Jalil Kianfar

Optimizing Freeway Traffic Sensor Locations by Clustering Global-Positioning-System-Derived Speed Patterns

This paper presents a new clustering-based methodology for sensor placements on freeways for estimating travel times. The proposed methodology is applicable to both freeways with no existing deployment and freeways with existing sensor deployments, where it identifies the critical sensors that need to be regularly maintained. The freeway sections are clustered based on speed data, with neighboring sections with identical speed profiles being grouped into a cluster. A new approach to estimate freeway travel times using the final clusters, which is called the optimal placement method, is also proposed. A family of k-means clustering algorithms and a hierarchical algorithm are then explored using real-world case studies of three freeway segments in Virginia. Speed and travel-time data are obtained using Global Positioning System (GPS)-equipped probe vehicles. The clustering results indicated that the hierarchical and k-means with a priori knowledge algorithms produced the best clusters. The tradeoff plots of travel-time measures (e.g., error) versus the number of freeway sensors were generated for two travel-time estimation methods: 1) optimal placement method and 2) midpoint placement method. The optimal placement method consistently produced better travel-time estimates than the midpoint placement method for all three case studies. The travel times were also estimated using three other methods found in the literature: 1) the zone of influence method; 2) the instantaneous method; and 3) the linear method. The results showed that the optimal placement method outperformed these methods in all three case studies.

Placement of Roadside Equipment in Connected Vehicle Environment for Travel Time Estimation

Vehicle-to-infrastructure (VTI) communication technology offers great potential to improve safety and mobility in transportation systems. One application of VTI communication to mobility is the measurement of travel time in an urban road network. The presence of traffic control at intersections in an urban network makes it challenging for traditional traffic-monitoring methods to provide accurate travel times. The use of several new technologies (e.g., Bluetooth) has been proposed to alleviate this problem. One of the features of the VTI communication technology is probe vehicle data collection, in which vehicles collect information on, for example, their location and speed. The speed information can be used for travel time estimation. This paper proposes a methodology for determining the optimal placement of roadside equipment (RSE) for travel time estimation in a VTI communication environment. A connected vehicle simulation test bed of Boise, Idaho, was developed in VISSIM traffic simulation software according to the SAE J2735 standard. A hybrid performance measure, the network coverage index, which combines travel time error and the number of links for which travel times are available, is proposed. A genetic algorithm-based solution method was implemented in conjunction with the simulation test bed to determine the optimal placement of different RSE deployments. The results indicate that the proposed methodology is capable of optimization of RSE locations in a VTI communication environment. Sensitivity analysis was also conducted by the use of various market penetration rates and travel time estimation intervals. The results indicate that higher penetration rates and larger estimation intervals produce better coverage index values.

Operational Analysis of a Freeway Variable Speed Limit System in St. Louis, Missouri

Variable speed limit (VSL) systems have been more widely implemented in Europe compared to the United States. Studies have demonstrated positive safety impacts of such systems; however, there are very few studies that used actual deployment data to investigate the operational benefits of VSL systems. This article presents the operational impacts of a VSL deployment on Interstate 270 in the state of Missouri in the United States. Techniques such as parametric curve fitting, nonparametric methods, and other statistical tests were used to identify the changes between before and after traffic conditions. The effect of VSL on traffic performance was investigated at eight heavily congested locations. The two-dimensional Kolmogorov–Smirnov test results indicated that flow–occupancy diagram changes were statistically significant at seven out of eight locations. The slopes of flow–occupancy plots for over critical occupancies were found to be steeper after VSL. Slight changes in critical occupancy were observed after VSL implementation. However, the changes were inconsistent across locations, with some witnessing an increase and others witnessing a decrease. The maximum flow prior to breakdown (the prebreakdown flow) decreased at four locations and increased at four locations after VSL. The maximum flow after breakdown (the postbreakdown flow) decreased at three locations and increased at five locations after VSL. The average daily duration of congestion decreased at five locations and increased at three locations after VSL. Findings of this study help to develop VSL control algorithms that are more efficient in improving VSL traffic operations benefits.

New Snapshot Generation Protocol for Travel Time Estimation in a Connected Vehicle Environment

Connected vehicle technology offers great potential to improve the safety and the mobility of a transportation system. Probe data collection is one feature of connected vehicle technology in which vehicles collect information such as their location and speed. Probe data could be used to support various traffic management and traveler information applications. This paper presents the novel R2 protocol, used to collect probe data in a connected vehicle environment. The core principle of R2 protocol is to collect only vehicle snapshots when a significant change occurs in vehicle speed. Data from a connected vehicle simulation test bed in Boise, Idaho, and a real-world test bed in Oakland County, Michigan, were used to evaluate the proposed protocol. An average speed method and a method with its basis in the reconstruction of vehicle time–speed plots were used to estimate link travel time. Linear regression, cubic spline, and piecewise cubic Hermite interpolation were applied to reconstruct time–speed plots. The proposed R2 protocol was compared with three existing protocols: fixed 2-s, fixed 4-s, and SAE J2735. The results from the simulation test bed indicated that the R2 protocol not only outperformed the three protocols in error measurement but also required fewer snapshots to achieve the lower-error value. The snapshots recorded by the R2 protocol were 30%, 26%, and 4% lower than those recorded by the other three J2735 protocols. The Michigan test bed case study showed that the R2 protocol produced fewer errors and needed 11% fewer snapshots than the SAE J2735 protocol.