Common use of Problem Statement Clause in Contracts

Problem Statement. Given the preliminary tests performed above, the problems of dual-loop data could be summarized as follows: • The total volume of vehicles detected and assigned to bins by the dual-loop system for a given time interval was consistently lower than the total volume of vehicles detected by the corresponding single loops for the same interval. In other words, the volume of classified vehicles was consistently lower than the volume of actual vehicles detected, reflecting vehicles that were dropped by the dual-loop system during the classification step. • The existing dual-loop detector systems had problems measuring vehicle lengths; hence the vehicle bin volume distribution for any time period might differ significantly from the actual distribution. Therefore, the detected trucking information did not reflect the true trucking activities during that period at that location. This study quantitatively evaluated the accuracy of WSDOT dual-loop data using video ground truth data and investigated the types and potential causes of dual-loop data inaccuracies. It also sought and recommended methods to improve the quality of dual- loop data. The objectives of the research are summarized below: • Identify dual-loop detectors in the FLOW system in good working condition to serve as representatives of WSDOT dual-loop detectors for analysis. • Quantitatively evaluate the accuracy of the sample dual-loop measurements of vehicle volumes and vehicle classifications. • Identify and interpret the potential causes of dual-loop data inaccuracies. • Recommend strategies for improving dual-loop data quality. Due to heavy weights and large turning radii, truck movements have characteristics very different from those of passenger cars and smaller vehicles. These differences make the collection of reliable and continuous vehicle classification data and truck volume data very important for reliable freeway performance monitoring. Of course, such data are critical to the specific monitoring of regional freight movements. To date, several different technologies have been used for vehicle detection and classification. These include single-loop detectors, dual-loop detectors, classification with vehicle acoustic signatures, video imaging systems, and laser and night vision systems. Each of these technologies has its own advantages and disadvantages. Dual-loop detectors are often used to collect vehicle classification data. As previously described, a dual-loop detector consists of two consecutive single loops that are only a few feet apart. An algorithm is applied to calculate vehicle speed and vehicle length on the basis of the information that these two single loops provide. The inductive loops are relatively cheap to install but at the cost of some inconvenience, as traffic has to be stopped. The loops can be damaged or broken and, thus, become unreliable. Vehicle classification can also be achieved using single-loop detectors. Xxxx and Xxxxx (Xxxx and Xxxxx 2000a) analyzed vehicle-length distributions and developed a nearest-neighbor algorithm to classify vehicles into a short-vehicle (SV) category and a large-truck (LT) category by using only single-loop measurements. This research development is especially useful in practice because most highway systems are, to date, equipped only with single-loop detectors. Comparison of the estimated large-truck volumes with dual-loop observed large-truck volumes showed that this algorithm worked well, especially when traffic volume was low. However, the method tended to underestimate large truck volumes under heavy traffic conditions. This was due to the fact that when traffic volume is heavy, speed is very unstable, and the uniform speed assumption, on which the algorithm is based, is violated.

Problem Statement. Given the preliminary tests performed above, the problems of dual-dual- loop data could be summarized as follows: • The total volume of vehicles detected and assigned to bins by the dual-dual- loop system for a given time interval was consistently lower than the total volume of vehicles detected by the corresponding single loops for the same interval. In other wordsI.e., the volume of classified vehicles was consistently lower than the volume of actual vehicles detected, reflecting vehicles that were dropped by the dual-dual- loop system during the classification step. • The existing dual-dual- loop detector systems had problems measuring vehicle lengths; hence the vehicle bin volume distribution for any time period might differ significantly from the actual distribution. Therefore, the detected trucking information did not reflect the true trucking activities during that period at that location. This study quantitatively evaluated the accuracy of WSDOT dual-dual- loop data using video ground truth data and investigated the types and potential causes of dual-dual- loop data inaccuracies. It also sought and recommended methods to improve the quality of dual- loop data. The objectives of the research are summarized below: • Identify dual-dual- loop detectors in the FLOW system in good working condition to serve as representatives of WSDOT dual-dual- loop detectors for analysis. • Quantitatively evaluate the accuracy of the sample dual-dual- loop measurements of vehicle volumes and vehicle classifications. • Identify and interpret the potential causes of dual-dual- loop data inaccuracies. • Recommend strategies for improving dual-dual- loop data quality. Due to heavy weights and large turning radii, truck movements have characteristics very different from those of passenger cars and smaller vehicles. These differences make the collection of reliable and continuous vehicle classification data and truck volume data very important for reliable freeway performance monitoring. Of course, such data are critical to the specific monitoring of regional freight movements. To date, several different technologies have been used for vehicle detection and classification. These include single-single- loop detectors, dual-dual- loop detectors, classification with vehicle acoustic signatures, video imaging systems, and laser and night vision systems. Each of these technologies has its own advantages and disadvantages. Dual-Dual- loop detectors are often used to collect vehicle classification data. As previously described, a dual-dual- loop detector consists of two consecutive single loops that are only a few feet apart. An algorithm is applied to calculate vehicle speed and vehicle length on the basis of the information that these two single loops provide. The inductive loops are relatively cheap to install but at the cost of some inconvenience, as traffic has to be stopped. The loops can be damaged or broken and, thus, become unreliable. Vehicle classification can also be achieved using single-single- loop detectors. Xxxx and Xxxxx (Xxxx and Xxxxx 2000a) analyzed vehicle-vehicle- length distributions and developed a nearest-nearest- neighbor algorithm to classify vehicles into a short-short- vehicle (SV) category and a large-truck (LT) category by using only single-single- loop measurements. This research development is especially useful in practice because most highway systems are, to date, equipped only with single-single- loop detectors. Comparison of the estimated large-truck volumes with dual-dual- loop observed large-truck volumes showed that this algorithm worked well, especially when traffic volume was low. However, the method tended to underestimate large truck volumes under heavy traffic conditions. This was due to the fact that when traffic volume is heavy, speed is very unstable, and the uniform speed assumption, on which the algorithm is based, is violated.