Master of Science / Department of Civil Engineering / Robert J. Peterman / This thesis details findings from testing done to determine bond and time-dependent characteristics of two lightweight concrete mixes. The lightweight mixes were tested to possibly provide a more cost-effective solution to replacing some of Kansas’ older bridges. Testing included use of a conventional lightweight mix and a self-consolidating lightweight mix. Sixteen Inverted T-beams were cast at a prestress plant to determine prestress losses that had occurred in the two lightweight mixes. These losses were compared to ACI, PCI, and AASHTO code equations. Creep and shrinkage prisms were also cast and measured to accurately determine creep and shrinkage variables for the two lightweight mixes. Twelve flexural beams were also cast at the prestress plant and tested at Kansas State University’s Civil Engineering Structures Laboratory to experimentally test development lengths of the lightweight mixes and to compare results with ACI code equations. This study found compressive strengths of the lightweight concrete mixes varied greatly from laboratory testing. Low concrete strengths caused the prestress losses to be greater than the predicted code values. Flexure beam testing showed several of the beams were subject to strand slip, causing a sudden violent failure.

Master of Science / Department of Civil Engineering / Hayder A. Rasheed / There is a high need to repair or replace many bridges in the state of Kansas. 23% of the bridges in Kansas are labeled structurally deficient or functionally obsolete. A majority of these bridges serve rural areas and are damaged due to overloading during harvest season. A state-of-the-art method of performing structural health monitoring on these bridges followed by an effective method of strengthening and repair was researched and presented in this thesis. The first phase of this research involved researching multiple devices to be used for state-of-the-art health monitoring. After deciding on an appropriate system, multiple tests were performed to determine the systems performance compared against conventional systems. The system was tested on a laboratory scale pre-stressed concrete T-beam. The system was tested on its ability to effectively record and transmit acceleration data. If this system were to be implemented on an actual bridge, KDOT could make a decision to repair or strengthen the bridge based on the results. The next phase of the research was to determine an effective strengthening procedure using carbon fiber reinforced polymer (CFRP). Reinforced concrete beam specimens were cast and tested in the lab. The specimens consisted of rectangular and T-shaped cross-sections to create different failure modes when tested in bending. The primary issue when strengthening with CFRP is the issue of early separation failure when using CFRP in the longitudinal direction only. In an effort to prove this, the specimens were strengthened with five layers of CFRP and tested in four-point bending until failure. In an effort to prevent early separation failure, CFRP “U-wraps” were applied to provide shear resistance and additional anchorage for the flexural CFRP. The beams were then tested in flexure until failure by FRP rupture or concrete crushing followed by FRP rupture. The test results indicate that the U-wraps allowed the FRP to reach full capacity and fail in FRP rupture. The use of CFRP provided a strength increase of about 220% over the control beam specimens while significantly reducing the ultimate deflection.

Master of Science / Department of Civil Engineering / Sunanda Dissanayake / In the United States, the mileage of unpaved roads is about 1.6 million miles. Total length of unpaved roads in Kansas is about 98,000 miles, of which about 78,000 miles are gravel roads. Most of the gravel roads are not posted with speed limit signs but regulated with a 55 mph blanket speed limit established by the Kansas Statutes. Surface conditions of gravel roads are very likely to change with time, space, and quality of maintenance work, making it even more necessary to have proper control of speeds on gravel roads. Normally used speed regulations and rules for freeways or other types of paved roadways might not be appropriate for gravel roads, especially for those local gravel roads which usually carry very low traffic in rural areas. Based on an extensive literature search, there was no specific rule or references to provide guidelines on how speed limits on gravel roads could be set. Therefore, an effort was made in this study to evaluate the effects of currently posted lower speed limits in some counties in Kansas, based on traffic characteristics and safety on gravel roads, with the intention of providing proper guidelines for setting speed limits on gravel roads in Kansas. In order to study traffic characteristics on gravel roads, field speed studies were conducted with automatic traffic counters on more than forty gravel road sections in seven counties in Kansas. Important speed measures, such as 85th-percentile speed and mean speed, were obtained from the raw data. A group of other related road characteristics were also recorded at the time of field data collection. Crash data on gravel roads were extracted from the Kansas Accident Recording System (KARS) database. Speed analysis on a number of gravel roads where the statutory imposed, unposted speed limit of 55 mph was utilized indicated that they are functioning at a reasonably acceptable level in terms of actual speeds. In order to evaluate whether there were differences in traffic speeds between two counties or groups which have different speed limit settings on gravel roads, t-test was used. The analysis found that there was no significant difference between the mean speeds in two counties, one of which has 35 mph posted speed limit on gravel roads while the other does not post any speed limits. Moreover, the mean speed on the sections with 35 mph posted speed was a little higher than that on gravel roads without any speed limits. Linear models to predict 85th-percentile speed and mean speed on gravel roads were developed based on speed data. Both models indicated that traffic speeds are not significantly affected by the speed limit, but are related with 90% confidence to road width, surface classification and percentage of large vehicles in traffic. Chi-square tests were conducted with the crash data, and the results indicated that the posted 35 mph speed limit on gravel roads had not resulted in either smaller total number of crashes or decreased proportion of severe crashes, compared to gravel roads where no speed limits were posted. Logistic regression models were also developed on four levels of crash severity, which indicated that gravel roads with higher speed limits are likely to experience higher probability of having injury crashes. Two mail-back surveys were also conducted to gather the opinions of county engineers and road users on the subject of suitable speed limits on gravel roads. The majority of county engineers believed that blanket speed limit should be used for gravel roads and does not need to be posted. Three restrictions: changeful road conditions, unpractical law enforcement, and limited funds, are basic reasons why they do not think that gravel roads should be posted. Besides that, a few respondents said 55 mph is too high for gravel roads and needs to be lowered. Majority of the road users suggested that all gravel roads be posted with lower speed limit signs. However, they were more concerned about law enforcement since they believe that posted speeds won't bring any benefits if no law enforcement patrol gravel roads. Based on all aspects looked into in this study, it does not appear that reducing the speed limits and posting it with signs, is going to improve either traffic operational or safety characteristics on gravel roads in Kansas, and therefore is not recommended for new situations.

Master of Science / Department of Civil Engineering / David R. Steward / A one-dimensional model was developed to study the flow of groundwater in the sloping Ogallala Aquifer at a steady state during predevelopment condition. The sloping base was approximated using a stepping base model. GIS applications were applied during data collection and preparation, and later during interpretation of model results. Analytical and numerical methods were employed in the development of this model which was used to try to understand long-term water balance in the study region. The conservation of mass was achieved by balancing groundwater input, output, and storage; this led to understanding the interactions of groundwater and surface water in the predevelopment conditions. The study resulted in identification of where natural discharge from groundwater to surface water occurred, and the quantity of these flows was obtained. The Ogallala Aquifer is thick in the south western part of Kansas, this region had an average saturated thickness of 100m during predevelopment conditions. The model found that groundwater flowed at a discharge per width of approximately 17 m[superscript]2/d in this region. The aquifer thickness tends to gradually decrease from west to east and from south to north. The northern part had an average saturated thickness of 40m during predevelopment conditions; the model found that groundwater flowed at a discharge per width of approximately 3 m[superscript]2/d in this region. It was also found that groundwater leaves the Ogallala Aquifer on the eastern side with discharge per width between 0-3 m[superscript]2/d. The discharge from groundwater to surface water was summed over contributing areas to river basins. The discharge to streams necessary to satisfy long-term conservation of mass computed by the model showed that Cimarron River has total baseflow of about 5.5 m[superscript]3/s; this was found to be almost 100% of the total streamflow recorded during predevelopment conditions. The Arkansas River was found to have total baseflow of about 0.97 m[superscript]3/s, which is approximately 14.3% of the total streamflow recorded during predevelopment conditions. The Smoky Hill River was found to have total baseflow of about 1.7 m[superscript]3/s, which is approximately 73.9% of the total streamflow recorded during predevelopment conditions. The Solomon River was found to have total baseflow of about 0.95 m[superscript]3/s, which is approximately 41.1% of the total streamflow recorded during predevelopment conditions. The Saline River was found to have total baseflow of about 0.25 m[superscript]3/s, which is approximately 62.5% of the total streamflow recorded during predevelopment conditions. The Republican and Pawnee River was found to have total baseflow of about 0.38 m[superscript]3/s and 0.22 m[superscript]3/s, which is approximately 18.5% and 12.6% of the total streamflow in the predevelopment conditions respectively. The model was found to be always within -16 to +12 meters between observed values and the model results, with an average value of 0.15m and a root mean square error of 1.98m. Results from this study can be used to advance this study to the next level by making a transient model that could be used as a predictive tool for groundwater response to water use in the study region.

Master of Science / Department of Civil Engineering / Sunanda Dissanayake / One-ninth of all traffic fatalities in the United States (U.S.) in the past five years have involved large trucks, although large trucks contributed to only 3% of registered vehicles and 7% of vehicle miles traveled. This crash overrepresentation indicates that truck crashes in general tend to be more severe than other crashes, though they constitute a smaller portion of vehicles on the road. To study this issue, fatal crash data from the Fatality Analysis Reporting System (FARS) was used to analyze characteristics and factors contributing to truck-involved crashes. Driver, vehicle, and crash-related contributory causes were identified, and as an extension, the likelihood of occurrence of these contributory causes in truck-involved crashes (with respect to non-truck crashes) was evaluated using the Bayesian Statistical approach. Likelihood ratios indicated that factors such as stopped or unattended vehicles and improper following have greater probability of occurrence in truck crashes than in non-truck crashes. Also, Multinomial Logistic Regression was used to model the type of fatal crash (truck vs. non-truck) to compare the relative significance of various factors in truck and non-truck crashes. Factors such as cellular phone usage, failure to yield right of way, inattentiveness, and failure to obey traffic rules also have a greater probability in fatal truck crashes. Among several other factors, inadequate warning signs and poor shoulder conditions were also found to have greater predominance in contributing to truck crashes than non-truck crashes. By addressing these factors through the implementation of appropriate remedial measures, the truck safety experience could be improved, which would eventually help in improving overall safety of the transportation system.

Master of Science / Department of Civil Engineering / Sunanda Dissanayake / In the United States, approximately 1,100 people die and 40,000 people are injured annually as a result of motor vehicle crashes in work zones. These numbers may be a result of interruption to regular traffic flow caused by closed traffic lanes, poor traffic management within work zones, general misunderstanding of problems associated with work zones, or improper usage of traffic control devices. In regard to safety of work zones, this study was conducted to identify characteristics and risk factors associated with work zone crashes in Iowa, Kansas, Missouri, Nebraska and Wisconsin, states currently included in the Smart Work Zone Deployment Initiative (SWZDI) region. The study was conducted in two stages. In the first stage, characteristics and contributory causes related to work zone crashes such as environmental conditions, vehicles, crashes, drivers, and roadways were analyzed for the five states for the period 2002-2006. An analysis of percentage-wise distributions was carried out for each variable based on different conditions. Results showed that most of the work zone crashes occurred under clear environmental conditions as during daylight, no adverse weather, etc. Multiple-vehicle crashes were more predominant than single-vehicle crashes in work zone crashes. Primary driver-contributing factors of work zone crashes were inattentive driving, following too close for conditions, failure to yield right of way, driving too fast for conditions, and exceeding posted speed limits within work zones. A test of independency was performed to find the relation between crash severity and other work zone variables for the combined states. In the second stage, a statistical model was developed to identify risk factors associated with work zone crashes. In order to predict injury severity of work zone crashes, an ordered probit model analysis was carried out using the Iowa work zone crash database. According to findings of the severity model, work zone crashes involving trucks, light duty vehicles, vehicles following too close, sideswipe collisions of same-direction vehicles, nondeployment of airbags, and driver age are some of the contributing factors towards more severe crashes.

Master of Science / Department of Civil Engineering / Sunanda Dissanayake / Speed management has been a challenge, particularly in places where high-speed highways pass through. Due to high rate of fatalities and low budgets available, it is therefore necessary to identify low-cost effective approaches in reducing speeds. Optical Speed Bar (OSB) treatment is one such technique. This research makes an attempt to evaluate the effectiveness of OSBs in reducing approach speeds on two-lane, rural highways approaching small communities. Speed data were collected and analyzed “before” and “after” periods at five sites. Effectiveness of OSBs was evaluated using changes in mean and 85th percentile speeds under different categories by considering all vehicles, vehicle classification (two axles vs. more than two axles), day of the week (weekdays vs. weekends), and time of day (daytime vs. nighttime), as well as proportions exceeding posted speed limit, using t-test mean speeds, F-test for analysis of variance, and Z-test for proportions of vehicles exceeding posted speed limit between “before” and “after” datasets. Even though motorists were found to slow down on the approaches, in response to speed zones, speeding was noted. “Before” speed data indicated higher speeds than desired at the sites. The 85th percentile speeds were between 50 and 63 mph while the posted speed limits on the approaches were 45 mph at four sites, and the 85th percentile speed was about 42 mph at one site with an approach posted speed limit of 30 mph. The “before” degrees of noncompliance were up to 90 % of free-flowing vehicles at the sites. Speed data analysis showed significant reductions in speeds at ends of OSBs at four test sites. Mean and 85th percentile speeds and standard deviations were found reduced in the after periods. Percent reductions in mean speeds were between 1.2 and 8.2 %, with 85th percentile reductions between 3.2 and 8.9 %. At one site, no notable change in mean and 85th percentile speeds occurred at the end of OSBs, but significant increases in standard deviations were noted. Speed reductions were higher for two-axle vehicles, during the daytime and on weekdays with few exceptions. Results of the study showed, as other previous studies did, OSBs may have some minor effects on vehicle speeds. The study provides an indication that it may be possible to create safety improvements as result of using OSBs on the approach to a rural community. However, magnitude of speed reductions was generally small, though the reductions were statistically significant at the 95% confidence level. Because of the non-consistence of the magnitude of speed reductions at the test sites, no conclusion can be drawn as to how much OSB treatment reduced speeds. These results were based on “after” periods up to five months. Therefore, further study would be required to determine whether these safety improvements are sustained over an even longer time period. Even though minor speed reductions occurred, speeds observed at the sites were still higher than the posted speed limits, indicating OSBs were not effective enough in providing the desired speed limit compliance. Additional studies would be helpful to identify combinations of countermeasures, for instance OSBs and other techniques, effective in providing speed limit compliance.

Master of Science / Department of Architectural Engineering and Construction Science / Sutton F. Stephens / Seismic hazards are a primary concern in some of the most populous regions in the United States. Performance-based seismic design has brought about new technology advances and introduced an innovative approach towards constructing seismic resistant buildings. Base isolation and structural damping systems are becoming increasingly utilized methods of advanced seismic resistance. This relatively new design approach presents various issues that must be addressed throughout the design and construction processes. A brief background on the origin, dynamics, and hazards of earthquakes and a discussion on designs of traditional, fixedbased structures is presented in this report. A description for selected types of new advanced seismic restraint systems, with an emphasis on base isolation, is also provided. Examples of current applications of buildings equipped with base isolation are presented. This report concludes with a review of the fundamental design methodology for structural base isolation along with additional requirements not addressed by the current building codes.

Doctor of Philosophy / Department of Civil Engineering / Yacoub M. Najjar / This study explored the potential use of ANNs for profiling and characterization of various environmental sites. A static ANN with back-propagation algorithm was used to model the environmental containment at a hypothetical data-rich contaminated site. The performance of the ANN profiling model was then compared with eight known profiling methods. The comparison showed that the ANN-based models proved to yield the lowest error values in the 2-D and 3-D comparison cases. The ANN-based profiling models also produced the best contaminant distribution contour maps when compared to the actual maps. Along with the fact that ANN is the only profiling methodology that allows for efficient 3-D profiling, this study clearly demonstrates that ANN-based methodology, when properly used, has the potential to provide the most accurate predictions and site profiling contour maps for a contaminated site. ANN with a back-propagation learning algorithm was utilized in the site characterization of contaminants at the Kansas City landfill. The use of ANN profiling models made it possible to obtain reliable predictions about the location and concentration of lead and copper contamination at the associated Kansas City landfill site. The resulting profiles can be used to determine additional sampling locations, if needed, for both groundwater and soil in any contaminated zones. Back-propagation networks were also used to characterize the MMR Demo 1 site. The purpose of the developed ANN models was to predict the concentrations of perchlorate at the MMR from appropriate input parameters. To determine the most-appropriate input parameters for this model, three different cases were investigated using nine potential input parameters. The ANN modeling used in this case demonstrates the neural network’s ability to accurately predict perchlorate contamination using multiple variables. When comparing the trends observed using the ANN-generated data and the actual trends identified in the MMR 2006 System Performance Monitoring Report, both agree that perchlorate levels are decreasing due to the use of the Extraction, Treatment, and Recharge (ETR) systems. This research demonstrates the advantages of ANN site characterization modeling in contrast with traditional modeling schemes. Accordingly, characterization task-related uncertainties of site contaminations were curtailed by the use of ANN-based models.

Master of Science / Department of Civil Engineering / Sunanda Dissanayake / The older population (>65 years) numbered 36.8 million in the United States in 2005. By 2030, the number is estimated to be 71.5 million, almost twice as many. An increase in the older population means an increase in older drivers as well. As a result of the natural aging process, the possibility of older drivers being involved in crashes and sustaining severe injuries increases, according to past findings. The objective of this study was to identify characteristics of older drivers involved in crashes in Kansas as well as associated safety issues, which can be used to suggest potential countermeasures for improving safety. A detailed characteristic analysis was carried out for older, middle-aged, and younger drivers involved in crashes, using crash data obtained from the Kansas Department of Transportation, and comparisons were made among the groups. However, the characteristic analysis had no basis with regard to injury severity and hence, univariate statistical analysis was carried out to highlight these severities. In addition, a survey was conducted focusing on identifying older-driver behaviors, potential problems, and level of exposure to various conditions. From the severity analysis, it was found that injury severity of older drivers in crashes occurring on rural roads was significantly higher compared to those on urban roads. Therefore, a detailed analysis was carried out using the decomposition method and ordered probit modeling to identify contributing factors leading to the situation. According to the findings, the number of older male drivers involved in crashes was higher compared to older female drivers, even though older driver licensees’ data indicate the opposite. Most of the older-driver-involved crashes occurred under good environmental conditions and at intersections. A majority of older drivers had difficulties associated with left- turn maneuvering and preferred to avoid high-traffic roads and other demanding conditions. Exposure to inclement weather conditions and difficulties associated with merging, diverging, and identifying speeds and distance of oncoming traffic have lead to higher crash propensity. In rural areas, driving in the wrong direction, failing to comply with traffic signs and signals, and speeding were identified as frequent contributing factors in high severe crashes.