Traffic Performance Measures Research Articles

Anonymous Vehicle Reidentification Using Heterogeneous Detection Systems

An innovative feature of this paper is the demonstration of the feasibility of real-time vehicle reidentification algorithm development at a signalized intersection, where different traffic detection technologies were employed at upstream and downstream locations. Previous research by the authors on vehicle reidentification has utilized the same traffic sensors (e.g., conventional square inductive loops) and detectors (e.g., high-speed scanning detector cards) at both locations. In this paper, an opportunity arose for the first time to collect a downstream data set from a temporary installation of a prototype innovative inductive loop sensor known as a ldquobladerdquo sensor in conjunction with conventional inductive loops upstream. At both locations, advanced high-speed scanning detector cards were used. Although the number of vehicles for which data could be collected was small, encouraging results were obtained for vehicle reidentification performance in this system of mixed traffic detection technologies. In future large-scale applications of vehicle reidentification approaches for real-time traffic performance measurement, management, and control, it would be most beneficial and practical if heterogeneous and homogeneous detection systems could be supported. This initial paper yielded many useful insights about this important issue and demonstrated on a small scale the feasibility of vehicle reidentification in a system with heterogeneous detection technologies.

Is revamping your web site worthwhile?

PurposeThe purpose of this paper is to study how web site quality affects traffic performance.Design/methodology/approachAn experimental design is employed to study how web quality affects traffic performance. A revamping of the experimental web site was used as the treatment, targeting visitors' perceived quality of the web site. Four traffic performance measures, page: views, visitor count, daily registrations, and average duration are tracked, and t‐tests are performed on pre‐treatment and post‐treatment data.FindingsThe analysis shows very positive responses among members; visitor count, page views and average duration increased for opt‐in and opt‐out members. For visitor count, even non‐members showed increases. However, daily registration, which measures how many non‐members become members each day, did not change. Non‐members visited more, but neither viewed more pages, nor stayed longer. Average duration is identified as the key factor for discerning visitor groups.Research limitations/implicationsThe experimental web site belongs to one web site category. The generalization is subject to reasoning by practitioners.Practical implicationsIt was found that: to increase membership, alternative schemes must be employed, perhaps along the lines of a non‐technical approach; to acquire more members, do not focus on converting known non‐members. Those with the same demographic profile as existing members should be targeted; and the question must be asked whether the fact that opt‐in members are stickier than opt‐out members is a trait or a consequence of opt‐in members receiving e‐mails periodically, while opt‐out members chose not to receive e‐mails.Originality/valueWith few existing traffic experiments in the literature, this study is unique, as are its implications.

IMAGIS, a PACS software solution for the National Health System in Cuba

Elastic Internet applications have the greatest share in the traffic transported over the Internet today. Web access and e-mail are the most popular applications among most people using the Internet. Although these applications can well adapt to a wide range of bandwidths, their performance is an issue to people using services interactively.In this paper, we use sophisticated application-level evaluations to extract traffic characteristics and performance measures for HTTP, SMTP and POP3 from traffic traces. The process of loading Web pages by real browsers using persistent and parallel connections is studied in detail, revealing statistics about the elements in Web pages as well as the number of parallel connections and the accumulated waiting time during Web page loads.The main results are: (1) High variance distributions are not only found as expected in file sizes, but also in the number of items in a Web page, the number of e-mails transmitted in one connection, the duration of SMTP command exchanges or even the users' viewing times for Web pages. (2) Web browsers utilize more parallel connections than usually expected to simultaneously load items. (3) A large portion of the delay in Web page retrievals or e-mail transfers is due to serial waiting and can therefore not be significantly reduced by increasing bandwidths. Therefore, low latency is as much an issue in access systems as high bandwidth.

Optical storage area networks

Storage area networks (SANs) are a promising technology to efficiently manage the ever-increasing amount of business data. Extending SANs over large distances becomes essential to facilitate data protection and sharing storage resources over large geographic distances. A WDM metropolitan ring network is examined as a suitable extension for SANs where it is shown that sectioning the ring can help deal with traffic asymmetry and hot node (SAN node on ring) scenarios. Several network architectures are studied: One of the architectures accommodates a single SAN and its mirror connected through a sectioning link. Another architecture accommodates two pairs of SANs and their mirrors with sectioning links connecting each pair. Issues investigated include impact of the number of co-existing IP (non SAN) nodes; traffic models: Poisson and self-similar; slotted regime: fixed-size (FS), variable-size (VS) and supersize (SS) slot schemes; MAC protocol design; handling traffic asymmetry and performance measures.

Analysis of Freeway Traffic Incident Conditions by Using Second-Order Spatiotemporal Traffic Performance Measures

The characteristics of preincident, postincident, and nonincident traffic conditions on freeways are investigated. The characteristics are defined by second-order statistical measures derived from spatiotemporal speed contour maps. Four performance measures are used to quantify properties such as smoothness, homogeneity, and randomness in traffic conditions in a manner similar to texture characterization of digital images. With real-world incident and traffic data sets, statistical analysis was conducted to seek distinctive characteristics of three groups of traffic operating conditions: preincident, postincident, and nonincident. The study results showed that the spatiotemporal characteristics of each of the three groups were not discernible. Although the distributions of performance measures within each group are statistically different, no consistent pattern was detected to imply that certain characteristics could increase the likelihood of incidents or identify precursory conditions to incidents.

Analysis of Freeway Traffic Incident Conditions by Using Second-Order Spatiotemporal Traffic Performance Measures

The characteristics of preincident, postincident, and nonincident traffic conditions on freeways are investigated. The characteristics are defined by second-order statistical measures derived from spatiotemporal speed contour maps. Four performance measures are used to quantify properties such as smoothness, homogeneity, and randomness in traffic conditions in a manner similar to texture characterization of digital images. With real-world incident and traffic data sets, statistical analysis was conducted to seek distinctive characteristics of three groups of traffic operating conditions: preincident, postincident, and nonincident. The study results showed that the spatiotemporal characteristics of each of the three groups were not discernible. Although the distributions of performance measures within each group are statistically different, no consistent pattern was detected to imply that certain characteristics could increase the likelihood of incidents or identify precursory conditions to incidents.

Anonymous Vehicle Tracking for Real-Time Traffic Surveillance and Performance on Signalized Arterials

One of the fundamental requirements for facilitating implementation of any advanced transportation management and information system (ATMIS) is the development of a real-time traffic surveillance system able to produce reliable and accurate traffic performance measures. This study presents a new framework for anonymous vehicle tracking capable of tracing individual vehicles by the vehicle features. The core part of the proposed vehicle tracking method is a vehicle reidentification algorithm for signalized intersections based on inductive vehicle signatures. The new vehicle reidentification system consists of two major components: search space reduction and probabilistic pattern recognition. Not only real-time intersection performance but also intersection origin–destination information can be obtained as the algorithm’s basic output. A systematic simulation investigation was conducted of the performance and feasibility of anonymous vehicle tracking on signalized arterials using the Paramics simulation model. Extensive research experience with vehicle reidentification techniques on single roadway segments was the basis for investigating the performance that could be obtained from tracking individual vehicles across multiple detector stations. The findings of this study serve as a logical and necessary precursor to possible field implementation of vehicle reidentification techniques. The proposed anonymous vehicle tracking methodology with existing traffic surveillance infrastructure would be an invaluable tool for operating agencies in support of ATMIS strategies for congestion monitoring, adaptive traffic control, system evaluation, and provision of real-time traveler information.

Deriving Traffic-Performance Measures and Levels of Service from Second-Order Statistical Features of Spatiotemporal Traffic Contour Maps

Little information has been successfully extracted from the wealth of data collected by intelligent transportation systems. Such information is needed for the efficiency of operations and management functions of traffic-management centers. A new set of second-order statistical measures derived from texture characterization techniques in the field of digital image analysis is presented. The main objective is to improve the data-analysis tools used in performance-monitoring systems and assessment of level of service. The new measures can extract properties such as smoothness, homogeneity, regularity, and randomness in traffic operations directly from constructed spatiotemporal traffic contour maps. To avoid information redundancy, a correlation matrix was examined for nearly 14,000 15-min speed contour maps generated for a 3.4-mi freeway section over a period of 5 weekdays. The result was a set of three second-order measures: angular second moment, contrast, and entropy. Each measure was analyzed to examine its sensitivity to various traffic conditions, expressed by the overall speed mean of each contour map. The study also presented a tentative approach, similar to the conventional one used in the Highway Capacity Manual, to evaluate the level of service for each contour map. The new set of level-of-service criteria can be applied in real time by using a stand-alone module that was developed in the study. The module can be readily implemented online and allows traffic-management center operators to tune a large set of related parameters.

Observed performance of elastic Internet applications

Elastic Internet applications have the greatest share in the traffic transported over the Internet today. Web access and e-mail are the most popular applications among most people using the Internet. Although these applications can well adapt to a wide range of bandwidths, their performance is an issue to people using services interactively. In this paper, we use sophisticated application-level evaluations to extract traffic characteristics and performance measures for HTTP, SMTP and POP3 from traffic traces. The process of loading Web pages by real browsers using persistent and parallel connections is studied in detail, revealing statistics about the elements in Web pages as well as the number of parallel connections and the accumulated waiting time during Web page loads. The main results are: (1) High variance distributions are not only found as expected in file sizes, but also in the number of items in a Web page, the number of e-mails transmitted in one connection, the duration of SMTP command exchanges or even the users' viewing times for Web pages. (2) Web browsers utilize more parallel connections than usually expected to simultaneously load items. (3) A large portion of the delay in Web page retrievals or e-mail transfers is due to serial waiting and can therefore not be significantly reduced by increasing bandwidths. Therefore, low latency is as much an issue in access systems as high bandwidth.

Capacity at Unsignalized Intersections Derived by Conflict Technique

A new simplified theoretical approach for the determination of capacities at unsignalized intersections has been developed on the basis of the method of additive conflict streams. This method is much easier to handle than the method of gap acceptance. It avoids many of the theoretical complications inherent in the method of gap acceptance, which, under certain circumstances, seems to be unrealistic. The new method has been developed for potential intersection configurations when one street has priority over the other. A calibration of the model parameters is given for German conditions. The new procedure can deal with shared lanes, short lanes, flared entries, and cases of so-called limited priority. For the estimation of traffic performance measures such as average delay and queue lengths, the classical methods can be applied.

A Rude Awakening: Internet Shakeout in 2000

This study explores various value-drivers of business-to-consumer ("B2C") Internet companies' share prices both before and after the market correction in the spring of 2000. Although many market observers had predicted that the shakeout would eventually occur (e.g., Perkins and Perkins 1999), the ultimate and previously unanswered challenge lay identifying which stocks would fall and which ones would survive the shakeout. We develop an empirical valuation model and provide evidence that the Internet stocks that this model suggests were relatively over-valued prior to the Internet stock market correction experienced relatively larger drops in their price-to-sales ratios when the shakeout occurred. This result is robust to the inclusion of competing explanatory variables suggested by the economics literature related to industry rationalizations. We examine the ability of a valuation model comprised of both financial (accounting) variables and nonfinancial web traffic metrics to explain Internet companies' market values during each of 1999 and 2000. Our findings suggest that the reach and stickiness web traffic performance measures are value-relevant to the share prices of Internet companies in each of 1999 and 2000. Our findings of significance for the year 2000 contradict the recent claims of some analysts that web traffic measures are no longer important. We also explore the valuation role of our proxy for B2C companies' current rate of "cash burn" and find that this proxy is a significant value-driver in each of 1999 and 2000, but with differential valuation implications for each period. Our results suggest that the market was favorably disposed towards Internet companies' aggressive cash expenditures in 1999, but appeared to adopt a more critical view of Internet companies' cash burn rates in 2000. Our results further suggest that investors adopted a more skeptical attitude towards expenditures on intangible investments as the Internet sector began to mature. We find that investors appear to implicitly capitalize product development (R&D) and advertising expenses (customer acquisition costs) during the earlier period when the market was more optimistic about the prospects of B2C companies. However, only product development costs are implicitly capitalized into value, on average, subsequent to the shakeout in the spring of 2000. Finally, we provide statistical evidence to support the conjecture that different parameter vectors characterize the estimated market valuation models for each of 1999 and 2000. Overall, our study provides a preliminary view of the shakeout and maturation of one of the most important New Economy industries to emerge to date-the Internet.

The Use of Predictive Displays for Aiding Controller Situation Awareness

This study provides an empirical evaluation of the effects of an enhanced predictive display concept for air traffic control. The predictive display informed the controller of the targeted altitude or heading of aircraft in a transitionary state The information provided simulated that which could be produced by a datalink from the aircraft's flight management system in a future air traffic system operation. The pilots in the simulation operated under self-separation rules and did not request clearances for altered flight plans. Use of the predictive display resulted in improvements in some air traffic control performance and situation awareness measures. Based on the results of the study, we recommend making improvements to the predictive display to better integrate it with the controller's radar display.

Traffic engineering of the radio interface for cellular mobile networks

The paper reviews traffic-related issues in the operation of the radio interface in cellular mobile networks operating with FDMA, TDMA, or CDMA technology. The emphasis is on issues which affect the traffic performance of the network, rather than on transmission technology, or transmission quality measures such as bit error rate. The author reviews the important traffic performance measures for cellular systems and the issues of channel assignment, handoff, access control, traffic modeling, and traffic management. Traffic modeling for all types of networks is considered in some depth, both by drawing analogies with fixed-wire networks, and by considering cellular-specific modeling. From the modeling results, he concludes that reuse-based networks such as FDMA and TDMA possess quite different traffic characteristics from CDMA networks. >

Traffic sensitive algorithms and performance measures for the generation of self-organizing radio network schedules

The generation of transmission schedules for self-organizing radio networks by traffic-sensitive algorithms is described. A centralized 'Traffic' algorithm that can be used as a performance benchmark is presented. Also described is a distributed 'degree' algorithm that is a traffic-sensitized version of an algorithm developed by A. Ephremides and T. Truong (1990). Two performance measures for comparing schedules and simulation results are also presented. >