Now showing 1 - 9 of 9
  • Publication
    Load effect of multi-lane traffic simulations on long-span bridges
    The traffic loading of long-span bridges is governed by congestion. Real-world observations show that congestion can take up different forms. Nevertheless, most previous studies on bridge traffic loading considered only a queue of standstill vehicles. In this paper, a micro-simulation tool is used for generating congested traffic on a two-lane same-way roadway. The total load is computed for a sample long-span bridge. Different congestion patterns are found and they are studied in relation to their effect on loading. It is found that very slow-moving traffic returns the highest loading events, rather than full stop conditions. The topic is especially relevant to existing bridges, where small differences in the loading may play an important role in the safety assessment and subsequent maintenance plans.
  • Publication
    Load effect of single-lane traffic simulations on long-span bridges
    It is well acknowledged that long-span road bridges (about 50 m long and more) are governed by congestion traffic rather than free-flow conditions. A conventional model for the design of new long-span bridges is to place over the bridge a load model representing a platoon of heavy vehicle with the gaps between them reduced to a minimum. This assumption is too conservative for existing bridges, given the large disruption costs faced by their closure for rehabilitation. In order to model the close gaps between vehicles, characteristic of congested traffic, microsimulation is needed to accurately capture drivers’ behaviour. In this work, a microsimulation model is studied and found to replicate many different known forms of congestion. As a first approach to the topic, single-lane simulations of identical vehicles have been carried out in order to obtain load effect on a sample bridge. This load effect is studied with reference to the form of traffic causing the load effect. It is found that the most extreme load effect may not be caused by purely congested traffic but also by non-stationary congested conditions
  • Publication
    Micro-simulation modelling of congestion due to lane closures
    (Irish Transport Research Network, 2012-08-30) ; ;
    Incident clearance and road work often require the closure of one or more of the available lanes on a highway. A lane-closure causes a significant capacity reduction, which often leads to heavy congestion. Simulation of congestion events due to lane - closures is relevant both for traffic and infrastructure management. This is especially valid when trucks are involved and they concentrate on bridges or in tunnels, thus generating critical situation s for loading and safety. A better understanding of the effects of lane closures requires a realistic simulation of the merging manoeuvre of vehicles occurring in the proximity of the lane closure. Micro-simulation allows for the motion of individual vehic les and it is therefore a suitable tool for studying traffic merging. In this paper, a micro-simulation tool made up of a car-following model and a lane-changing model is used for simulating a lane closure on a two-lane one direction stretch of road. The effects on traffic are studied, in terms of average speed, lane change rates, and truck distribution. It is found that the lane-changing model requires an appropriate parameter calibration when applied to lane-closures. These parameters are quite different from the ones reported in literature. An alternative means of causing congestion is also tested and it is found that it can replicate the overall congestion features upstream the closure. However, there are some differences about details of the traffic features
  • Publication
    Micro-simulation modelling of traffic loading on long-span bridges
    Traffic loading on long-span bridges is governed by congestion. An accurate modelling of the traffic loading to which a bridge is actually exposed allows to plan maintenance operations more effectively. However, most previous research uses simplified traffic models, typically consisting of a queue of vehicles at minimum bumper-to-bumper distances, in fact neglecting lighter and more frequently observed forms of congestion. Moreover, the available load models assume conservative frequencies of occurrence of congestion. In this research, micro-simulation is used to reproduce several observed congestion patterns. Different combinations of traffic streams and congestion strengths are simulated, thus allowing the analysis of several traffic features and their effects on long-span bridge loading. It is found that the queue of vehicles at a standstill, widely used in previous research, is not always the most critical congested state for long-span bridge loading. In fact, also slow-moving traffic can be critical. It is also found that the load does not significantly depend on the inflow, as long as congestion is triggered, implying that critical loading events may also happen out of rush hours. A new methodology is proposed to compute the characteristic load based on site-specific traffic data, such as incident patterns and hourly flow distributions. This methodology is applied to traffic data available from the literature. It is found that flows with high truck percentage, typically occurring at night-time in combination with low traffic volumes, largely contribute to the characteristic load, even though congestion in such conditions is quite rare. Finally, some issues related to traffic data collection for bridge loading are dealt with and a promising application for bridge loading control is proposed.
  • Publication
    The effect of controlling heavy vehicle gaps on long-span bridge loading
    (Transportation Research Board, 2013-01) ; ;
    Long-span road bridges are governed by congested traffic rather than free-flowing conditions. During congestions, heavy vehicles can get quite close to each other, thus giving potential critical loading events for the bridge. In this paper, the effects of a system capable of warning truck drivers when the gap falls below a certain threshold are investigated. The effects are studied both in terms of increase in traffic disruption and reduction in loading. The minimum distance between trucks should be ideally adjusted in relation to the site-specific traffic features and to the load the bridge is able to carry in safety. Doing so, it is possible to allow for future increase in truck weight regulations and/or heavy traffic volumes, by adjusting the control gap value. Importantly, the system does not presume any restriction to the truck weight. By contrast, the system is meant to be an alternative way of limiting the load on long-span bridges by keeping the trucks apart, rather than by limiting the truck weight. The introduction of such a gap control system is studied by means of micro-simulation. The car-following model used here has been shown able to replicate many observed congestion patterns. Results show that the introduction of the gap control system does not significantly disrupt the traffic further. On the other hand, having only 10% of equipped trucks beneficially reduces the total traffic loading by about 10%. When most trucks are equipped, nearly 50% reduction in the total load can be attained.
  • Publication
    A comparative study of a bridge traffic load effect using micro-simulation and Eurocode load models
    Micro-simulation, a process by which individual vehicles and driver behaviour are simulated, is used in this study to assess the effects of congested traffic load on bridges. The Eurocode for traffic loading on long-span bridges was derived from simulations of convoys of trucks at small bumper-to-bumper distances. Real congested traffic includes a mix of cars and trucks with a variety of congestion patterns, which reduce the load effects. Micro-simulation is used in this study to generate more realistic traffic loading scenarios, based on real traffic data from a Polish site. The results are then extrapolated to determine characteristic maximum effects for two return periods. The Eurocode is shown to be conservative, and micro-simulation based assessment is therefore valuable for problem long-span bridges.
  • Publication
    Micro-simulation of single-lane traffic to identify critical loading conditions for long-span bridges
    The traffic loading of long-span bridges is governed by congestion. Real-world observations show that congestion can take several different forms. Nevertheless, most previous studies on bridge traffic loading consider only queues of vehicles at minimum bumper-to-bumper distances. In fact, such full-stop queues are rare events, while in most cases congestion waves propagate through the traffic stream, so that on a bridge there are periodically times of closely-spaced vehicle concentrations and times of flowing traffic, where vehicles are more distant. In this paper, an acknowledged traffic micro-simulation model is used for generating congested traffic on a single-lane roadway encompassing two bridges (200 and 1000 m long). Two truck percentages are considered (20% and 50%) and different congestion patterns are analysed in relation to their traffic features and effects on bridge loading. It is found that for the case of 200 m span and 20% trucks slow-moving traffic results in greater loading than full-stop conditions. Finally, the frequency of occurrence of different forms of congestion is taken into account based on recent available data, rather than being assumed as in most previous research. It is found that considering only the widely-used full-stop conditions leads to an over-estimation of the characteristic total load by about 10% for the cases of 200 m span with 50% trucks, and 1000 m with 20% trucks; for the case of 1000 m span with 50% trucks, the over-estimation drops to nearly 5%. However, for the case of 200 m span with 20% trucks, considering only the full-stop conditions leads to a slight under-estimation of the total load.
    Scopus© Citations 38  713
  • Publication
    Effect of Single-Lane Congestions on Long-Span Bridge Traffic Loading
    It is well known that traffic loading of long-span bridges is governed by congestion. In spite of the fact that field observations in the past decades have shown that congestion can take up different forms, most previous studies on bridge traffic loading consider only a queue of standstill vehicles. In this paper, a micro-simulation tool is used for generating congested traffic on a single-lane roadway. The underlying micro-simulation model has been found capable of successfully replicating observed congestion patterns on motorways by simulating single-lane traffic with identical vehicles. Here trucks are introduced into the model, in an investigation of the total load for a 200m span bridge. Different congestion patterns are found and studied in relation to their effect on loading. It is found that the bumper-to-bumper queue is not necessarily the most critical situation for the sample long-span bridge, since it does not allow the flowing of vehicles and therefore decreases the probability of observing critical loading events. Slow-moving traffic, corresponding to heavy congestion, can be more critical, depending on the truck proportion.
  • Publication
    Determination of Minimum Gap in Congested Traffic
    Accurate evaluation of site-specific loading can lead to cost and material savings in rehabilitation and replacement of bridges. Currently, bridge traffic load assessment is carried out using long run traffic simulations based on weigh-in-motion (WIM) data obtained at the site. Congestion is the governing load condition for long-span bridges. To correctly model congestion, a minimum gap between vehicles is usually assumed. Where the gap is overestimated, the calculated characteristic load is smaller than the actual characteristic load leading to an unsafe assessment. If the gap is underestimated, the safety assessment is too conservative, which is both costly and wasteful of finite resources. This paper outlines the development of an optical method to measure parameters required to model driver behaviour in congestion. Images are obtained using a camera with a wide angle, aspherical lens. Edge detection and Hough transforms are used to location wheels and bumpers. The resulting data can increase the accuracy of traffic microsimulation and hence, the assessment of long span bridge traffic loading.