Search

Innovations in the transport and mobility sector have emerged over the past decades under the umbrella term of the sharing economy. Ride-sourcing and car-sharing are two examples of such innovations. Ride-sourcing companies, such as Uber and Didi Chuxing, have attracted attention mostly in media even though academic research on ride-sourcing is relatively young and rare, especially considering its social sustainability from the perspective of consumers. Likewise, car-sharing has been developed in the western markets over a longer period that the service in China only started over the past few years. Therefore, there is a research gap on social aspects of car-sharing in the Chinese context. This research aims to evaluate the social impact of these two types of vehicle sharing services in the urban China. The research was conducted through mixed-methods. The researcher targeted on two cities in China, Beijing and Guangzhou. To explore the services' impact on social sustainability, a series of qualitative interviews were conducted with key stakeholders, such as service providers, transport planning experts, drivers and passengers of ride-sourcing and users of car-sharing. Additionally, a quantitative approach was used to assess the impact of either service on consumers' attitude and behaviour towards travelling and towards vehicle ownership. Through logistic regression analysis, the relationship between different mobility styles and their corresponding influence over attitude and behaviour was established. For instance, the research results indicated that the more frequent usage of carpooling mode within ride-sourcing was linked with a positive attitude on the sharing practice and this mode displayed significant influence on forgoing or delaying private vehicle ownership. Whereas frequent use of car- sharing in China did not necessarily lead to any significant vehicle ownership reduction. To author's knowledge, this is the first research that explored ride-sourcing and car-sharing in detail in Chinese social sustainability context. With different sharing modes, the results can help policymakers to tailor the design of sustainable mobility strategies for transport sector. For example, efforts should be made on improving the convenience of public transport, especially buses, in order to compete with the increasing use of sharing services. At the same time, the shared use of car-sharing fleets in the form of carpooling should also be promoted to potentially curb the demand for private car ownership. The emphasis on urban spaces in this research can also inspire other regions undergoing urbanisation to build a more sustainable transport and mobility system.

Today, the activities relevant to performance management (PM) can be found in every corner of business, and its importance could be described by a famous business motto that whether a company measures its workforce in hundreds or thousands, its success relies solely on performance. Despite its importance as an enabler of successful business, some issues and shortcomings still exist in the performance management research and its implementation, which can be largely categorized into two main challenges. The first challenge is reflected in the PM dilemma of SMEs. Most existing PM frameworks focus on mechanistic organisations of significant size, yet small and medium enterprises, which comprise 99 percent of business in the UK and 94.15 percent in China, benefit little from the extant PM research. The second PM challenge is combining PM with business and management process innovations. Even for an organisation with simple operating cores, current PM frameworks provide little guidance on how to introduce innovations during performance management system (PMS) building up and management. This issue further causes difficulties in managing performance in complex operating cores, which is exemplified by the challenges of carrying out PM in an R&D unit. The above challenges are quite typical and should be dealt with in the level of performance management framework. We believe that existing PM frameworks are not built around an organisation's performance generation processes and therefore may not be able to handle many issues effectively, including those outlined above. Thus, the research objectives of this thesis are to develop a PM framework that is built around performance generation and also has mechanisms to address the above issues. Furthermore, we need to develop implemental approaches within the framework that can effectively deal with these challenges in real business cases. To accomplish the aforementioned research objectives, a comprehensive typological literature review was carried out to analyze the characteristics and features of the existing PM frameworks. Next, based on the literature research, a new PM framework, namely as the performance tree (PT) framework was introduced in Chapter five, which focuses on the performance generation processes of organisations and also contains mechanisms to accommodate different approaches for a wide range of organisations. In addition, two implemental approaches of PT frameworks were developed in the thesis for the sake of solving the pressing PM issues in SMEs and R&D unit. This research has the following four main contributions: 1. A significant research gap was identified that the existing PM frameworks largely ignored the procedures of performance generation which could lead organisations to be near-sighted, unsustainable, and even experience strategic failure. 2. A new performance management framework, PT framework was developed in this thesis. The PT framework adopts a performance-based perspective to explain performance generation and management processes; also it contains mechanisms to accommodate different approaches for a wide range of organisations seeking to handle the pressing PM issues 3. An implemental approach of PT for classic Chinese manufacturer SMEs was developed in the thesis. Comparing with the existing PM approaches, the new one fully considers the managerial and operational characteristics of SMEs, such as fast-changing organisational chart and high demand for organisational adjustments. 4. An implemental approach of PT for Chinese R&D units under was developed in the thesis. The approach considers both the characteristics of R&D management and the specific PM needs, and hence a PMS in accordance with core operation of R&D units can be developed under its guidance. Meanwhile, a behavioural evidence-based performance measurement system is accommodated in the approach to better measure and evaluate R&D staff's performance.

Genomic clones were isolated and characterized using the human monoamine oxidase A (MAOA) cDNA to screen a phage library, constructed from a human 4X cell line (48, XXXX). The genomic contig derived from overlapping phage clones showed that the size of the MAOA gene is over 80 kb. Exon-containing fragments from these phage clones were subcloned and sequenced. The data from this showed that the MAOA gene consists of 15 exons. A YAC (yeast artificial chromosome) isolated using the MAOA cDNA was characterized. This YAC was found to contain both the MAOA and the MAOB genes. Using PFGE (pulsed-field gel electrophoresis) to investigate the YAC, it was found that the MAOA and the MAOB genes are located within 50 kb and adjacent to each other. The two genes are localized in a 3'-to-3' fashion, suggesting their expression may be regulated independently. The analysis of the homology shown by the two genes clearly demonstrated that they were derived from duplication of a common ancestral gene. A CpG island was discovered to be associated with the 5' end of both genes. A restriction map of -2.5 Mb of genomic DMA around the MAO genes was generated by PFGE. Long-range mapping defined the physical relationship between the marker L1.28 and the MAO genes as L1.28_MAOA_MAOB. A number of genetic diseases have been linked to the Xp11.3 region. Strong linkage was known to exist between the Norrie disease locus and L1.28. Studies showed that some of the Norrie patients have deletions encompassing the region which contains L1.28 as well as the MAO genes. Another YAC isolated by using L1.28 as the probe was also characterized. A phage library was constructed from the L1.28 YAC and the end clones were isolated. Studies on some of the Norrie deletion patients showed that the proximal end clone of the YAC was retained in one of the deletion patients. Previous studies had shown that the Norrie disease locus was also localized proximal to the 5' end of the MAOB gene. The combined information placed the disease locus to an interval of 240 kb within the YAC. More phage clones were characterized in order to define further the region for the Norrie locus which was finally localized within 160 kb. A YAC fragment of 160 kb was isolated and used to screen two human retinal cDNA libraries. Among the cDNAs isolated, one group was found to be deleted in some of the Norrie patients previously without any known deletion, which established their candidacy as the transcripts of the Norrie disease locus. Further characterization of the candidate gene showed that it is conserved across species. The expression of the gene was detected in various tissues. The homology shared between the NDP gene and some of the growth factor binding proteins suggests its role in neural cell proliferation and differentiation.

Determining the molecular mechanisms at the origin of protein function remains a challenge due to the complex non-covalent interactions that shape their structure. Since the non-covalent interactions arise from charge fluctuations, electric fields can be used as a tool to quantify the interactions between a target and its environment. The contribution of each component of the system is reflected in the direction and strength of the electric field exerted on the target, which can be calculated from molecular dynamics simulations. The interactions experienced by ligands in enzymatic active sites determine the catalytic activity of the enzyme. Ligands in synthetic enzymes lack interactions with the protein scaffold, which limit their efficiency. To substitute for the role of non-effective protein scaffold, we introduced a polar DNA fragment to the enzyme vicinity, inducing electrostatic interactions that will facilitate the reaction. We found that the introduction of a DNA fragment enhanced the original interactions between the residues in the active site and the ligand, without creating new interaction hot spots. Using electric fields, we calculated a reduction in activation energy of 2.0 kcal/mol when introducing the DNA fragment, indicating a promising avenue for catalytic improvement. Inspired by the success in using electric fields to understand enzyme catalysis in the context of electrostatic preorganization theory, we generalized these fundamental concepts to another type of proteins: voltage-gated ion channels. Our results indicate that electric fields also report on channel activity. We find an asymmetry in the number of active residues for channel function between the four domains and between the two gating motifs of the permeation pathway, with domain I being the major contributor in both cases. The importance of residues for channel activity is not a simple linear correlation of their distance with the functional motif, but a relationship dominated by non-covalent interactions. Finally, we investigate the effects of loop dynamics on enzyme product inhibition. We modify the chemical nature of the unstructured loops that obstruct the active site of DszB by glycosylating serine and threonine residues. We monitor the corresponding variations in loop dynamics and their effect on the interaction between the enzyme and the product. Overall, promising results were found using electric fields in the investigation of protein mechanisms that are mainly dominated by non-covalent interactions and provide insight into the role of the individual components in the system.

Polymers are a group of materials that consist of large molecules (macromolecules) arranged in a pattern of repeating small subunits. They can be classified into natural polymers and synthetic polymers, degradable and non-degradable polymers. A variety of processing techniques are available to process the polymers, including hot melt extrusion, additive manufacturing, and molding. The versatility of polymer science enables its broad applications to range from packaging, and automotive to biomedical concerns. Particularly, the application in biomedical fields has been of the utmost interest because of the hope of increased longevity and improved quality of life. We here investigated novel polymer engineering methods to improve the interface between materials science and biology and presented a couple of biomedical applications examples based on protein/ polymer composites and non-degradable conductive polymers. Virus nanoparticles (VNPs) (plant viruses in this thesis) are virus-based nanoparticles with highly symmetrical, polyvalent, and monodisperse structures. They can have icosahedral or rod-shaped geometry. Interestingly, the rod-shaped virus, tobacco mild green mosaic virus (TMGMV) presents as high as 1 GPa Young’s modulus and has been utilized as a hydrophilic nanofiller to strengthen the hydrogel networks in this thesis. Cowpea mosaic virus (CPMV) by itself is a potent cancer immunotherapy agent, however, multi-dosage injections are generally required for CPMV treatment. This brings challenges for patients’ compliance as well as difficulty in treating hard-to-inject tumors. Here, we introduced a hot melt extrusion (HME) fabrication method to manufacture a sustained CPMV delivery device. A lyophilization-based genomic material elimination method was accidentally discovered in the development of the CPMV implant. A small molecule STING agonist sustained delivery implant was further fabricated based on the reported HME method and the in vivo efficacy of the biomedical device was examined with murine models. Poly(3,4-ethylenedioxythiophene) (PEDOT) is so far the most conductive commercial conductive polymer, however, 3D printable PEDOT is still under development. We here developed a coagulation-bath-assisted method to pattern PEDOT hydrogel, which resulted in high conductivity and biological tissue-matched mechanical properties. A cortex-wide brain-machine interface was fabricated based on the developed technology, and its feasibility has been studied with an electrical stimulation experiment observed by a wide-field microscope. Polymer science is a highly multidisciplinary subject, we here separately presented the engineering methods of polymers to fabricate protein/polymer composites for drug delivery, and conductive polymers for bioelectronics. However, the idea of protein/polymer composites and bioelectronics can be combined and enabled a “living material” based on bioelectronics. The concept of wirelessly controllable contact lenses for immunotherapy has been proposed in the outlook of the thesis. With the hope of improved human health and increased living quality, the development from the perspective of polymer engineering was studied in this thesis and served for further research investigations.

Effective multi-modal integration of single cell datasets is critical for uncovering the biological properties of cells from different molecular perspectives. However, this poses significant challenges, including how to preserve shared information and account for differences between differently distributed datasets, how to integrate datasets linked by different anchors (cells or features) and how to improve the quality of datasets for integration. In this dissertation, we introduce two novel models that address these challenges. First, we present scDMVAE, a neural network model that can capture both shared and data-specific aspects of datasets in a latent space. scDMVAE can handle both cell-linked and feature-linked datasets through its embedding learning and attention-based matching components, respectively. We demonstrate the effectiveness of scDMVAE on a cell-linked CITE-seq dataset to reveal different cell type relations between mRNA and protein, and on feature-linked SCoPE2 proteomics and scRNA-Seq mRNA human testis datasets to transfer labels from mRNA to protein. Additionally, we present PCRID, a principal curve based model that aligns the retention time of peptides to improve confidence estimates of peptide-spectrum-matches (PSMs) in SCoPE2 technology. PCRID outperforms existing models like DART-ID by handling non-linearities in retention time more effectively, increasing the identification rate of peptides by 154.53 % at a PEP threshold of 0.01 while controlling false discoveries. Together, these models represent significant advances in single cell data analysis and have broad applications across related fields.

Perceiving and manipulating deformable objects with the sense of touch are essential skills in everyday life. However, it remains difficult for robots to autonomously manipulate deformable objects using tactile sensing because of numerous perception, modeling, planning, and control challenges. We believe this is partially due to two fundamental challenges: (1) Establishing a physics-based model describing physical interactions between deformable tactile sensors and deformable objects is difficult; (2) Modern tactile sensors provide high-dimensional data, which is beneficial for perception but impedes the development of practical planning and control strategies. To address these challenges, we developed systematic frameworks for the tactile-driven manipulation of deformable objects that integrates state-of-the-art tactile sensing with well-established tools used by other robotics communities.In Study \#1, we showed how a robot can learn to manipulate a deformable, thin-shell object via tactile sensor feedback using model-free reinforcement learning methods. A page flipping task was learned on a real robot using a two-stage approach. First, we learned nominal page flipping trajectories by constructing a reward function that quantifies functional task performance from the perspective of tactile sensing. Second, we learned adapted trajectories using tactile-driven perceptual coupling, with an intuitive assumption that, while the functional page flipping trajectories for different task contexts (page sizes) might differ, similar tactile sensing feedback should be expected.In Study \#2, we showed how a robot can use tactile sensor feedback to control the pose and tension of a deformable linear object (elastic cable). For a cable manipulation task, low-dimensional latent space features were extracted from high-dimensional raw tactile sensor data using unsupervised learning methods, and a dynamics model was constructed in the latent space using supervised learning methods. The dynamics model was integrated with an optimization-based, model predictive controller for end-to-end, tactile-driven motion planning and control on a real robot.In summary, we developed frameworks for the tactile-driven manipulation of deformable objects that either circumvents sensor modeling difficulties or constructs a dynamics model directly from tactile feedback and uses the model for planning and control. This work provides a foundation for the further development of systematic frameworks that can address complex, tactile-driven manipulation problems.

This paper examines the relation between banks' dividends and their future financial health. Using banks' Nonperforming Loans Ratio, Loan Loss Provision Ratio, and Z-score as proxies for their financial health, I show that there is a strong positive relation between banks' dividends lagged by one quarter and their financial health in the current quarter. This main finding continues to hold following several additional tests, including the application of an instrumental variable approach, the use of change in dividends as the key independent variable, the exclusion of banks that are subject to stress test, the addition of macroeconomic variables, the exclusion of too-big-to-fail banks, and the exclusion of non-depository banks. I also find that the positive relation between banks' dividends and their future financial health is more pronounced for banks with a higher degree of opacity, a lower Tier 1 capital ratio, and during the 2007-2009 financial crisis. This paper contributes to three strands of the finance literature, including the Risk Reduction Hypothesis of dividend signaling in corporate finance, bank dividend policies, and the determinants of banks' financial stability. First, I show that there is a positive relation between banks' dividends lagged by one quarter and their financial health in the current quarter, also meaning that banks' dividends are negatively associated with their future risk conditions. This finding is consistent with the Risk Reduction Hypothesis regarding dividend signaling. Second, Floyd, Li, and Skinner (2015) propose a new idea that banks use dividends to signal financial health, and they rely on this idea to explain why banks have a higher and more stable propensity to pay dividends vis-à-vis industrials during the past several decades. My finding that banks' dividends are positively associated with their future financial health empirically supports this idea proposed by Floyd, Li, and Skinner (2015). Last, to my knowledge, no prior study has attempted to extensively detect a direct relation between banks' dividends and their financial stability. I fill this gap by investigating whether this relation exists. I show that banks' dividends have significantly positive explanatory power on their future financial stability, as proxied by three risk conditions.

Carbon nanotubes (CNTs) have been regarded as one of the most promising materials for electronics applications during the past two decades according to their outstanding electrical properties. However, the fabrication method of CNTs is an important matter since the materials need to be managed easily as well as placed precisely when being applied in electronics. Xu and Hamilton [1] - [2] found an effective solution based application method which can meet those requirements while the process is done at room temperature. This is better than some of the most commonly found methods, such as chemical vapor deposition (CVD), which needs high fabrication temperatures. High fabrication temperatures can cause deformation due to any coefficient of thermal expansion (CTE) mismatches. Their groundbreaking work showed that the electrical properties of CNTs didn’t have a large influence during the fabrication process through both experiment and modeling. However, the thermal interfacial behavior of solution deposited CNT films needed to be characterized. In this thesis, the interfacial thermal impedance of the CNTs, which were fabricated by the method of Xu and Hamilton [1] - [2] was studied. A TIM Tester Model 1400 using ASTM D5470 method by Analysis Tech Inc. was used to characterize the thermal resistance of inkjet CNT films. Copper Alloy 110 disks were used as fabrication substrates on which the ink was printed. The volumes of CNT dispersion were controlled for 2 mL, 4mL, 6 mL and 8 mL. Two groups of samples which were single sided and double sided were tested under four different pressures: 20 psi, 40 psi, 60 psi and 80 psi. Every single test was repeated for 10 times in order to obtain reliable data. The average heat resistances of the last five tests are shown in Table 3.10. All data is listed in appendix. A group of bare copper substrates without MWNTs films was also tested under same conditions as a control reference. The results show a heat insulating, electrical conducting CNTs film using this kind of solution-based application method was obtained.

Fast and accurate diagnosis is always in demand by modern medical professionals and in the area of national defense. At present, limitations of testing speed, sample conditions, and levels of precision exist under current technologies, which are usually slow and involve testing the specimen under laboratory conditions. Typically, these methods also involve several biochemical processing steps and subsequent detection of low energy luminescence or electrical changes, all of which reduce the speed of the test as well as limit the precision. In order to solve these problems and improve the sensing performance, this project proposes an innovative CMOS magnetic biological sensor system for rapidly testing the presence of potential pathogens and bioterrorism agents (zoonotic microorganisms) both in specimens and especially in the environment. The sensor uses an electromagnetic detection mechanism to measure changes in the number of microorganisms--tagged by iron nanoparticles--that are placed on the surface of an integrated circuit (IC) chip. Measured magnetic effects are transformed into electronic signals that count the number and type of organisms present. This biosensor introduces a novel design of a conical-shaped inductor, which achieves ultra-accuracy of sensing biological pathogens. The whole system is integrated on a single chip based on the fabrication process of IBM 180 nm (CMOS_IBM_7RF), which makes the sensor small-sized, portable, high speed, and low cost. The results of designing, simulating, and fabricating the sensor are reported in this dissertation.

This dissertation is composed of three chapters centering on nonparametric econometrics with applications to consumer demand system analysis, value-at-risk analysis of commodity future prices, and credit risk analysis of home mortgage portfolios.The first chapter, based on my joint research with Abdoul Sam considers a semiparametric estimation model for a censored consumer demand system with micro data. A common attribute of disaggregated household data is the censoring of commodities. Maximum likelihood and existing two-step estimators of censored demand systems yield biased and inconsistent estimates when the assumed joint distribution of the disturbances is incorrect. This essay proposes a semiparametric estimator that retains the computational advantage of the two-step methods while circumventing their potential distributional misspecification. The key difference between the proposed estimator and existing two-step counterparts is that the parameters of the binary censoring equations are estimated using a distribution-free single-index model. We implement the proposed estimator using household-level data obtained from the Hainan province in China. Horrowitz and Härdle (1994)'s specification test lends support to our approach.The second chapter is an empirical application of a nonparametric estimator of Value-at-Risk on the cattle feeding margin. Value-at-Risk, known as VaR is a common measure of downside market risk associated with an asset or a portfolio of assets. It has been used as a standard tool of predicting potential portfolio losses for twenty years in the financial industry. Recently VaR has gained popularity in agricultural economics literature since the market price risks associated with agricultural commodities are under evaluation. As initial empirical findings suggest that the performance of any VaR estimation technique is sensitive to the types of data set (portfolio composition) used in developing and evaluating the estimates, agricultural data provides a unique laboratory to further explore VaR and its estimation approaches. This essay as a first attempt applies a distribution-free nonparametric kernel estimator of VaR in an agricultural context, the cattle feeding margin using futures data. The empirical results suggest that the nonparametric VaR estimates enjoy a significant efficiency gain without losing much accuracy compared to the parametric estimates.The third chapter measures credit risks associated with residential mortgage loans. Credit risk is the primary source of risk for real estate lenders. Recent advancements in the measurement and management of credit risk give lenders with sophisticated internal risk models a significant comparative advantage over other lenders in terms of capital optimization and risk controlling. This manuscript helps understand the determinants of credit risk and acquire perspectives on how it is distributed in the current or future loan portfolios. This essay contributes to the existing volume of literature as it incorporates the nonparametric estimation technique into default risk analysis. The CreditRisk model is modified and estimated using the consumer side of information. The model identifies the factors determining household default risks and generates a full loan loss distribution at the portfolio level using consumer finance survey data. In the end, portfolio management strategies are discussed.