A Paper Submitted for Presentation and Publication at the 96th Annual Meeting of the Transportation Research Board, Washington, D.C.

Transcription

1 Estimating the Capacity Impacts of Urban Street Incidents Aidin Massahi. (Corresponding Author) Research Assistant & PhD candidate Department of Civil and Environmental Engineering Florida International University 0 W. Flagler Street, EC Miami, FL Phone: amass0@fiu.edu Mohammed Hadi, Ph.D., P.E. Professor Department of Civil and Environmental Engineering Florida International University hadim@fiu.edu Maria Adriana Cutillo MSc Student Department of Civil and Environmental Engineering Florida International University mcuti00@fiu.edu Yan Xiao, Ph.D., PE Research Assistant Professor Department of Civil and Environmental Engineering Florida International University yxiao00@fiu.edu Word Count = words + figures + tables = words A Paper Submitted for Presentation and Publication at the th Annual Meeting of the Transportation Research Board, Washington, D.C. August 0, 0

2 0 0 ABSTRACT Incident impacts on capacity is the most critical parameter to estimate the influence of incidents on network performance. The Highway Capacity Manual (HCM 00) provides estimates of the drops in capacity due to incidents, as a function of the number of the blocked lanes and the total number of lanes of the freeway section. However, there is limited information regarding the impacts of incidents on the capacity of urban streets. This study investigates the capacity impacts of the interaction between the drop in capacity below demands at midblock urban street segment location and upstream and downstream signalized intersection operations. A model was developed to estimate the drop in capacity at the incident location as a function of the number of blocked lanes, distance from downstream intersection, and green time to cycle length (g/c) ratio of the downstream signal. A second model was developed to estimate the reduction in the upstream intersection capacity due to the drop in capacity at midblock incident location as estimated by the first model. The second model estimates the drop in capacity of the upstream links feeding the incident locations as a function of incident duration time, volume to capacity (v/c) ratio at the incident location, and distance from an upstream signalized intersection. The models were developed based on data generated utilizing a microscopic simulation model that was calibrated by comparison to parameters suggested in the HCM 00 for incident and no-incident conditions and by comparison with field measurements. Key Words: Highway Capacity, Urban Street, Incident Impacts, Incident Management

3 INTRODUCTION Active Transportation Management (ATM) is an important component of Transportation system management and operations (TSM&O). It provides significant benefits in terms of travel time, travel time reliability, emission, fuel consumption, safety, and other performance measures of transportation systems. Although, the impact of ATM strategies on mobility measures have been widely investigated for freeway facilities, there is a lack of studies, that address ATM strategies for urban streets (-). As Transportation agencies start to move their focus to these streets. Agencies have encountered challenges associated with traffic signals configuration and timing. Incident management is a vital part of ATM strategies. Estimating the impact of incidents and incident management strategies allows traffic management agencies to determine the need for various incident management strategies and technologies to justify the decisions to invest in their programs. These strategies can include incident detection, response, and clearance strategies. Special signal control plans implemented during incident conditions can increase the roadway capacity using arterial incidents and diversion due to freeway incidents. In addition, modifying signal timing can be combined with traveler information that guides motorists to alternative routes ().Incident management strategies was found to reduce the incident duration by to 0 percent (-), delay by to 0 percent (), and secondary incidents by to percent (,). These strategies also reduced emission. For instance, the Atlanta s NAVIGATOR system, reduced. kg of hydrocarbons (HC),. kg of carbon monoxide (CO) and.0 kg of nitrous oxides NOx per incident (). Birst and Smadi () utilized simulation to show that adaptive signal control combined with DMS (Dynamic Message Sign) can decrease travel times by percent and increase speeds by percent. Glassco, et al., (0) found based on simulation that adaptive signal control of alternative routes around an incident reduced delay by 0 to 0 percent for the affected paths. With an emphasis on implementing incident management strategies for signalized urban streets in recent years, there has been an increasing interest in models that estimate incident impacts on these streets. Such estimation will allow agencies to determine the need for special signal timing plans and other management strategies during incident conditions. In the absence of field measurements of the incident impact on arterial network performance, microscopic traffic simulation has been used as a powerful method to estimate these impacts (, ). However, the use of simulation models is expensive in terms of data collection, model input preparation, and calibration, particularly when the incident management systems need to be assessed at the regional levels and when the stochastic nature of incident attributes and locations need to be considered in the analysis. Queuing analysis has been more widely used to identify incident benefits than shock wave analysis (,). A variety of examples of the use of queuing analysis (-) and shock wave theory analysis (, ) for freeway incident impact assessments are available. Rakha and Zhang (0) concluded that queuing theory provides a simple and accurate technique for estimating delays at highway bottlenecks. However, when this estimation is done for arterial streets, simple queuing theory and shock wave analysis do not consider the impacts of the incidents on upstream and downstream signals. Earlier work by the research team of the current study (, ) concluded, that simple queuing analysis equations underestimate incident delays on signalized urban networks and recommended using a multiplication a factor of. to the incident delays calculated based on queuing equations.

4 Incident impacts on capacity is the most critical parameter to estimate the influence of incidents on network performance. The Highway Capacity Manual (HCM 00) () provides estimates of the drops in capacity due to incidents, as a function of the number of the blocked lanes and the total number of lanes of the freeway section. It should be noted that HCM 00 defines incident as any occurrence on a roadway that impedes normal traffic flow. For the purpose of this study, an incident is a recurrent event that results in drop in capacity due to lane or shoulder blockage due for example to crashes, debris, disabled vehicles, and so on. Other studies also reported estimates of the drop in capacity due to incidents (,0, ). The SHRP L0 study () developed equations to estimate saturation flow rate adjustment factor for incident presence at the stop line of a traffic signal for use as part of the assessment of incident impacts on travel time reliability of urban street facilities. The equation estimates the saturation flow adjustment factor as a function of number of lanes, number of lanes blocked by the incident, and coefficients related to incident severity. The equations, however, does not address incident locations other than at the stop line. This study investigates the capacity impacts of the interaction between the drop in capacity below demands at midblock urban street segment location and upstream and downstream signalized intersection operations. The models were developed based on data generated utilizing a microscopic simulation model that was calibrated by comparison to parameters suggested in the HCM 00 and SHRP-L0 project(,)for incident and no-incident conditions and by comparison with field measurements. The developed models will assist agencies in estimating incident impacts on urban street facilities to support the development and use of special signal plans based on incident attributes and urban network characteristics. In addition, the proposed models will allow agencies to assess the benefits of various incident management strategies. PROBLEM STATEMENT An incident occurring at a mid-block locations can decrease the throughput of the incident link and upstream and downstream intersections. Upstream intersection throughput capacity is affected, if the queue from the incident spillbacks to the upstream links. During the first parts of the upstream link green phases, the vehicles will be able to leave the stop lines of the feeding links at the saturation flow rates of these links until the queue due to the downstream incident spills back to the upstream signal. From the moment that this happens until the end of the green, the throughput of the upstream links will be controlled by the allowable throughput at the downstream incident location. Thus, the upstream movement greens can be divided into two parts. The first part, referred to as the unconstrained green in this paper. It refers to the green time that vehicles from upstream links can leave at the saturation flow rates of these links due to the availability of queuing storage at the downstream link. In the second part, referred to as the constrained green, the movements from the upstream links are controlled by the capacity at the incident location due to the spillback of the queues from the downstream incident. The result of having this constrained green is a reduction in the capacity of the upstream intersection feeding links, which causes an increase in the upstream movement delays. The unconstrained green portion is expected to increase as the incident is located further downstream from the stop line. In addition, higher reduction in incident capacity and higher demand at the incident location will result in an increase in the

5 constrained green of upstream intersection. It should be noted that this only happens when there is a spillback from the incident location to the upstream signal. On the other hand, incident capacity can be effected by downstream intersection queue, if it spillbacks to the upstream incident. In addition, a reduction in link capacity due to incidents can prevent vehicles from arriving at downstream signalized intersections. Consequently, this decreases downstream intersection queue discharges. The reduction in the upstream intersection throughputs are expected to be a function of how far is the incident from the upstream intersection, the volume to capacity (v/c) ratio at the incident location, and incident duration. Whereas, the capacity at the incident location (IC) is expected to be a function of how far is the incident from downstream intersection, its green to cycle-length (g/c) ratio, and number of total and blocked lanes. This study develops two models to estimate the impacts of incidents on capacity taking into consideration the interactions with upstream and downstream signalized intersections. The first model estimates the remaining capacity with lane-blockage incidents located at different distances from the downstream signal. This model considers the impact of the queue from the downstream signal and is a function of the g/c ratio of the downstream signal, distance of incident to downstream intersection, and number of the blocked lanes due to the incident. The second model estimates the reduction in the upstream intersection throughput due to incidents at different downstream locations. This model is a function of the v/c ratios, at the incident location, distance to the incident, and incident duration times. The v/c ratio is estimated based on the drop in capacity reported in the SHRP L0 project () and HCM [00] () as explained in the methodology section below. The models can be used together to estimate the delays at the incident location considering the drop in capacity at the upstream signal due to queue spillback from the incident locations. These models are derived based on microscopic simulation modeling. METHODOLOGY As stated above, this study estimates the interaction between incidents and upstream and downstream signalized intersections on capacity. The first step is to estimate the drop in link capacity due to incidents without this interaction. This estimation was done in this study based on the SHRP Program L0 project () and the HCM 00 () recommendations to estimate remaining capacity due to incidents. The estimates are % remaining capacity for one lane blockage incidents and % capacity remaining for two-lane blockage incidents of three lane segments. The SHRP L0 equations can be used only to estimate the capacity drop due to incidents at the movement stop lines (). This study uses these mentioned values to estimate the throughput at an incident location, calibrate a microscopic simulation model to produce the capacity at the incident location, and then examine the interactions with upstream and downstream signals to develop the models, as described in this section. Calibration of VISSIM for No-Incident Conditions In order to develop the mentioned models, the study used the VISSIM microscopic simulation tool. The used model was first calibrated for no-incident conditions based on real-world data to reflect measured traffic parameters for the utilized network. The manual calibration process was divided into two parts. The first is based on saturation flow rate and the second is based on traffic flow

6 performance measures. A three mile segment of Glades Road in Boca Raton an urban street in Boca Raton, FL was used as a case study. The segment has signalized intersections with an average signal spacing of,0 feet. The posted speed limit is mile per hour. The Glades Road network was coded in VISSIM and the results from the simulation was compared with real-world measurements. The first step was to perform an initial fine-tuning of the model parameters to produce saturation flow rates that are in agreement with the Highway Capacity Manual 00 (HCM 00) procedures () and previous observations from the field in South Florida. The target saturation flow rate for the purpose of this calibration was set to,0 to,00 passenger cars per hour per lane. The parameters of the VISSIM model were fine-tuned to produce this value. Two of the urban driver car following model in VISSIM (the Wiedemann driver behavior model) were fine-tuned in this study in order to obtain the target saturation flow rate: ) the additive part of desired safety distance (bx_add) and ) the multiplicative part of desired safety distance (bx_mult) (,, ). These parameters determine the target desired safety distance, which has a direct impact on saturation flow rate (). The most appropriate combination of the two parameters was found to be. feet for the additive part of the desired safety distance and. for the multiplicative part of the desired safety distance. The resulting saturation flow rate for the simulated thru movement was, passenger cars per hour per lane based on the average of ten simulation runs with different random seeds. The second part of the calibration process is to compare the traffic flow performance according to the Glades Road simulation model with real-world traffic data. Three traffic data types were used in the calibration including: ) signal timing plans of all intersections; ) historical turning movement counts of all intersections; ) Data from permanently installed magnetometers (from SENSYS ). These data types provide point measurements of speed, occupancy, and volume; as well as travel time measurements based on automatic vehicle re-identification. Virtual detectors were included in the simulation at the same locations of the detectors in the field. The measured and simulated parameters based on ten simulation runs with different seed numbers were examined to determine if further fine-tuning of the parameters was needed. It was determined that the Root-Mean-Square-Percentage-Errors (RMSPE) for the simulated versus measured volume and speed values were below % for both directions of travel. Modeling Incidents in Microscopic Simulation This section describes the calibration of microscopic simulation models for incident conditions in VISSIM. VISSIM does not allow the user to specify incidents in the model. Previous studies modeled incidents by setting up a red signal at the incident lane (), using vehicle(s) with zero speed at the time and location of the incident (), or using buses with dwelling times equal to the lane blockage durations (). In this study, the incidents were simulated in VISSIM using bus stops with dwell time equal to incident duration on the lanes blocked. This was combined with reduced speed area on the adjacent lanes to emulate driver slowing to observe the incident. (,, ). The length of reduce speed area in the vicinity lanes of the incident was modified by trial and error to achieve the mentioned expected drop in capacity due to the uninterrupted incident. Uninterrupted incident refers to the incident which is not be effected by downstream signal queue spillbacks.

7 INCIDENT CAPACITY Massahi, Hadi, Xiao 0 Assessing the Impacts of Downstream signal on Upstream Incident Capacity As discussed above, the first model developed in this study estimates the impact of queue spillbacks from downstream signalized intersection on the capacity of midblock incident locations. The modeled incidents are one and two lane blockage incidents on a three lane urban street segment at different distances from downstream intersection and different g/c ratios at the downstream intersection. The model is applicable for conditions, in which the capacity with incident conditions is below the demand at the incident location. The simulation started with a minutes warm-up period, followed by minutes analyzing period. In all simulated scenarios, a one hour incident was assumed to occur minutes after the simulation starts. The simulation model was used to assess the impacts of incidents on upstream intersection throughputs. The incident location and the downstream signal g/c ratio were varied and the incident capacity at the upstream link were assessed using the microscopic simulation. Figure shows the variation in the incident capacity with incident location and the g/c ratio of downstream signalized intersection, considering the segment number of lane blockage. UPSTREM INTERRUPTED INCIDENT CAPACITY DISTANCE FROM DOWNSTREM SIGNAL one lane blockage & g/c=0. 0 One lane blockage & g/c= Two lane blockage& g/c=0. 0 Two lane blokage & g/c= Figure:Interrupted Incident Capacity Effected by Downstream Intersection As can be seen from Figure, the drop in the incident capacity decreases with the increase in the distance from the downstream intersection. In addition, the capacity increases with the increase in the g/c ratio at the downstream intersection and decreases with the increase in the number of blocked lanes. Based on the simulation results for one lane blockage, the upstream incident capacity was not found to be affected by the downstream intersection, when the incidents

8 0 0 0 occurs at,0 ft and,0 ft from the downstream signal with the g/c ratio at the downstream signal equals 0. and 0., respectively. These values were selected based on typical g/c ratios for the main street in the analysis area. Reducing the g/c ratio to a level that constrains the departing volumes from the upstream link is expected to reduce the traffic volume arriving at the downstream incident location and this may reduce the portion of unconstrained green. For the two lane blockage incident, the capacity at the incident location was not affected significantly by the downstream intersection, when the incident occurs at 0 ft or more form the downstream signal. This distance is the same for both g/c ratios (0. and 0.). The data presented in Figure was used to develop regression models to estimate the incident capacity (IC) based on the investigated influencing factors. The developed regression models are presented in Table. The developed regression models show that there is a significant relationship between the incident capacity and the three independent variables, mentioned earlier, as indicated by the Coefficient of Determination (R-Squared) values and the t-test of the significance of the independent variable coefficients. If the g/c ratio for an assessed conditions is between the two g/c ratios assessed in this study, as presented in Table, interpolation can be used to estimate the capacity. TABLE : Upstream Interrupted Incident Capacity Regression Models Number of lane blockage One lane blockage Two lane blockage Downstream signal g/c Upstream Incident Capacity R 0. IC = 0.x IC = 0.x IC = E-0x x + 0.x IC = E-0x x +.0x Assessing the Impacts of Incidents on Upstream Intersection Throughputs A second model was developed in this study to estimate upstream intersection saturation flow as impacted by the drop in capacity due to a downstream signal. The saturation flow rate was assessed for incidents with different distances from the upstream intersection, different v/ic ratios at the incident location, and different incident durations. This was done by introducing incidents at different locations with different capacity drops and different incident durations in a VISSIM model. The used incident durations were minutes, minutes, and minutes. It should be noted that minutes is the median incident duration at the study location. The network used in the testing is part of the Glades Road network in Boca Raton, FL, mentioned earlier First, the network was simulated without an incident and then incidents with different attributes were introduced in the model. In all simulated incident scenarios, the incidents were assumed to occur minutes after the simulation starts. The simulation model was used to assess the impacts of incidents on upstream intersection throughputs. The maximum throughputs at the upstream intersection were assessed using the microscopic simulation. Figures to show the variation in the upstream intersection saturation flow rate with the incident location, incident duration and the v/c ratio at the incident location. In these figures, the incident location references the distance from the upstream signal stop line. It should be also noted that the saturation flow of the upstream signal without incident is, veh/hr. Thus, incidents

9 Incident Duration Intersection Saturation Flow Incident Duration Intersection Saturation Flow Massahi, Hadi, Xiao with downstream locations and v/c ratios that produce this upstream intersection saturation flow in the simulation are recognized as incidents that do not impact upstream signal operations. For example the investigated minute incident did not affect the upstream intersection throughput when it is located more than 0 ft, 00 ft and 000 ft from the upstream signalized intersection with the v/c ratio at the incident location of.,. and., respectively. v/ic= Distance minutes Minutes 0 Minutes 0 0 Minutes minutes Minutes minutes Minutes Minutes FIGURE : Upstream intersection saturation flow rate variation with incident location and duration with v/ic ratio equal to. at the incident location v/ic= Distance Minutes Minutes Minutes Minutes Minutes Minutes 0 Minutes Minutes Minutes FIGURE : Upstream intersection saturation flow rate variation with incident location and duration with v/ic equal to. at the incident location

10 Intersection Saturation Flow Incident Duration Massahi, Hadi, Xiao v/ic= Distance Minutes Minutes Minutes 00 Minutes Minutes Minutes 0 0 FIGURE : Upstream intersection saturation flow rate variation with incident location and Duration with v/ic equal to. at the incident location As can be seen from Figures to, the saturation flow rate decreases with the increase in the v/c ratio at the incident location and the decrease in the distance between the upstream intersection and the incident location. The incident duration effect on the saturation flow rate can be explained by the fact that in the first few minutes after the incident happen, the queue length does not always reach the upstream signals and thus the upstream intersection operates at close to the saturation flow rate during these few minutes. Since the proportions of these minutes compared to the incident duration are higher in case of shorter incidents, the calculated saturation flow rates of shorted incidents are somewhat higher. The duration of the unconstrained green, described in the problem statement was calculated based following relationships: MT = SF UG (TG UG) + IC C C Thus, the unconstrained green can be calculated as: UG = (MT c CI TG) (SF IC) Minutes Minutes Minutes Where UG is upstream intersection unconstrained green time, MT is intersection maximum throughput, TG is upstream intersection green time, C is its cycle length, IC is capacity at incident location and SF is saturation flow. () ()

11 Table shows the regression analysis models developed in this study to estimate the saturation flow rate and the unconstrained green of the upstream intersection as a function of the independent variables. x in the table is the incident distance from the upstream intersection.. TABLE : Regression Models to Estimate Upstream Intersection Saturation Flow and Unconstrained Green Regression Models Intersection Incident v/c at incident movement saturation R Duration location Unconstrained Green R Flow Minutes. SF = 0.x UG = 0.0x SF = 0.x UG = 0.00x SF = 0.x UG = 0.0x Minutes. SF = 0.x UG= 0.00x SF = 0.x UG = 0.0x SF = 0.x UG = 0.0x Minutes. SF = 0.x UG = 0.0x SF = 0.x UG = 0.0x SF = 0.x UG = 0.0x UTILIZATION OF THE DEVELOPED MODELS To estimate the incident delay, first the capacity at the incident location can be estimated using the models presented in Table. This capacity can be used as an input to the models presented in Table to determine the impact of the midblock incident on upstream intersection saturation flow rate. Once this done, the delay due to the incident can be estimated as a combination of the delay due to mid-block queuing on the incident link and the increase in upstream intersections delay due the reduction in the saturation flow rate resulting from queue spillback to the upstream intersection(). The delay due to queuing on the incident link can be calculated using deterministic queueing analysis equations, as is used in estimating incident delays on freeways. The increase in upstream intersection delay can be calculated using the signalized intersection control delay method presented in the HCM 00 (). The increase in upstream control delay can be calculated as the difference in delay when utilizing the HCM model with and without the reduction in saturation flow rate, estimated as in Table. CONCLUSIONS This study investigates the capacity impacts of the interaction between the drop in capacity below demands at midblock urban street segment location and upstream and downstream signalized intersection operations. A model was developed to estimate the drop in capacity at the incident location as a function of the number of blocked lanes, distance from downstream intersection, and 0

12 g/c ratio of the downstream signal. A second model was developed to estimate the reduction in the upstream intersection capacity due to the drop in capacity at midblock incident location as estimated by the first model. The second model estimates the drop in capacity of the upstream links feeding the incident locations as a function of incident duration time, v/c ratio at the incident location, and distance from an upstream signalized intersection. The models were developed based on data generated utilizing a microscopic simulation model that was calibrated by comparison to parameters suggested in the HCM 00 and SHRP-L0 project for incident and no-incident conditions and by comparison with field measurements. The developed regression models show significant relationships between the drops in incident capacity and the drops in upstream saturation flow as dependent variables in the models and the independent variables. The drop in capacity at the incident location increases as the incident gets closer to the downstream signal, as the downstream signal g/c ratio decreases, and as the number of blocked lanes increase. The drop in the upstream intersection link capacity flow increases as the incident gets closer to the upstream signal and as the incident duration and the v/c ratio at the incident location increases. The focus of this study was on three lanes arterial facility because three lane facilities are the most common type of the principal urban street facilities. However, the second model which estimate upstream intersection flow rate is applicable to facilities with other number of lanes. The developed models in this study can be used to assess the impacts of incidents and incident management strategies on system performance and can be used to support agencies in developing special signal timing plans to be applied under different incident conditions. Such plans can be applied for both downstream and upstream intersections of the incident locations to reduce incident impacts. The model developed in this study can be used to better assess the incident impacts and thus the benefits of incident management strategies. The estimation of these benefits can be used to estimate the return of incident management investment. In addition, the developed model can be used by traffic management agencies to estimate the saturation flow rates during incident that can be used to develop special timing plans to be implemented during incidents. Another application of the developed model is a part of dynamic traffic assignment applications to estimate the diversion due to the drop in link capacity caused by incidents. REFERENCES. Hong, C., Yue, Y, and Bin, Z. A Capacity Assessment Method on Urban Expressway after Traffic Incident. Procedia - Social and Behavioral Sciences., Vol, 0,pp... Highway Capacity Manual. TRB, National Research Council. Washington, D.C., Turnbull, K. F. Managing Travel for Planned Special Events: First National Conference Proceedings, New Orleans, Louisiana., December 00.. Chang, G. L. & Rochon, S. Performance Evaluation of CHART the Real Time Incident Management System in Year 00. Maryland Department of Transportation,00.. Khattak, A. J., Wang, X., Zhang, H. & Cetin, M. Primary and secondary incident management: Predicting durations in real time.report No. VCTIR -R.,0.

13 Sorensen, P., Wachs, M., Min, E. Y., Kofner, A. & Ecola, L. Moving Los Angeles: Short-term policy options for improving transportation, Rand Corporation., 00.. Systematics, C. Analytical procedures for determining the impacts of reliability mitigation strategies, Transportation Research Board., Report:S-L0-RR-., 0.. Economic Development Research Group, I., Cambridge Systematics, I., International, I., Institute, T. T. A. M. T. & Weris, I. Development Of Tools For Assessing Wider Economic Benefits Of Transportation,0.. Birst, S. & Smadi, A. An Evaluation of ITS for Incident Management in Second-Tier Cities: A Fargo, ND Case Study. ITS America 0th Annual Meeting and Exposition: Revolutionary Thinking, Real Results, Glassco, R. A., Proper, A., Shah, V. & Wunderlich, K. Studies Of Potential Intelligent Transportation Systems Benefits Using Traffic Simulation Modeling-Volume.,.. McCandless, J.The Calibration, Validation, and Comprision Of VISSIM Simulations Using The Two -Fluid Model. Master of science thesis, 00.. Vissim Version Manual. PTV Planning Transport, Verkehr AG, Karlsruhe, Germany, 0.. Yang, S., Mei, F, and Hadi, M. Estimation of Incident Delays on Arterial Streets. Presented at th Annual Meeting of the Transportation Research Board, Washington, D.C., 00.. Xiao, Y., Hadi, M, and Ozen, H., Mysore, V. An ITS Evaluation Tool in the FSUTMS Regional Demand Modeling Environment. In Transportation Research Record: Journal of the Transportation Research Board, No., Transportation Research Board of the National Academies, Washington, D.C., 00, pp. -.. Hadi, M., Sinha, P, and Wang, A. Modeling Reductions in Freeway Capacity due to Incidents in Microscopic Simulation Models In Transportation Research Record: Journal of the Transportation Research Board, No., Transportation Research Board of the National Academies, Washington, D.C., 00, pp. -.. Gomes, Gabriel., May, Adolf,. and Horowitz, Roberto. Congested Freeway Microsimulation model using Vissim. In Transportation Research Record: Journal of the Transportation Research Board, No., Transportation Research Board of the National Academies, Washington, D.C., 00, pp. -.. Zhou, H, and feng, T. Modeling Analysis of Incident and Roadway Clearance Time. presented at th International conference on traffic and transportation studies, China, Procedia - Social and Behavioral Sciences vol, 0, pp.. Rakha, H, and Zhang, W. Consistency of Shock-Wave and Queuing Theory Procedures for Analysis of Roadway Bottlenecks. Presented at th Annual Meeting of the Transportation Research Board, Washington, D.C,00.. May, A. D. Traffic Flow Fundamentals. Prentice Hall Inc., Englewood Cliffs, New Jersey, Knoop, V., Hoogendoorn, S, and van Zuylen, H. Capacity Reduction at Incidents, Empirical Data Collected from a Helicopter., In Transportation Research Record: Journal of the Transportation Research Board,No. 0, Transportation Research Board of the National Academies, Washington, D.C., 00, pp. -.. Highway Capacity Manual. TRB, National Research Council. Washington, D.C., 00.. Qin, L., and B. L. Smith. Characterization of Accident Capacity Reduction. Report UVACTS Center for Transportation Studies, University of Virginia, Charlottesville, 00.

14 0. Kittelson & Associates, Inc.SHRP Project L0 Incorporation of Travel Time Reliability into the Highway Capacity Manual Kim, Seung-Jun, Wonho Kim, and L. Rilett. "Calibration of microsimulation models using nonparametric statistical techniques." Transportation Research Record: Journal of the Transportation Research Board, 00, pp.-.. Al-Nuaimi, H.,Ayers, R, and Somasundaraswaran, K., Traffic Flow Modeling For Urban Arterial Road in BRISBANE City. Presented at th REAAA Conference, 0.. Oregon Department of Transportation, Protocol for VISSIM Simulation. June, 0.. Avetisyan, G., Hooks, E., Melanta, S, and Qi, B. Effects of vehicle technologies, traffic volume changes, incidents and work zones on greenhouse gas emissions production., Transportation Research Part D., Vol., 0, pp. 0.. Massahi, A., Ai., Hadi, M., Xiao, Y., Tao, W., and Chen X,. Improved Model for Estimating Incident Impact on Urban Street Travel Time with Consideration of Upstream Intersection Capacity Reduction. Presented at the th Annual Meeting of the Transportation Research Board, National Research Council, Washington, D.C., January 0.