Grinding is a complex material removal process with a large number of parameters influencing each other. In the process, the grinding wheel surface contacts the workpiece at high speed and under high pressure. The complexity of the process lies in the multiple microscopic interaction modes in the wheel-workpiece contact zone, including cutting, plowing, sliding, chip/workpiece friction, chip/bond friction, and bond/workpiece friction. Any subtle changes of the microscopic modes could result in a dramatic variation in the process. To capture the minute microscopic changes in the process and acquire better understanding of the mechanism, a physics-based model is necessary to quantify the microscopic interactions, through which the process output can be correlated with the input parameters. In the dissertation, the grinding process is regarded as an integration of all microscopic interactions, and a methodology is established for the physics based modeling. To determine the engagement condition for all micro-modes quantitatively, a virtual grinding wheel model is developed based on wheel fabrication procedure analysis and a kinematics simulation is conducted according to the operational parameters of the grinding process. A Finite Element Analysis (FEA) is carried out to study the single grain cutting under different conditions to characterize and quantify the grain-workpiece interface. Given the engagement condition on each individual grain with the workpiece from the physics-based simulation, the force, chip generation, and material plastic flow can be determined through the simulation results. Therefore, the microscopic output on each discrete point in the wheel-workpiece contact zone can be derived, and the grinding process technical output is the integrated product of all microscopic interaction output. From the perspective of process prediction and optimization, the simulation can provide the output value including the tangential force and surface texture. In terms of the microscopic analysis for mechanism study, the simulation is able to estimate the number of cutting and plowing grains, cutting and plowing force, probability of loading occurrence, which can be used as evidence for process diagnosis and improvement. A series of experiments are carried out to verify the simulation results. The simulation results are consistent with the experimental results in terms of the tangential force and surface roughness Ra for dry grinding of hardened D2 steel. The methodology enables the description of the 'inside story' in grinding processes from a microscopic point of view, which also helps explain and predict the time dependent behavior in grinding. Furthermore, the process model can be used for grinding force (or power) estimation for multiple-stage grinding cycles which includes rough, semi-finish, finish, and spark out. Therefore, the grinding process design can be carried out proactively while eliminating 'trial and error'. In addition, the grinding wheel model itself can be used to guide the recipe development and optimization of grinding wheels. While the single grain micro-cutting model can be used to study the mechanism of single grit cutting under various complex conditions, it can also be used to derive the optimal parameters for specific grains or process conditions.

In the United States, influenza causes approximately 36,000 deaths and over 200,000 hospitalizations each year with elderly most often affected. Clostridium difficile infection (CDI) is another major health care challenge and pressing public health issue associated with 14,000 deaths and over 335,000 hospitalizations annually. The use of antibiotics has been implicated in the development of CDI. This study's purpose was to test the relationship of seasonal influenza incidence and antiviral/antibiotic use in CDI development among hospitalized patients. Grounded in the epidemiologic wheel model of man-environment interactions, this retrospective observational study described and analyzed data from a proprietary, laboratory, and pharmacy-based system from a cohort of hospitals. The association between 147 patients with a diagnosis and/or positive test for influenza, the independent variables of delivery of antivirals/antibiotics (n = 130) during the patient's hospitalization, and the dependent variable of positive test or diagnosis of CDI (n = 17) was tested using multiple logistic regressions. The study results did not prove to be significant for the 3 research questions, suggesting no impact of antiviral use (R2 = .05, p = .336), antibiotic use (R2 = .05, p = .290), or antiviral and/or antibiotic use (R2 = .04, p = .382) on development of CDI within 60 days of discharge. However, findings indicated that recommended antiviral medication was inconsistently administered to influenza positive patients and that inappropriate prescribing patterns for antimicrobial agents coincided with seasonal influenza. Implications for positive social change include confirming the importance of antibiotic stewardship as an essential aspect of quality healthcare.

In this work, measurements and simulations were used to investigate the effects of grinding wheel topography on the geometric aspects of the grinding process. Since existing methods for measuring the grinding wheels were either not accurate enough or could only measure a small portion of a grinding wheel, a novel grinding wheel measurement system was developed. This system consists of a white light chromatic sensor, a custom designed positioning system and software. The resulting wheel scanning system was capable of measuring an entire grinding wheel with micron level accuracy. The system was used to investigate the effects of fine, medium and course dressing on grinding wheel surface topology and the resulting workpiece surface. New techniques were also developed to simulate metal removal in grinding. The simulation software consisted of a stochastic wheel model, dressing model and metal removal model. The resulting software could determine the uncut chip thickness, contact length for every cutting edge on a grinding wheel as well as the resulting surface roughness of the grinding wheel. The simulation was validated by comparing the wheel model used in the simulation to grinding wheel measurements and by comparing the simulated surface finish to the measured surface finish. There was excellent agreement between the predicted and experimentally measured surface topology of the workpiece. The results suggested that only 22 to 30% of the cutting edges exposed on the grinding wheel are active and that the average grinding chip is as much as 10 times thicker and 5 times shorter than would be produced by a grinding wheel with a regular arrangement of cutting edges as assumed by existing analytical approaches.