Faced with continuously increasing maintenance due to aging infrastructure, the Texas Department of Transportation (TxDOT) is evaluating the potential impact of reduced funding on highway safety. The main objective of this thesis is to develop a methodological procedure to identify threshold levels of pavement skid resistance for highways in the context of traffic crashes, assisting TxDOT Administration and engineers in making proper maintenance decisions. As a result, the efficiency and safety of the highway system could be preserved. The scope of this study covers all types of state-maintained highways in Texas. The primary objectives of this thesis include: 1) synthesis of literature; 2) quantification of the relationship between crash risk and pavement skid resistant; 3) determination of critical skid resistant threshold levels; and, 4) benefit cost analysis. A detailed methodology framework was developed and a comprehensive database was generated from four data files containing pavement, geometry, traffic, and crash information to support this research. The impact of skid resistance level on crash risks was proven to be significant based on the results of regression analysis and insights provided by TxDOT experts. The quantitative relationships between crash risk and skid resistance were quantified using the Crash Rate Ratio method. Hierarchical structure grouping was used to categorize the entire network into homogeneous groups based on traffic level, roadway alignment and other factors. Critical skid resistance threshold levels were determined for the whole state as well as for stratified highway groups. Finally, benefit/cost ratio analyses were conducted to evaluate the effectiveness of pavement maintenance treatments to restore or increase skid resistance. / text

Increases in bicycling in the United States results in increased exposure to crashes and injuries. This research focuses on the factors involved in bicycle crashes in the United States and the state of Iowa. Data from the U.S. Nationwide Inpatient Sample and the Iowa Department of Transportation were used to address three aims: 1) estimate the burden and examine the outcomes of bicycle crashes resulting in hospitalizations nationwide by motor vehicle involvement, 2) describe how bicycle motor vehicle crashes vary by intersection and non-intersection in Iowa, and 3) identify the impact of on-road bicycle facilities on bicycle-motor vehicle crashes in Iowa. Using the U.S. Nationwide Inpatient Sample, years 2002-2009, the estimated annual burden of injury from bicycle-related hospitalizations equated to a billion dollars in hospital charges, over 100,000 days in the hospital, and over 300 in-hospital deaths. We also found that bicycling crashes involving motor vehicles had more hospital charges, longer stays, and greater odds of in-hospital death. We also used the Iowa Department of Transportation crash database, 2001 to 2010, to examine risk factors for bicycle-motor vehicle (BMV) crash locations. We found that BMV crashes involve risk factors at person, crash, environment, and population levels and these vary by intersection and non-intersection. Compared to intersections, non-intersection crashes were more likely to involve young bicyclists (0-9 years), locations outside city limits, with driver vision obscured, reduced lighting on the roadway and less likely involve failure to yield right of way. Finally, we conducted a case site-control site study in Iowa, using crash data from 2007 to 2010 to investigate the impact of pavement markings (bicycle lanes and shared lane arrows) and bicycle-specific signage on crash risk. Our results suggest that bicycle facilities are protective against crashes, with the most protective being the combination of both pavement markings and signage, followed by pavement markings alone, and then signage alone. This project shows that bicycling carries a large burden of injury in the United States and that there are many contributing factors to bicycle crashes. It also provides evidence suggesting that infrastructure changes can decrease crash occurrence and there opportunities to intervene at other levels (e.g., person factors) to have an even greater impact overall.

Reducing the number of motor vehicle crashes is one of the major challenges of our times. Current strategies to reduce crash rates can be divided into two groups: identifying risky driving behavior prior to crashes to proactively reduce risk and automating some or all human driving tasks using intelligent vehicle systems such as Advanced Driver Assistance Systems (ADAS) and Automated Driving Systems (ADS). For successful implementation of either strategy, a deeper understanding of human driving behavior is essential. This dissertation characterizes human driving behavior through an analysis of a large naturalistic driving study and offers four major contributions to the field. First, it describes the creation of the Surface Accelerations Reference, a catalog of all longitudinal and lateral surface accelerations found in the Second Strategic Highway Research Program Naturalistic Driving Study (SHRP 2 NDS). SHRP 2 NDS is the largest naturalistic driving study in the world with 34.5 million miles of data collected from over 3,500 participants driving in six separate locations across the United States. An algorithm was developed to detect each acceleration epoch and summarize key parameters, such as the mean and maxima of the magnitude, roadway properties, and driver inputs. A statistical profile was then created for each participant describing their acceleration behavior in terms of rates, percentiles, and the magnitude of the strongest event in a distance threshold. The second major contribution is quantifying the effect of several factors that influence acceleration behavior. The rate of mild to harsh acceleration epochs was modeled using negative binomial distribution-based generalized linear mixed effect models. Roadway speed category, driver age, driver gender, vehicle class, and location were used as fixed effects, and a unique participant identifier was as the random effect. Subcategories of each fixed effect were compared using incident rate ratios. Roadway speed category was found to have the largest effect on acceleration behavior, followed by driver age, vehicle class, and location. This methodology accounts for the major influences while simultaneously ensuring that the comparisons are meaningful and not driven by coincidences of data collection. The third major contribution is the extraction of acceleration-based long-term driving styles and determining their relationship to crash risk. Rates of acceleration epochs experienced on ≤ 30 mph roadways were used to cluster the participants into four groups. The metrics to cluster the participants were chosen so that they represent long-term driving style and not short-term driving behavior being influenced by transient traffic and environmental conditions. The driving style was also correlated to driving risk by comparing the crash rates, near-crash rates, and speeding behavior of the participants. Finally, the fourth major contribution is the creation of a set of interactive analytics tools that facilitate quick characterization of human driving during regular as well as safety-critical driving events. These tools enable users to answer a large and open-ended set of research questions that aid in the development of ADAS and ADS components. These analytics tools facilitate the exploration of queries such as how often do certain scenarios occur in naturalistic driving, what is the distribution of key metrics during a particular scenario, or what is the relative composition of various crash datasets? Novel visual analytics principles such as video on demand have been implemented to accelerate the sense-making loop for the user. / Doctor of Philosophy / Naturalistic driving studies collect data from participants driving their own vehicles over an extended period. These studies offer unique perspectives in understanding driving behavior by capturing routine and rare events. Two important aspects of understanding driving behavior are longitudinal acceleration, which indicates how people speed up or slow down, and lateral acceleration, which shows how people take turns. In this dissertation, millions of miles of driving data were analyzed to create an open access acceleration database representing the driving profiles of thousands of drivers. These profiles are useful to understand and model human driving behavior, which is essential for developing advanced vehicle systems and smart roadway infrastructure. The acceleration database was used to quantify the effect of various roadway properties, driver demographics, vehicle classification, and environmental factors on acceleration driving behavior. The acceleration database was also used to define distinct driving styles and their relationship to driving risk. A set of interactive analytics tools was developed that leverage naturalistic driving data by enabling users to ask a large set of questions and facilitate open-ended analysis. Novel visualization and data presentation techniques were developed to help users extract deeper insight about driving behavior faster than previously exiting tools. These tools will aid in the development and testing of automated driving systems and advanced driver assistance systems.

This thesis has evaluated the effectiveness of GDL programs both in New Jersey and across the United States using several metrics. The New Jersey GDL program was analyzed because it is considered one of the most stringent programs in the country. It was found that GDL indeed reduces the per capita rate of crashes for teen drivers in New Jersey. However, no statistical difference was seen in the rate of fatalities in teen driver crashes. The per capita rate of violations for 16 and 17 year old drivers was lower after GDL, but the rate of point-carrying violations increased for 19 and 20 year old drivers who were licensed under GDL. The September, 2008 directive by the New Jersey Attorney General banning plea-agreements for teens significantly reduced the rate of violations further for 16 and 17 year old GDL drivers. The factors that led to teen crashes did not change in the United States after GDL. Teen drivers are still prone to distractions and inappropriate behavior while driving. Teen drivers also have higher rates of control loss and road departure crashes when compared to adults. Finally, it was found changes in the number teen driver crashes and fatalities are associated with similar changes in travel exposure. Teen crashes and fatalities have dropped with the implementation of GDL but teen VMT has also dropped. Graduated driver's licensing did not change the reasons for teen driver crashes. Also, it is likely that any reductions in the number of teen crashes or fatalities are associated with reductions in exposure and not changes in teen driver behavior. / Ph. D.

The overall aim was to elucidate the effects of vehicle passengers on young drivers. This generated two specific aims and four papers. The first aim was to investigate the crash risk for young drivers with passengers and to establish whether such accidents involve any special circumstances compared to those that occur without passengers. This goal was achieved by analysing accident and exposure data from two registers. The second objective was to analyse the group processes that develop between four young men in a vehicle and to ascertain how those interactions affect driving behaviour. Those issues were addressed by performing an observational study of twelve young men driving an instrumented vehicle in real traffic with and without passengers. The interactions between the vehicle occupants were video and audio recorded, and the driving behaviour was registered in various ways. The results show that drivers with passengers have a lower crash risk compared to those driving alone regardless of the driver’s age, although this effect is weaker for young drivers (especially males) than for other age groups. Compared to driving alone, driving with passengers for young drivers is more extensively associated with single-vehicle crashes that occur at night, on weekends, and in rural areas on roads with higher speed limits, and it leads to more severe outcomes. It has also been found that the passengers sometimes try to induce the young drivers to act in either safer or more dangerous ways, although the drivers very often resist urging and coaxing from their passengers. Cohesion is another factor that affect the driver-passenger group: a high level of cohesion, especially task cohesion, is associated with a low number of unsafe driving actions. Consequently, it seems that the presence of passengers is not enough to ensure safe driving—substantial group cohesion is also necessary for such behaviour. / Det övergripande syftet med denna avhandling var att studera passagerares effekt på unga förare, vilket genererade två delsyften och fyra delarbeten. Det första delsyftet var att undersöka olycksrisken för unga förare med passagerare och att ta reda på om dessa olyckor skedde under några speciella omständigheter. För att få svar på syftet gjordes en registerstudie där olycks- och exponeringsdata från två olika register analyserades. Det andra delsyftet var att analysera de grupprocesser som utvecklas mellan fyra unga män i en bil och att studera hur dessa interaktioner påverkar körbeteendet. Detta undersöktes med en observationsstudie där tolv unga män fick köra en instrumenterad bil i verklig trafik, både med och utan passagerare. Interaktionerna som uppstod i bilen spelades in med hjälp av videokameror och mikrofoner och körbeteendet registrerades med olika mätinstrument. Resultatet visar att förare med passagerare har en lägre olycksrisk jämfört med förare utan passagerare oavsett förarens ålder. Denna effekt är inte lika stark för unga förare (speciellt inte unga män) som den är för övriga åldersgrupper. Unga förares olyckor med passagerare är också mer vanligt förekommande under speciella omständigheter. Jämfört med olyckor utan passagerare sker de i större utsträckning under nattetid, under veckoslut, i tätbebyggt område, på vägar med hög hastighetsbegränsning, de är oftare singelolyckor och de får mer allvarliga konsekvenser. Vidare visar resultaten att passagerare ibland, på olika sätt, försöker få den unga föraren att köra på ett annat sätt än vad de gör vilket kan betyda ett säkrare eller ett mer trafikfarligt beteende. Det verkar dock som att förarna oftast står emot dessa övertalningsförsök och förolämpningar. En annan faktor som påverkar gruppen är kohesion: en hög grad av kohesion, speciellt uppgiftskohesion, visar sig leda till färre trafikfarliga körbeteenden. Med andra ord verkar det som att blotta närvaron av passagerare inte räcker för att få ett säkert körbeteende; det behövs även en betydande grad av kohesion.

<p>This dissertation is a collection of papers with the unifying objective being to better understand crash risk in foreign exchange markets. I investigate how exposure to the risk of currency crashes is able to provide a unified rationalization of the returns of various sorted currency portfolios.</p><p>In the first chapter, I identify an aggregate global currency skewness risk factor, which I denote SKEW. Currency portfolios that have higher average excess returns covary more positively with this risk factor. They suffer losses in times when high interest rate investment currencies have a greater tendency to depreciate sharply as a group relative to low interest rate funding currencies. Consequently, they earn higher average excess returns as reward for exposure to this risk. I create three sets of sorted currency portfolios reflecting three distinct sources of variation in average excess currency returns. The first set sorts currencies based on interest rate differentials. The second set sorts currencies based on currency momentum. The third set sorts currencies based on currency undervaluedness relative to purchasing power power parity (PPP) implied exchange rates. I find that differences in exposure to the global currency skewness risk factor can explain the systematic variation in average excess currency returns within all three groups of portfolios much better than existing foreign exchange risk factors in the literature.</p><p>In the second chapter, I build on the first chapter by studying the extent to which currency crash risk is predictable or unpredictable and whether the pricing power of aggregate currency skewness, uncovered in the first chapter, is due to unpredictable or predictable crash risk. Focusing on currency crash risk proxied using realized currency skewness at both the individual currency level and at the aggregate level using the SKEW risk factor introduced in the first chapter, I investigate whether either form of crash risk is predictable using only past information about crash risk. In particular, I use past information on both individual currency level and aggregate level measures based on both lagged realized currency skewness and lagged option implied risk neutral skewness. I find evidence that there is not much predictability at the individual country level or at the aggregate level over the full sample period considered. However, there is some evidence of predictability at the aggregate level since 1999, and especially so when option implied risk neutral skewness measures are used. Additionally, I use the predictions of SKEW and conduct asset pricing similar to that in chapter 1 using predicted and unpredicted SKEW to see whether its pricing power comes from predictable or unpredictable components. I find evidence that it is unpredictable currency crash risk that is very important, as the asset pricing results are largely identical when either SKEW or SKEW forecast errors are used. and whether the pricing power of</p> / Dissertation

In the United States, intersections are among the most frequent locations for crashes. One of the major problems at signalized intersection is the dilemma zone, which is caused by false driver behavior during the yellow interval. This research evaluated driver behavior during the yellow interval at signalized intersections and compared different dilemma zone countermeasures. The study was conducted through four stages. First, the driver behavior during the yellow interval were collected and analyzed. Eight variables, which are related to risky situations, are considered. The impact factors of drivers' stop/go decisions and the presence of the red-light running (RLR) violations were also analyzed. Second, based on the field data, a logistic model, which is a function of speed, distance to the stop line and the lead/follow position of the vehicle, was developed to predict drivers' stop/go decisions. Meanwhile, Cellular Automata (CA) models for the movement at the signalized intersection were developed. In this study, four different simulation scenarios were established, including the typical intersection signal, signal with flashing green phases, the intersection with pavement marking upstream of the approach, and the intersection with a new countermeasure: adding an auxiliary flashing indication next to the pavement marking. When vehicles are approaching the intersection with a speed lower than the speed limit of the intersection approach, the auxiliary flashing yellow indication will begin flashing before the yellow phase. If the vehicle that has not passed the pavement marking before the onset of the auxiliary flashing yellow indication and can see the flashing indication, the driver should choose to stop during the yellow interval. Otherwise, the driver should choose to go at the yellow duration. The CA model was employed to simulate the traffic flow, and the logistic model was applied as the stop/go decision rule. Dilemma situations that lead to rear-end crash risks and potential RLR risks were used to evaluate the different scenarios. According to the simulation results, the mean and standard deviation of the speed of the traffic flow play a significant role in rear-end crash risk situations, where a lower speed and standard deviation could lead to less rear-end risk situations at the same intersection. High difference in speed are more prone to cause rear-end crashes. With Respect to the RLR violations, the RLR risk analysis showed that the mean speed of the leading vehicle has important influence on the RLR risk in the typical intersection simulation scenarios as well as intersections with the flashing green phases' simulation scenario. Moreover, the findings indicated that the flashing green could not effectively reduce the risk probabilities. The pavement marking countermeasure had positive effects on reducing the risk probabilities if a platoon's mean speed was not under the speed used for designing the pavement marking. Otherwise, the risk probabilities for the intersection would not be reduced because of the increase in the RLR rate. The simulation results showed that the scenario with the pavement marking and an auxiliary indication countermeasure, which adds a flashing indication next to the pavement marking, had less risky situations than the other scenarios with the same speed distribution. These findings suggested the effectiveness of the pavement marking and an auxiliary indication countermeasure to reduce both rear-end collisions and RLR violations than other countermeasures.

Relevance of reactive traffic management strategies such as freeway incident detection has been diminishing with advancements in mobile phone usage and video surveillance technology. On the other hand, capacity to collect, store, and analyze traffic data from underground loop detectors has witnessed enormous growth in the recent past. These two facts together provide us with motivation as well as the means to shift the focus of freeway traffic management toward proactive strategies that would involve anticipating incidents such as crashes. The primary element of proactive traffic management strategy would be model(s) that can separate 'crash prone' conditions from 'normal' traffic conditions in real-time. The aim in this research is to establish relationship(s) between historical crashes of specific types and corresponding loop detector data, which may be used as the basis for classifying real-time traffic conditions into 'normal' or 'crash prone' in the future. In this regard traffic data in this study were also collected for cases which did not lead to crashes (non-crash cases) so that the problem may be set up as a binary classification. A thorough review of the literature suggested that existing real-time crash 'prediction' models (classification or otherwise) are generic in nature, i.e., a single model has been used to identify all crashes (such as rear-end, sideswipe, or angle), even though traffic conditions preceding crashes are known to differ by type of crash. Moreover, a generic model would yield no information about the collision most likely to occur. To be able to analyze different groups of crashes independently, a large database of crashes reported during the 5-year period from 1999 through 2003 on Interstate-4 corridor in Orlando were collected. The 36.25-mile instrumented corridor is equipped with 69 dual loop detector stations in each direction (eastbound and westbound) located approximately every ½ mile. These stations report speed, volume, and occupancy data every 30-seconds from the three through lanes of the corridor. Geometric design parameters for the freeway were also collected and collated with historical crash and corresponding loop detector data. The first group of crashes to be analyzed were the rear-end crashes, which account to about 51% of the total crashes. Based on preliminary explorations of average traffic speeds; rear-end crashes were grouped into two mutually exclusive groups. First, those occurring under extended congestion (referred to as regime 1 traffic conditions) and the other which occurred with relatively free-flow conditions (referred to as regime 2 traffic conditions) prevailing 5-10 minutes before the crash. Simple rules to separate these two groups of rear-end crashes were formulated based on the classification tree methodology. It was found that the first group of rear-end crashes can be attributed to parameters measurable through loop detectors such as the coefficient of variation in speed and average occupancy at stations in the vicinity of crash location. For the second group of rear-end crashes (referred to as regime 2) traffic parameters such as average speed and occupancy at stations downstream of the crash location were significant along with off-line factors such as the time of day and presence of an on-ramp in the downstream direction. It was found that regime 1 traffic conditions make up only about 6% of the traffic conditions on the freeway. Almost half of rear-end crashes occurred under regime 1 traffic regime even with such little exposure. This observation led to the conclusion that freeway locations operating under regime 1 traffic may be flagged for (rear-end) crashes without any further investigation. MLP (multilayer perceptron) and NRBF (normalized radial basis function) neural network architecture were explored to identify regime 2 rear-end crashes. The performance of individual neural network models was improved by hybridizing their outputs. Individual and hybrid PNN (probabilistic neural network) models were also explored along with matched case control logistic regression. The stepwise selection procedure yielded the matched logistic regression model indicating the difference between average speeds upstream and downstream as significant. Even though the model provided good interpretation, its classification accuracy over the validation dataset was far inferior to the hybrid MLP/NRBF and PNN models. Hybrid neural network models along with classification tree model (developed to identify the traffic regimes) were able to identify about 60% of the regime 2 rear-end crashes in addition to all regime 1 rear-end crashes with a reasonable number of positive decisions (warnings). It translates into identification of more than ¾ (77%) of all rear-end crashes. Classification models were then developed for the next most frequent type, i.e., lane change related crashes. Based on preliminary analysis, it was concluded that the location specific characteristics, such as presence of ramps, mile-post location, etc. were not significantly associated with these crashes. Average difference between occupancies of adjacent lanes and average speeds upstream and downstream of the crash location were found significant. The significant variables were then subjected as inputs to MLP and NRBF based classifiers. The best models in each category were hybridized by averaging their respective outputs. The hybrid model significantly improved on the crash identification achieved through individual models and 57% of the crashes in the validation dataset could be identified with 30% warnings. Although the hybrid models in this research were developed with corresponding data for rear-end and lane-change related crashes only, it was observed that about 60% of the historical single vehicle crashes (other than rollovers) could also be identified using these models. The majority of the identified single vehicle crashes, according to the crash reports, were caused due to evasive actions by the drivers in order to avoid another vehicle in front or in the other lane. Vehicle rollover crashes were found to be associated with speeding and curvature of the freeway section; the established relationship, however, was not sufficient to identify occurrence of these crashes in real-time. Based on the results from modeling procedure, a framework for parallel real-time application of these two sets of models (rear-end and lane-change) in the form of a system was proposed. To identify rear-end crashes, the data are first subjected to classification tree based rules to identify traffic regimes. If traffic patterns belong to regime 1, a rear-end crash warning is issued for the location. If the patterns are identified to be regime 2, then they are subjected to hybrid MLP/NRBF model employing traffic data from five surrounding traffic stations. If the model identifies the patterns as crash prone then the location may be flagged for rear-end crash, otherwise final check for a regime 2 rear-end crash is applied on the data through the hybrid PNN model. If data from five stations are not available due to intermittent loop failures, the system is provided with the flexibility to switch to models with more tolerant data requirements (i.e., model using traffic data from only one station or three stations). To assess the risk of a lane-change related crash, if all three lanes at the immediate upstream station are functioning, the hybrid of the two of the best individual neural network models (NRBF with three hidden neurons and MLP with four hidden neurons) is applied to the input data. A warning for a lane-change related crash may be issued based on its output. The proposed strategy is demonstrated over a complete day of loop data in a virtual real-time application. It was shown that the system of models may be used to continuously assess and update the risk for rear-end and lane-change related crashes. The system developed in this research should be perceived as the primary component of proactive traffic management strategy. Output of the system along with the knowledge of variables critically associated with specific types of crashes identified in this research can be used to formulate ways for avoiding impending crashes. However, specific crash prevention strategies e.g., variable speed limit and warnings to the commuters demand separate attention and should be addressed through thorough future research.

Recent research at the University of Central Florida addressing crashes on Interstate-4 in Orlando, Florida has led to the creation of new statistical models capable of calculating the crash risk on the freeway (Abdel-Aty et al., 2004; 2005, Pande and Abdel-Aty, 2006). These models yield the rear-end and lane-change crash risk along the freeway in real-time by using static information at various locations along the freeway as well as real-time traffic data that is obtained from the roadway. Because these models use the real-time traffic data, they are capable of calculating the respective crash risk values as the traffic flow changes along the freeway. The purpose of this study is to examine the potential of two Intelligent Transportation System strategies for reducing the crash risk along the freeway by changing the traffic flow parameters. The two ITS measures that are examined in this research are route diversion and ramp metering. Route diversion serves to change the traffic flow by keeping some vehicles from entering the freeway at one location and diverting them to another location where they may be more efficiently inserted into the freeway traffic stream. Ramp metering alters the traffic flow by delaying vehicles at the freeway on-ramps and only allowing a certain number of vehicles to enter at a time. The two strategies were tested by simulating a 36.25 mile section of the Interstate-4 network in the PARAMICS micro-simulation software. Various implementations of route diversion and ramp metering were then tested to determine not only the effects of each strategy but also how to best apply them to an urban freeway. Route diversion was found to decrease the overall rear-end and lane-change crash risk along the network at free-flow conditions to low levels of congestion. On average, the two crash risk measures were found to be reduced between the location where vehicles were diverted and the location where they were reinserted back into the network. However, a crash migration phenomenon was observed at higher levels of congestion as the crash risk would be greatly increased at the location where vehicles were reinserted back onto the network. Ramp metering in the downtown area was found to be beneficial during heavy congestion. Both coordinated and uncoordinated metering algorithms showed the potential to significantly decrease the crash risk at a network wide level. When the network is loaded with 100 percent of the vehicles the uncoordinated strategy performed the best at reducing the rear-end and lane-change crash risk values. The coordinated strategy was found to perform the best from a safety and operational perspective at moderate levels of congestion. Ramp metering also showed the potential for crash migration so care must be taken when implementing this strategy to ensure that drivers at certain locations are not put at unnecessary risk. When ramp metering is applied to the entire freeway network both the rear-end and lane-change crash risk is decreased further. ALINEA is found to be the best network-wide strategy at the 100 percent loading case while a combination of Zone and ALINEA provides the best safety results at the 90 percent loading case. It should also be noted that both route diversion and ramp metering were found to increase the overall network travel time. However, the best route diversion and ramp metering strategies were selected to ensure that the operational capabilities of the network were not sacrificed in order to increase the safety along the freeway. This was done by setting the maximum allowable travel time increase at 5% for any of the ITS strategies considered.

Recent research at the University of Central Florida involving crashes on Interstate-4 in Orlando, Florida has led to the creation of new statistical models capable of determining the crash risk on the freeway (Abdel-Aty et al., 2004; 2005, Pande and Abdel-Aty, 2006). These models are able to calculate the rear-end and lane-change crash risks along the freeway in real-time through the use of static information at various locations along the freeway as well as the real-time traffic data obtained by loop detectors. Since these models use real-time traffic data, they are capable of calculating rear-end and lane-change crash risk values as the traffic flow conditions are changing on the freeway. The objective of this study is to examine the potential benefits of variable speed limit implementation techniques for reducing the crash risk along the freeway. Variable speed limits is an ITS strategy that is typically used upstream of a queue in order to reduce the effects of congestion. By lowering the speeds of the vehicles approaching a queue, more time is given for the queue to dissipate from the front before it continues to grow from the back. This study uses variable speed limit strategies in a corridor-wide attempt to reduce rear-end and lane-change crash risks where speed differences between upstream and downstream vehicles are high. The idea of homogeneous speed zones was also introduced in this study to determine the distance over which variable speed limits should be implemented from a station of interest. This is unique since it is the first time a dynamic distance has been considered for variable speed limit implementation. Several VSL strategies were found to successfully reduce the rear-end and lane-change crash risks at low-volume traffic conditions (60% and 80% loading conditions). In every case, the most successful treatments involved the lowering of upstream speed limits by 5 mph and the raising of downstream speed limits by 5 mph. In the free-flow condition (60% loading), the best treatments involved the more liberal threshold for defining homogeneous speed zones (5 mph) and the more liberal implementation distance (entire speed zone), as well as a minimum time period of 10 minutes. This treatment was actually shown to significantly reduce the network travel time by 0.8%. It was also shown that this particular implementation strategy (lowering upstream, raising downstream) is wholly resistant to the effects of crash migration in the 60% loading scenario. In the condition approaching congestion (80% loading), the best treatment again involved the more liberal threshold for homogeneous speed zones (5 mph), yet the more conservative implementation distance (half the speed zone), along with a minimum time period of 5 minutes. This particular treatment arose as the best due to its unique capability to resist the increasing effects of crash migration in the 80% loading scenario. It was shown that the treatments implementing over half the speed zone were more robust against crash migration than other treatments. The best treatment exemplified the greatest benefit in reduced sections and the greatest resistance to crash migration in other sections. In the 80% loading scenario, the best treatment increased the network travel time by less than 0.4%, which is deemed acceptable. No treatment was found to successfully reduce the rear-end and lane-change crash risks in the congested traffic condition (90% loading). This is attributed to the fact that, in the congested state, the speed of vehicles is subject to the surrounding traffic conditions and not to the posted speed limit. Therefore, changing the posted speed limit does not affect the speed of vehicles in a desirable manner. These conclusions agree with Dilmore (2005).