The use of F1 hybrids for commercial production of such cross-pollinated crops as corn, sorghum, sugar beets, onions, and pearl millet suggests the feasibility of using this technique for alfalfa. Production of F1 hybrids of commercial value is dependent on the use of breeding material expressing good combining ability. In order to obtain precise estimates of combining ability for quantitative characters in alfalfa, it is necessary to produce all possible single crosses among a number of parents. The single cross seed required is difficult to obtain due to the vegetative reproduction and isolation required; and, subsequently, limited testing of this type has been conducted in alfalfa. Testing breeding material for combining ability based on seed production has been more limited than testing based on forage yield or various other measurements. In this experiment a diallel crossing system was used to test the general and specific combining ability of five alfalfa clones previously selected for good general combining ability. The report is based on first-year data of a three-year study, and the results are subject to errors which may occur due to variability inherent in the year of establishment. This is especially true for conclusions based on seed production. However, first-year data should be valid for such characteristics as flower color and seedling height. The analysis of seed production and seedling height is designed to measure the relative amount of general and specific combining ability of the cones involved. Reciprocals of the single crosses were evaluated for flower color, seedling height, and seed yield to check if reciprocal cross progeny give equal performance.

Les convertisseurs de fréquences optiques utilisant les processus non linéaires d’ordre deux permettent d’étendre la gamme spectrale accessible aux sources lasers. Celle-ci est en effet limitée par les bandes de gain des milieux lasers disponibles. Par ailleurs, la combinaison cohérente par contrôle actif de la phase est une technique efficace permettant la montée en puissance des sources lasers. Elle nécessite toutefois des modulateurs de phase rapides qui ne sont disponibles commercialement qu’aux longueurs d’onde standard.L’objectif de cette thèse est d'appliquer la combinaison cohérente à des convertisseurs de fréquences en utilisant la relation de phase inhérente au processus non linéaire.Cela permet de contrôler la phase de l’onde générée en agissant sur la phase de l’onde de pompe. C’est ce qu’on appelle le contrôle indirect de la phase.Pour cela, une étude théorique a été menée afin de s’assurer de la compatibilité de la technique de combinaison cohérente par marquage en fréquence avec le contrôle indirect de la phase.La démonstration expérimentale a d’abord été effectuée dans le cas le plus simple, la génération de seconde harmonique, qui met en jeu trois ondes dont deux dégénérées.Enfin, on s’est intéressé au cas de la génération de différence de fréquences dans le moyen infrarouge, où trois ondes sont mises en jeu.Dans ces deux cas, la qualité de mise en phase mesurée est excellente. / Laser emission wavelength is limited by the gain bandwidth of available laser media. Optical frequency converters rely on second order nonlinear processes to overcome this limitation, and give access to new wavelengths outside of the emission range of lasers.Besides, coherent beam combining with active phase control is an efficient technique to power scale laser sources. However, it requires fast phase modulators, some commercially available devices, but only at standard laser wavelengths.The objective of this thesis is to perform coherent combining of frequency converters, thanks to the phase-matching condition required for efficient nonlinear processes to take place.This relation allows for indirect phase control, where the converted-wave phase is tuned through direct phase control of the pump wave.A theoretical study has been carried out to confirm that indirect phase control was compatible with frequency-tagging coherent combining.Then, coherent combining through indirect phase control has been demonstrated experimentally in the simplest case of the second harmonic generators, where two of the three waves involved are degenerate.Finally, coherent combining has been experimentally performed in the non-degenerate case of mid-infrared difference frequency generators.In both experimental demonstrations, an excellent beam combination efficiency has been achieved.

Allier de fortes puissances moyennes et crêtes donne accès à un champ applicatif très large pour un système laser ultrarapide. Une technique qui s’est avérée capable de satisfaire ces exigences est la combinaison cohérente de faisceaux (CBC). Elle permet de séparer spatialement les faisceaux avant l’amplification pour les recombiner ensuite d’une manière cohérente en un unique faisceau. Afin d’obtenir une recombinaison efficace, les propriétés spatiales et spectrales de tous les faisceaux doivent être parfaitement en accord.Pour des applications comme l’accélération de particules, le recours à plusieurs milliers de fibres doit être envisagé. Il est donc nécessaire d’étudier des architectures CBC fortement évolutives en termes de canaux amplificateurs.Le projet XCAN vise à une première démonstration d’un tel système en réalisant la combinaison cohérente de 61 fibres amplificatrices. Afin d’étudier les défis scientifiques et techniques d’une telle architecture, une version de taille réduite comprenant sept fibres a été mise en place.La conception et la réalisation de ce prototype sont le sujet de cette thèse.Dans un premier temps, des simulations ont été effectués afin d’estimer les désaccords tolérables entre les propriétés spatiales et spectrales des différents faisceaux.Basé sur ce travail de modélisation, un système laser de combinaison cohérente de sept fibres a été ensuite assemblé et caractérisé. Les résultats obtenus sont très prometteurs et montrent que notre architecture est bien adaptée pour accueillir les 61 fibres du démonstrateur final XCAN. / Future applications of high power ultrafast laser systems require simultaneously high average and peak powers. A technique which has proved to be capable of meeting these demands is coherent beam combining (CBC).In this technique, the beam is spatially split prior to amplification, and coherently recombined in one single beam afterwards. In order to achieve an efficient recombination, the spatial and spectral properties of all beams need to be perfectly matched.For applications such as particle acceleration, the coherent combining of several thousands of fibers needs to be considered. It is thus necessary to investigate highly scalable CBC architectures.The XCAN project aims at a first demonstration of such a scalable setup by coherently combining 61 fiber amplifiers. In order to study the scientific and technical challenges of such a system, a downscaled version consisting of seven fibers has been implemented.The design and characterization of this prototype is the subject of this thesis.As a starting point, numerical simulations have been performed in order to estimate the maximum tolerable mismatches between the spatial and spectral properties of the beams.Based on this modeling work, a seven fiber CBC system has been assembled and characterized. The obtained results are very promising and imply that our setup is well suited for the accommodation of all 61 fibers of the final XCAN demonstrator.

Les lasers à fibre optique délivrant des impulsions femtoseconde sont aujourd'hui utilisés dans de nombreuses applications scientifiques ou industrielles. Pour étendre l'éventail de ces applications, l'augmentation des performances en termes de durée, énergie par impulsion, et puissance moyenne délivrée par ces sources a fait l'objet de nombreux développements. Ce travail a pour objectif d'utiliser l'idée de combinaison cohérente d'impulsions femtoseconde dans le but de poursuivre l'amélioration des caractéristiques des amplificateurs à fibre ultrabrefs selon deux axes.La première partie est consacrée à la montée en énergie en utilisant des architectures de combinaison cohérente passives, c'est à dire ne nécessitant pas de boucle de contrôle de la phase optique entre les impulsions à combiner. Ces systèmes exploitent à la fois le domaine spatial, (combinaison de faisceaux) et le domaine temporel (combinaison de répliques de l'impulsion initiale décalées dans le temps). Nous démontrons que ces techniques permettent effectivement la montée en énergie, et étudions les limites de leur utilisation liées aux non-linéarités optiques et à la saturation du gain des amplificateurs. Nous proposons également des perspectives pour outrepasser ces limites.La deuxième partie du manuscrit est dédiée à la réduction de la durée des impulsion émises par ces sources à fibre. Nous utilisons dans un premier temps la combinaison cohérente active de deux impulsions femtoseconde amplifiées ayant des contenus spectraux différents et décalés. Ainsi, une impulsion plus courte que chacune des impulsions individuelles est synthétisée. Une autre approche consistant à sculpter le contenu spectral de l'impulsion à amplifier afin de compenser le profil de gain de l'amplificateur est également étudiée. Enfin, nous appliquons les architectures de combinaison cohérente passive étudiées dans la première partie à des systèmes de compression temporelle non-linéaire afin d'outrepasser leurs limites en énergie. / Optical fiber-based ultrafast laser sources are nowadays used in numerous scientific and industrial applications. To extend further the number of possible applications, it is essential to improve their performances in terms of pulsewidth, energy per pulse, and average power. Extensive research work has been performed over the last years on this area. The work described in this manuscript is a contribution to these research efforts, and aims at improving ultrafast fiber laser sources by using coherent combination of several femtosecond pulses.The first part is devoted to energy scaling by using passive coherent combining architectures, that do not require a feedback control loop of the relative optical phase between pulses to be combined. This idea is used both in the space (combining beams) and time (combining replicas of the pulse at different delays) domains. We demonstrate that these techniques allow energy scaling and study their limitations related to optical nonlinearities and gain saturation in the amplifiers. We also propose ways to circumvent these limitations.In the second part, we study various ways of decreasing the output pulsewidth of fiber-amplified femtosecond sources. First, we implement an active coherent combining system that performs combination of two amplified femtosecond pulses with different, shifted spectral content. This allows the synthesis of a pulse that is shorter than any of the individual pulses to be combined. Another studied approach consists in tailoring the spectrum of input pulses to precompensate the spectral gain shape of the amplifier. Finally, by using passive combining architectures described in the first part, we demonstrate energy scaling of temporal nonlinear compression setups.

La capacité des sources laser à concentrer une quantité d’énergie énorme intéresse beaucoup le secteur industriel pour l’usinage et la structuration de la matière. Il faut pour cela rassembler une forte puissance optique sur une surface infime: on parle alors de luminance. La combinaison cohérente permet de répondre à la problématique de l’augmentation de la luminance d’un système laser. Dans le cadre du projet européen BRIDLE, ces travaux sont focalisés sur la combinaison cohérente de lasers à semi-conducteur. Ce type de combinaison nécessite un accord de phase stable entre les différents émetteurs. Plusieurs techniques permettent cette mise en phase; nous étudions en particulier les techniques d’amplification en parallèle ainsi que l’utilisation d’une cavité externe commune. L’originalité se situe dans le développement d’une architecture nouvelle, pensée pour optimiser l’extraction de puissance. La technique consiste à utiliser une cavité étendue commune aux émetteurs à combiner pour leur mise en phase, placée sur leur face arrière. Tout en fournissant un fort retour optique arrière nécessaire à la mise en phase, l’extraction de puissance est maximisée sur la face avant où les faisceaux sont par la suite combinés extracavité. Ce document démontre la bonne adéquation de cette architecture avec les meilleures diodes laser en termes de luminance : les émetteurs à section évasée. L’étude est étendue à une barrette de diodes par l’utilisation d’éléments diffractifs optique permettant la séparation et la combinaison des faisceaux. / Scaling up the brightness of laser diodes is a major research objective in the laser community. The coherent beam of several emitters is the most efficient technique to increase the brightness by constructive interference. An efficient combination can only be achieved in an arrangement that forces the required phase relation between the emitters. Different approaches are investigated: either active phase-locking of amplifiers seeded by a single-frequency laser split into N beams and amplified in parallel, or passive selforganization of emitters in a common laser cavity. We investigate a new coherent combining architecture using a common extended cavity on the back side of diode lasers for phase locking. As a result, the efficiency of the phase-locked laser cavity is increased as compared to standard front-side configurations. Moreover, such an extended cavity placed on the rear-side provides the strong optical feedback required for phase-locked operation. This configuration is demonstrated with high-brightness tapered devices, highlighting the capability of such setup for high power operation. This architecture is then extended to diode laser arrays by the use of diffractive optical elements.

Recent data indicate that less than 50% of American secondary students are able to demonstrate grade-level proficiency in reading, writing, and science (National Center for Educational Statistics [NCES], 2007, 2011, 2012a, 2012b). Secondary students* are expected to develop advanced literacy skills, especially in writing, in order to be ready for college and careers. Students are expected to develop these advanced literacy skills, within all academic subjects. In other words, they are expected to develop disciplinary literacy skills. The statistics are alarming overall, but they are particularly alarming in the area of science. Students need strong literacy skills, including written expression, to be prepared for employment opportunities in science fields, which currently are being filled by graduates of other industrialized nations, who have a more advanced skill set. This loss of occupational opportunity poses a threat for the U.S. to remain globally competitive in science innovation and advancement, which ultimately secures economic prosperity. Despite these staggering concerns, there is little research conducted to evaluate effective instructional methods to develop complex writing skills in academic disciplines such as science. To address this critical issue, the present study examined the effects of a metalinguistic approach to the writing intervention of sentence combining with eighth-grade students who struggle with literacy. The researcher conducted the study in a typical science classroom in an urban American school setting. The focus of the intervention was to increase students* metalinguistic awareness of science text, to improve written sentence complexity in science, as well as the written expression and determination of comparison and contrast of science content. The study employed a quasi-experimental design. The participants consisted of an experimental group (two classes) who received the treatment during typical science instruction and a comparison group (three classes) who did not receive treatment, but participated in their typical science instruction. There were four participating teachers and 84 participating students. The researcher conducted the study over a period of seven weeks within regularly scheduled science classes. Twenty intervention sessions were conducted for a length of 20 minutes each, totaling 400 minutes or 6.6 hours. Hierarchical repeated measures ANOVA and hierarchical repeated measures MANOVA analyses revealed that the experimental group performed significantly better than the comparison group on their ability to determine similarities and differences (compare and contrast) related to science content, with a medium effect. The experimental group achieved a slightly higher marginal mean over the comparison group on their ability to combine sentences, with a small effect. Multiple statistical analyses revealed a trend of higher marginal means in favor of the experimental group over the comparison group on several measures of written sentence complexity on both the science compare and contrast writing prompt (small-medium effect) and the science expository essay (medium to large effect). One experimental class also demonstrated higher scores in their overall sentence correctness on science expository essay as compared to all the other classes. These findings suggest that sentence combining, utilizing a metalinguistic approach, may hold promise as an effective writing intervention in a content area classroom, for secondary students who struggle with literacy. Furthermore, the findings suggest that a metalinguistic approach to sentence combining can be successfully embedded within a content area class, which may result in increased concept knowledge and writing skills in that academic discipline. Implications for practice and future research directions are discussed.

Liquid crystals have been a major subject of research for the past decades. Aside from the variety of structures they can form, they exhibit a vast range of optical phenomena. Many of these phenomena found applications in technology and became an essential part of it. In this dissertation thesis we continue the line to propose a number of new applications of optical effects in liquid crystals and develop their theoretical framework. One such application is the possibility of beam combining using Orientational Stimulated Scattering in a nematic liquid crystal cell. Our numerical study of the OSS process shows that normally this possibility does not exist. However, we found that if a number of special conditions is satisfied efficient beam combining with OSS can be done. These conditions require a combination of special geometric arrangement of incident beams, their profiles, nematic material, and more. When these conditions are fulfilled, power of the beamlets can be coherently combined into a single beam, with high conversion efficiency while the shape and wave-front of the output beam are still of good quality. We also studied the dynamics of the OSS process itself and observed (in a numerical model) a number of notorious instabilities caused by effects of back-conversion iv process. Additionally, there was found a numerical solitary-wave solution associated with this back-conversion process. As a liquid crystal display application, we consider a nematic liquid crystal layer with the anisotropy axis modulated at a fixed rate in the transverse direction with respect to light propagation direction. If the layer locally constitutes a half-wave plate, then the thinscreen approximation predicts 100% -efficient diffraction of normal incident wave. If this diffracted light is blocked by an aperture only transmitting the zero-th order, the cell is in dark state. If now the periodic structure is washed out by applying voltage across the cell and light passes through the cell undiffracted, the light will pass through the aperture as well and the cell will be in its bright state. Such properties of this periodically aligned nematic layer suggest it as a candidate element in projection display cells. We studied the possibility to implement such layer through anchoring at both surfaces of the cell. It was found that each cell has a thickness threshold for which the periodic structure can exist. The anchored periodic structure cannot exist if thickness of the cell exceeds this threshold. For the case when the periodic structure exists, we found the structure distortion in comparison with the preferable ideal sinusoidal profile. To complete description of the electromechanical properties of the periodic cell, we studied its behavior at Freedericksz transition. Optical performance was successfully described with the coupled-mode theory. While influence of director distortion is shown to be negligibly small, the walk-off effects appear to be larger. In summary, there are good prospects for use of this periodically v aligned cell as a pixel in projection displays but experimental study and optimization need to be performed. In the next part we discuss another modulated liquid crystal structure in which the director periodically swings in the direction of light propagation. The main characteristic of such structure is the presence of bandgap. Cholesteric liquid crystals are known to possess bandgap for one of two circular polarizations of light. However, unlike the cholesterics the bandgap of the proposed structure is independent of polarization of normally incident light. This means that no preparation of light is needed in order for the structure to work in, for example, liquid crystal displays. The polarization universality comes at the cost of bandgap size, whose maximum possible value ∆ωPTN compared to that of cholesterics ∆ωCh is approximately twice smaller: ∆ωPTN ≈ 0.58∆ωCh if modulation profile is sinusoidal, and ∆ωPTN ≈ 0.64∆ωCh if it is rectangular. This structure has not yet been experimentally demonstrated, and we discuss possible ways to make it.

The main objective of this thesis is to explore the use of volume holographic elements recorded in photo-thermo-refractive (PTR) glass for power scaling of narrow linewidth diffraction-limited fiber lasers to harness high average power and high brightness beams. Single fiber lasers enable kW level output powers limited by optical damage, thermal effects and non-linear effects. Output powers can be further scaled using large mode area fibers, however, at the cost of beam quality and instabilities due to the presence of higher order modes. The mechanisms limiting the performance of narrow-linewidth large mode area fiber lasers are investigated and solutions using intra-cavity volume Bragg gratings (VBG) proposed. Selfpulsations-free, completely continuous-wave operation of a VBG-stabilized unidirectional fiber ring laser is demonstrated with quasi single-frequency ( < 7.5 MHz) output. A method for transverse mode selection in multimode fiber lasers to reduce higher order mode content and stabilize the output beam profile is developed using angular selectivity of reflecting VBGs. By placing the VBG output coupler in a convergent beam, stabilization of the far-field beam profile of a 20 μm core large mode area fiber laser is demonstrated. Beam combining techniques are essential to power scale beyond the limitations of single laser sources. Several beam combining techniques relevant to fiber lasers were compared in this study and found to be lacking in one or more of the following aspects: the coherence of the individual sources is compromised, the far-field beam quality is highly degraded with significant power in iv side lobes, spectrally broad and unstable, and uncertainty over scaling to larger arrays and higher power. Keeping in mind the key requirements of coherence, good far-field beam quality, narrow and stable spectra, and scalability in both array size and power, a new passive coherent beam combining technique using multiplexed volume Bragg gratings (M-VBGs) is proposed. In order to understand the mechanism of radiation exchange between multiple beams via these complex holographic optical elements, the spectral and beam splitting properties a 2nd order reflecting M-VBG recorded in PTR glass is experimentally investigated using a tunable single frequency seed laser. Two single-mode Yb-doped fiber lasers are then coherently combined using reflecting M-VBGs in both linear and unidirectional-ring resonators with > 90% combining efficiency and diffraction-limited beam quality. It is demonstrated that the combining bandwidth can be controlled in the range of 100s of pm to a few pm by angular detuning of the M-VBG. Very narrow-linewidth ( < 210 MHz) operation in a linear cavity and possibility of singlefrequency operation in a unidirectional ring cavity of the coherently combined system is demonstrated using this technique. It is theoretically derived and experimentally demonstrated that high combining efficiency can be achieved even by multiplexing low-efficiency VBGs, with the required diffraction efficiency of individual VBGs decreasing as array size increases. Scaling of passive coherent beam combining to four fiber lasers is demonstrated using a 4th order transmitting M-VBG. Power scaling of this technique to 10 W level combined powers with 88% combining efficiency is demonstrated by passively combining two large mode area fiber lasers using a 2nd order reflecting M-VBG in a unidirectional ring resonator. High energy compact single-frequency sources are highly desired for several applications – one of which is as a seed for high power fiber amplifiers. Towards achieving the goal of a monolithic solid-state laser, a new gain medium having both photosensitive and luminescence properties is investigated – rare-earth doped PTR glass. First lasing is demonstrated in this new gain element in a VBG-stabilized external cavity.

High power thulium fiber lasers are useful for a number of applications in both continuous-wave and pulsed operating regimes. The use of thulium as a dopant has recently gained interest due to its large bandwidth, possibility of high efficiency, possibility of high power and long wavelength ~1.8 – 2.1 μm. The longer emission wavelength of Tm-doped fiber lasers compared to Yb- and/or Er-doped fiber lasers creates the possibility for higher peak power operation due to the larger nonlinear thresholds and reduced nonlinear phase accumulation. One primary interest in Tm-doped fiber lasers has been to scale to high average powers; however, the thermal and mechanical constraints of the fiber limit the average power out of a single-fiber aperture. One method to overcome the constraints of a single laser aperture is to spectrally combine the output from multiple lasers operating with different wavelengths into a single beam. In this thesis, results will be presented on the development of three polarized 100 W level laser systems that were wavelength stabilized for SBC. In addition to the development of the laser channels, the beams were combined using bandpass filters to achieve a single near diffraction-limited output. Concurrently, with the development of high average power systems there is an increasing interest in femotosecond pulse generation and amplification using Tm- doped fiber lasers. High peak power sources operating near 2 µm have the potential to be efficient pump sources to generate mid-infrared light through supercontinuum generation or optical parametric oscillators. This thesis focuses on the development of a laser system utilizing chirped pulse amplification (CPA) to achieve record level energies and peak powers for ultrashort pulses in Tm-doped fiber. iv A mode-locked oscillator was built to generate femtosecond pulses operating with pJ energy. Pulses generated in the mode-locked oscillator were limited to low energies and contained spectral modulation due to the mode-locking mechanism, therefore, a Raman-soliton self-frequency shift (Raman-SSFS) amplifier was built to amplify pulses, decrease the pulse duration, and spectrally clean pulses. These pulses were amplified using chirped pulse amplification (CPA) in which, limiting factors for amplification were examined and a high peak power system was built. The primary limiting factors of CPA in fibers include the nonlinear phase accumulation, primarily through self-phase modulation (SPM), and gain narrowing. Gain narrowing was examined by temporally stretching pulses in a highly nonlinear fiber that both stretched the pulse duration and broadened the spectrum. A high peak power CPA system amplified pulses to 1 µJ energy with 300 fs compressed pulses, corresponding to a peak power >3 MW. High peak power pulses were coupled into highly nonlinear fibers to generate supercontinuum

Research recently conducted at the University of Central Florida involving crashes on Interstate-4 in Orlando, Florida has led to the creation of new statistical and neural networks models that are 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 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 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 combining two ITS strategies (Ramp Metering and Variable Speed Limits strategies) for reducing the crash risk (both rear-end and lane-change crash risks) along the I-4 freeway. Following this aspect, a 36.25-mile section of I-4 running though Orlando, FL was simulated using the PARAMICS micro-simulation program. Gayah (2006) used the same network to examine the potential benefits of two ITS strategies separately (Route Diversion and Ramp Metering) for reducing the crash risk along the freeway by changing traffic flow parameters. Cunningham (2007) also used the same network to examine the potential benefits of implementing Variable Speed Limits strategy for reducing the crash risk along the freeway. Since the same network is used, the calibration and validation procedures used in this study are the same as these previous two studies. This study simulates three volume loading scenarios on the I-4 freeway. These are 60, 80 and 90 percent loading scenarios. From the final experimental design for the 60 % loading, it was concluded that implementing VSL strategy only was more beneficial to the network than either implementing Ramp Metering everywhere (through the whole network) in conjunction with VSL everywhere or implementing Ramp Metering downtown (in downtown areas only) in conjunction with VSL everywhere. This was concluded from the comparison of the results of this study with the results from Cunningham (2007). However, either implementing Ramp Metering everywhere or downtown in conjunction with VSL everywhere showed safety benefits across the simulated network as well as a reduction in the total travel time. The best case for implementing Ramp Metering everywhere in conjunction with VSL everywhere was using a homogeneous speed zone threshold of 2.5 mph, a speed change distance of half speed zone and a speed change time of 5 minutes in conjunction with a 60 seconds cycle length for the Zone algorithm, a critical occupancy of 0.17 and a 30 seconds cycle length for the ALINEA algorithm. And the best case for implementing Ramp Metering downtown in conjunction with VSL everywhere was using a homogeneous speed zone threshold of 2.5 mph, a speed change distance of half speed zone and a speed change time of 10 minutes in conjunction with a 60 seconds cycle length for the Zone algorithm, a critical occupancy of 0.17 and a 30 seconds cycle length for the ALINEA algorithm. For the 80 % loading, it was concluded that either implementing Ramp Metering everywhere in conjunction with VSL everywhere or implementing Ramp Metering downtown in conjunction with VSL everywhere was more beneficial to the network than implementing VSL strategy only. This was also concluded from the comparison of the results of this study with the results from Cunningham (2007). Moreover, it was concluded that implementing Ramp Metering everywhere in conjunction with VSL everywhere showed higher safety benefits across the simulated network than implementing Ramp Metering downtown in conjunction with VSL everywhere. Also, both of them increased the total travel time a bit, but this was deemed acceptable. Additionally, both of them had successive fluctuations and variations in the average lane-change crash risk vs. time step. The best case for implementing Ramp Metering everywhere in conjunction with VSL everywhere was using a homogeneous speed zone threshold of 5 mph, a speed change distance of half speed zone and a speed change time of 30 minutes in conjunction with a 60 seconds cycle length for the Zone algorithm, a critical occupancy of 0.17 and a 30 seconds cycle length for the ALINEA algorithm. And the best case for implementing Ramp Metering downtown in conjunction with VSL everywhere was using a homogeneous speed zone threshold of 5 mph, a speed change distance of half speed zone and a speed change time of 30 minutes in conjunction with a 60 seconds cycle length for the Zone algorithm, a critical occupancy of 0.17 and a 30 seconds cycle length for the ALINEA algorithm. Searching for the best way to implement both Ramp Metering and VSL strategies in conjunction with each other, an indepth investigation was conducted in order to remove the fluctuations and variations in the crash risk with time step (through the entire simulation period). The entire simulation period is 3 hours, and each time step is 5 minutes, so there are 36 time steps representing the entire simulation period. This indepth investigation led to the idea of not implementing VSL at consecutive zones (using either a gap of one zone or more). Then this idea was applied for the best case of implementing Ramp Metering and VSL everywhere at the 80 % loading, and the successive fluctuations and variations in the crash risk with time step were removed. Moreover, much better safety benefits were found. So, this confirms that this idea was very beneficial to the network. For the 90 % loading, it was concluded that implementing Ramp Metering strategy only (Zone algorithm in downtown areas, and ALINEA algorithm in non downtown areas) was more beneficial to the network than implementing Ramp Metering everywhere in conjunction with VSL everywhere. This was concluded from the comparison of the results of this study with the results from Gayah (2006). However, implementing Ramp Metering everywhere in conjunction with VSL everywhere showed safety benefits across the simulated network as well as a reduction in the total travel time. The best case was using a homogeneous speed zone threshold of 2.5 mph, a speed change distance of the entire speed zone and a speed change time of 20 minutes in conjunction with a 60 seconds cycle length for the Zone algorithm, a critical occupancy of 0.17 and a 30 seconds cycle length for the ALINEA algorithm. In summary, Ramp Metering was more beneficial at congested situations, while Variable Speed Limits were more beneficial at free-flow conditions. At conditions approaching congestion, the combination of Ramp Metering and Variable Speed Limits produced the best benefits. These results illustrate the significant potential of ITS strategies to improve the safety and efficiency of urban freeways.