Traffic data

Historic traffic data and traffic counts are also very important because they are necessary to determine past traffic and predict future pavement condition and remaining pavement life. 

Historic traffic data are relatively easy to be found in most countries. If not available, past traffic can be estimated from traffic counts executed during surveying period, the average annual increase of commercial vehicles from date of construction or last intervention and number of years elapsed. 

---

This paper presents applying of neural networks for intrusion detection through an examination of network traffic data. It has been shown that denial of service and other network-based attacks are presented in the network traffic data. Therefore using neural networks permits to extract nonlinear relationships between variables from network traffic and to design real-time intrusion detection systems. 

Let s examine the block-diagram of the intrusion detection system (Figure 1). It consists of several stages. At the beginning the system reads traffic data and sends it to the preprocessing module. The task of preprocessing module is to collect necessary data for neural networks from network traffic. 

Vector 2008 Shapefile From National Road Traffic Data... 

The above eases collection of traffic data since only traffic composition and frequency need to be recorded. 

Mintsis G., C. Taxiltaris, S. Basbas, A. Filaktakis, K. Koutsoukos, S. Guy, and E. Viskos. 2007. Temporal evolution of HGV traffic data along Egnatia Odos motorway. Proceedings of the 4th International Conference on Bituminous Mixtures and Pavements, pp. 591-601.Thessaloniki, Greece Aristotle University of Greece. 

The traffic composition may be estimated from past surveys or representative area traffic data or may be determined from new counts. The conversion to ESAL may be carried out by using equivalency factors or truck factors provided. The equivalency factors provided by the methodology are those given in Table 12.3. 

In case there are no analytical traffic data per type of vehicle, a combined wear factor for each of the two distinguished categories (OGVl + PSV) and OGV2 can be used. These factors are as shown at the bottom of Table 13.4. 

It is emphasized that the traffic load model may be significantly dependent on the road category and thus should be always developed considering available traffic data. 

One proposal focuses on a system known as Applications for the Environment Real-Time Information Synthesis (AERIS) that would help people make decisions about the best and most efficient mode of transportation to use for a particular trip. Under the proposal, AERIS would provide information comparing current travel times from one point to another depending on the mode of transportation. The information would be drawn from traffic data on highways and roads, public transit schedules, up-to-the-minute information on delays, and weather data. The system s interface might list a traveler s options by transportation mode, ranking the options by travel time, cost, or carbon footprint. These options could be viewed on a mobile device such as a cell phone. 

Some of the sources in the United States for collecting vessel traffic data are as follows ... 

Urban Mobility Information from the Texas Transportation Institute (TTI) http //mobility. tamu.edu/ (accessed September 15,2010) provides information and resources about traffic monitoring and congestion, with emphasis on the analysis of traffic data. 

In order to examine the applicability of the complexity factors introduced in Section 3, we decided to use a neural network based approach similar to the one seen in Gianazza Guittet (2006). To provide input data to our networks, air traffic data of the Hungarian airspace was obtained for two days in the past (29th July 2011 and 30th July 2012) and a few more hours from 25th October 2012 when military TRAs were in use. We took snapshots of the radar data every 30 minutes for the two full days and every 10 minutes for the time when TRAs were open. After taking these samples, we had radar data for 107 different air traffic situations which included a wide variety of traffic levels. 

The historic overview clearly shows that a wide range of perspectives has been applied in risk assessments for maritime transportation. Most of the work is rooted in the idea that a true, mind-independent risk exists in line with realist perspectives on the continuum of Table 1. Using different modeling approaches, the majority of the methods aim to estimate this true risk. While the use of expert judgment has gained steady support, many of the applications rely heavily on accident and traffic data. Even when judgment is applied, it is considered to be truth-oriented. From the overview, it is also found that constructivist views, where the assessment is seen rather as a reflection of an assessor s interpretation of the system risk, are rare. The uncertainty view, where an assessor expresses his uncertainty about the occurrence of events and consequences, is not found in the application area. 

ABSTRACT In this paper we introduce a method for the assessment of the complexity of ship-ship encounters, linked with the risk level of such encounters. The method combines traffic data obtained from the Automatic Identification System and experts judgment. The collected data describes maritime traffic over the Northern Baltic Sea, including the Gulf of Finland, which is potentially one of the most heavily trafficked sea areas in the world. 

The method presented in this paper applies experts judgment and micro-level ship traffic data... 

In this chapter, the methodological framework is provided for the developed model, which detects and ranks the complexity of ship-ship encounters. First, the available micro-level ship traffic data, as obtained from the Automatic Identification System (AIS), is outlined. Subsequently, the process and outcome of an expert elicitation related to the development of a quantitative Risk Indicator (I). The latter is a measure of risk in the sea area, based on detection of vessels encounters and their complexity. Then, the mathematical formulation of the indicator is presented, both in regards to the model structure and model parameters. Finally, the obtained results need to be clustered to arrive at qualitative Risk Indicator. Thus, the purpose and rationale of the applied clustering technique is outlined. 

Micro-level ship traffic data Automatic... 

Finally, we carried out the assessment of ship collision risk for the Northern Baltic Sea, with the use of the Risk Indicator introduced here. For this purpose we utilized the traffic data for May and July 2011 with the primary aim to assess the historic collision risk level in various locations of the Northern Baltic Sea in order to draw a spatial risk map for the analyzed sea area. 

The method applies experts judgment together with micro-level ship traffic data obtained from the Automatic Identification System corresponding to the Northern Baltic Sea and summer season (Many, July 2011). 

The traffic composition is based on traffic data provided by the Road and Motorway Directorate of the Czech Republic. 

In the former analysis of road safety trends in Poland - due to the unavailability of traffic data, and to the occurrence of two economic breakdowns in 2001 and 2007 onwards - researchers tomed to economic factors to explain the simultaneous decrease that occurred in the trend of the number of fatalities in Poland - as in other countries in Europe. Variables such as the GDP and unemployment rate have been used for modeling changes in the number of fatalities in the short- or medium term... 

In total, 345 injury accidents, described in the French accident database BAAC (Bulletins d Analyse des Accidents Corporels), occurred during the period of 2003-2010 on this 23 km long part of the ALLEGRO Network. Traffic data from 15 stations were available only during the period of 2009-2010, covering lane flow, speed and occupancy every 6 minute. However, it was not possible to link an accident to the traffic conditions before, because very few traffic stations were available for this research. 

On the Marius network where detailed traffic data exist, a density can be assigned by accident. This was not the case for the ALLEGRO Network, where traffic data are more rare however, that available traffic data are assumed to be... 

As traffic data covers two years and accident data covers eight years, risks per vehicle kilometer are computed with the assumption that the traffic during the period of 2003-2008 was the same as during the period of 2009-2010. 

---