. On each of these days, the merge became a bottleneck with queue discharge rates that were substantially lower than the flows that had passed the merge prior to the bottleneck's activation. On some days, these earlier high flows persisted for many minutes. The bottleneck always occurred when inflows from the on-ramp surged in the presence of high flows arriving from the freeway. Often, the on-ramp surges persisted for no longer than a minute or two and a wide range of these surges was observed. The data show strong correlation between the magnitudes of the on-ramp surges and the merge area's outflows that were measured during the final minutes before the bottleneck activations. These short-run outflows were markedly higher on days when surges from the on-ramp were low. This implies on-ramp metering can be an effective means of postponing this bottleneck's activation, thereby prolonging higher outflows from the merge. Further study of the data indicates that vehicles often maneuvered from the shoulder lane to the center lane at locations just downstream of the merge. It appears that bursts in this lane-changing activity were what triggered the bottleneck's daily activations and that these bursts coincided with surges in inflow from the on-ramp. (These lane changes were evidently negotiated by drivers who had just originated from the on-ramp and/or by other drivers over-taking them). Thus, the benefits of metering inflows at this on-ramp seem to stem from the effects this has on limiting disruptive lane changing downstream.

Measurements taken downstream of freeway/on-ramp merges have verified that discharge flow diminishes when a merge becomes an active bottleneck. We show that metering the on-ramp can recover the higher discharge flow and thereby increase merge capacity. Detailed observations collected using video revealed that the outflow drop following activation was triggered by a queue that formed near the merge in the freeway shoulder lane and then spread laterally, as drivers changed lanes to maneuver around slow traffic. Once restrictive metering mitigated this shoulder lane queue, high outflows often returned to the median lane. Merge outflow could be increased to levels measured prior to the bottleneck activation by then relaxing the metering rate so that inflows from the on-ramp increased. Although outflows recovered in this fashion were unstable and never persisted for periods greater than 13 mins, the findings are the first real evidence that ramp metering can favorably affect the capacity of an isolated merge. The findings point to control strategies that might stabilize outflow and increase merge capacity even more.