This paper describes the collection and use of 1.4 million travel time records that were collected over a 12-week period in 2009 to evaluate and communicate quantifiable travel mobility metrics for a rural interstate highway work zone along I-65 in northwestern Indiana. The effort involved the automated collection and processing of Bluetooth probe data from multiple field collection sites, communicating travel delay times to the motoring public, assessing driver diversion rates, and developing proposed metrics for a state transportation agency to evaluate work zone mobility performance. Collected travel time profiles were compared with traditionally measured hourly flows in both incident and nonincident conditions. Through the 12-week period over which work zone performance was measured, the work zone had 422 h of congested conditions in which travel time delay was greater than 10 min. Despite the display of real-time delay measurements to the motoring public through portable dynamic message signs, a negligible percentage of the travel probes were observed to divert in advance of the congested work zone through self-guidance. Implementation of a targeted alternate route starting the weekend of July 24 resulted in an increase of observed probes diverting along the trail-blazed route from none to more than 30%. The paper concludes by suggesting that acquisition of work zone travel time data provides a mechanism for assessing the relationship between crashes and work zone queuing. Real-time monitoring of these travel time data may also enable future contracts to include innovative travel time reliability clauses.

Route choice is often assessed with either a modeling technique or field observations. Field observations have historically used a variation of license plate matching. The proposed technique assesses route choice and travel time that uses an anonymous Bluetooth media access control (MAC) address sampling technique as a surrogate for license plate matching to assess route choice. The Bluetooth sampling technique was used to evaluate the impact of an unexpected bridge closure in northwest Indiana, including an assessment of the proportion of vehicles using each of four alternate routes. The Bluetooth technology also provided a means to collect travel time data for each alternate route; these observed travel times were also compared with travel time estimates obtained by route classification and link distance. In general, the route choice behavior was consistent with observed travel time estimates. The Bluetooth sampling technique is cost-effective to deploy, and although results are approximate, direct measurement of travel times and route choice is useful for public agencies to assess mobility and travel time reliability along alternate routes.

Signal offsets are a signal-timing parameter that has a substantial impact on arterial travel times. The traditional technique is to optimize offsets with an offline software package, implement the settings, and then possibly observe field operations. It is not uncommon for a traffic engineer to fine-tune the settings by observing the arrivals of platoons at an intersection and making adjustments to the offset from this qualitative visual analysis. This paper discusses two tools to assist the engineer in managing arterial offsets. First, it introduces the Purdue coordination diagram (PCD) as a means of visualizing a large amount of controller and detector event data to allow investigation of the time-varying arrival patterns of coordinated movements. The second technique is arterial travel time measurement by vehicle reidentification via address matching by Bluetooth media access control. This technique is used to evaluate existing offsets and assess the impact of implemented offset changes. These tools are demonstrated with a case study involving a before-and-after comparison of an offsettuning project. PCDs were used to identify causes of poor progression in the before case, as well as to visualize both the predicted and the actual arrival patterns associated with the optimized offsets. More than 300 travel time measurements from Bluetooth probes were used for statistical assessment of before-and-after travel time. The statistical comparison showed a significant (at the 99% level) 1.7-min reduction (28%) in mean northbound travel time, corresponding to a 1.9-min reduction in median northbound travel time. Southbound travel times were not negatively affected by the offset changes.

Typically, speed limits are reduced in work zones to safely accommodate construction activities and motorists on the roadway. This paper presents a methodology to evaluate the temporal and spatial effects of techniques designed to encourage compliance with work zone speed limits. The evaluations were performed over short and long segments within and adjacent to an Interstate construction work zone in suburban Indianapolis, Indiana, with the use of vehicle probe data. Space mean speed was measured by using 13 Bluetooth probe data acquisition stations, which provided a random sample of unique identifiers for approximately 11% of the passing vehicles. These space mean speed data were used to compute a series of comparisons between a day with no enforcement activity and a day with exceptionally high enforcement. During enforcement, the space mean speed decreased by approximately 5 mph throughout the 12.2-mi study segment. Within 30 min after the enforcement detail ended, however, space mean speeds increased, and there was no statistically significant residual impact on the space mean speed. Even at the absolute peak of enforcement, 75% of the probe vehicles exceeded the speed limit in all but one of the segments that had a posted speed limit of 45 mph. In addition, 25% of the probe vehicles exceeded the posted limit by more than 5 mph in all 45-mph segments during peak enforcement. The study is perhaps the largest ever conducted with respect to concurrent enforcement and extensive space mean measurement. The data represent an upper bound on the impact of enforcement activity on work zone speeds and should be of interest to public agencies as they consider compliance techniques.

This research is focused on: (1) drivers' compliance with the system, (2) delays and travel times on approaches to work zones, (3) optimal configuration of the system, and (4) warrants for the system's use. The simulation and field studies indicate a significant reduction in the number of merging maneuvers near work zones after the IMLS is applied. Also, the travel time on continuous lanes is reduced. The increased fairness of the system improves the perception of the traffic conditions among the majority of drivers. A slight reduction in the capacity of the merge point is the second finding of the field observations. This finding should be confirmed through long-term measurements of capacity during regular use of the IMLS units.The final report is divided into two parts. Part I presents the performed research, including the simulation model development and simulation experiments. Part II contains the system description, guidelines for its use, and rules for its setting. The system description includes presentation of the concept and the system components. The guidelines for the system use provide the traffic conditions where the system is expected to provide benefit. Finally, the manual gives a set of simple rules useful in setting all the system parameters to achieve the maximum reduction in the travel time in the continuous lane. DOT F 1700.7 (8-69) (TRB, 1975; TRB, 1996). Evidently, the assumption requires the qualification that speed is a function of density, but only at equilibrium. Because equilibrium can rarely be observed in practice, a satisfactory speeddensity relationship is hard to obtain, and it is often assumed or inferred theoretically. Higher-order Continuum ModelsThe higher order models take into account acceleration/deceleration and inertia characteristics of traffic mass by replacing the equilibrium speed-density relationship with a momentum equation (Whitham, 1974; Payne, 1979 where v is the viscosity coefficient. (Whitham, 1974; Payne, 1979 -v,) Cl dk,where U, is the equilibrium speed and T is the relaxation time. Typically, the value of U, is calculated using an empirical speed-density relationship developed from field data.The left-hand side of the above equation gives the acceleration/deceleration rate of the flow observed at location x. The acceleration/deceleration rate is a sum of several components: