To help solve the climate crisis, municipalities are increasingly modifying their building codes and offering incentives to create greener buildings in their cities. But, city planners find it difficult to set and assess these policies, as most municipalities do not have the types of data used in urban building energy modelling (UBEM) that would allow their planners to forecast the impacts of various building policies. This thesis offers techniques for operating in this data-poor environment, presenting best practices for developing data-driven archetypes with machine learning, demonstrating inference of parameter values to improve archetypes by using surrogate modelling and genetic algorithms, and a demonstration of techniques for assessing residential retrofit impact in a data-limited environment, where data is neither detailed enough to create an in-depth single archetype study, nor broad enough to create an UBEM model. It will be shown that inference techniques have potential, but need a certain amount of detailed data to work, though far less than traditional UBEM techniques. For performing residential retrofit, it will be shown the lack of ideal detailed data does not present an overwhelming obstacle to drawing useful conclusions and that meaningful insight can be extracted despite the lack of precision. Overall, this thesis shows a data-poor environment, while challenging, is a viable environment for both research and policy modelling. / Graduate

To improve throughput, due date adherence, or tool usage in semiconductor manufacturing, it is crucial to model the duration of individual processes such as coating, diffusion, or etching. Equipped with such data, production planning can develop dispatch schemes and schedules for optimized material routing. However, just a few tools indicate how long a process will take. Many variables affect the runtime of tool recipes that are used to realize processes. These variables include wafer processing mode, historical context, batch size, and job handling. In this thesis, a model that allows inferring tool recipe runtimes with adequate accuracy shall be developed. Firstly, predictive models shall be built for selected tools with known runtime behavior to establish a baseline for the methodology. Tools will be selected to cover a broad spectrum of processing modalities. The main predictors will be revealed using variable importance analysis. Furthermore, the analysis shall reveal under which conditions recipe runtime modeling is most accurate. Secondly, a generic approach shall be created to model recipe runtime. By accounting for tool, process, and material context, methods would be investigated from feature selection and automatic model selection. Finally, a pipeline for data cleansing, feature engineering, model building, and metrics will be developed using historical data from a wide range of factory data sources. Finally, a scheme to operationalize the findings shall be outlined. In particular, this requires establishing model serving to enable consumption in applications such as dispatching or operator interfaces.

No description available.

Classification tasks are an integral part of science, industry, medicine, and business; being such a pervasive technique, its smallest improvement is valuable. Artificial Neural Network (ANN) is one of the strongest techniques used in many disciplines for classification. The ANN technique suffers from drawbacks such as intransparency in spite of its high prediction power. In this dissertation, motivated by learning styles in human brains, ANN’s shortcomings are assuaged and its learning power is improved. Self-Organizing Map (SOM), an ANN variation which has strong unsupervised power, and Feedforward ANN, traditionally used for classification tasks, are hybridized to solidify their benefits and help remove their limitations. These benefits are in two directions: enhancing ANN’s learning power, and improving decision-making. First, the proposed method, named Self-Organizing Error-Driven (SOED) Artificial Neural Network (ANN), shows significant improvements in comparison with usual ANNs. We show SOED is a more accurate, more reliable, and more transparent technique through experimentation with five famous benchmark datasets. Second, the hybridization creates space for inclusion of decision-making goals at the level of ANN’s learning. This gives the classifier the opportunity to handle the inconclusiveness of the data smarter and in the direction of decision-making goals. Through three case studies, naming 1) churn decision analytics, 2) breast cancer diagnosis, and 3) quality control decision making through thermal monitoring of additive manufacturing processes, this novel and cost-sensitive aspect of SOED has been explored and lead to much quantified improvement in decision-making.

This thesis mainly addresses the problems of thermal management in data centers (DCs) through data-driven techniques. For thermal management, a temperature prediction model in the facility is very important, while the thermal modeling based on first principles in DCs is quite difficult due to the complicated air flow and heat transfer. Therefore, we employ multiple data-driven techniques including statistical methods and deep neural networks (DNNs) to represent the thermal dynamics. Then based on such data-driven models, temperature estimation and control are implemented to optimize the thermal management in DCs. The contributions of this study are summarized in the following four aspects: 1) A data-driven model constructed through multiple linear Autoregression exogenous (ARX) models is adopted to describe the thermal behaviors in DCs. On the basis of such data-driven model, an observer of adaptive Kalman filter is proposed to estimate the temperature distribution in DC. 2) Based on the data-driven model proposed in the first work, a data-driven fault tolerant predictive controller considering different actuator faults is developed to regulate the temperature in DC. 3) To improve the modeling accuracy, a deep input convex neural network (ICNN) is adopted to implement thermal modeling in DCs, which is also specifically designed for further control design. Besides, the algorithm of elastic weight consolidation (EWC) is employed to overcome the catastrophic forgetting in continual learning. 4) A novel example reweighting algorithm is utilized to enhance the robustness of ICNN against noisy data and avoid overfitting in the training process. Finally, all the proposed approaches are validated in real experiments or experimental-data-based simulations. / Dissertation / Doctor of Philosophy (PhD) / This thesis mainly investigates the applications of data-driven techniques for thermal management in data centers. The implementations of thermal modeling, temperature estimation and temperature control in data centers are the key contributions in this work. First, we design a data-driven statistical model to describe the complicated thermal dynamics of data center. Then based on the data-driven model, efficient observer and controller are developed respectively to optimize the thermal management in data centers. Moreover, to improve the nonlinear modeling performance in data centers, specific deep input convex neural networks capable of good representation capability and control tractability are adopted. This thesis also proposes two novel strategies to avoid the influence of catastrophic forgetting and noisy data respectively during the training processes. Finally, all the proposed techniques are validated in real experiments or experimental-data-based simulations.

One of the crucial challenges for Data Center (DC) operation is inefficient thermal management which leads to excessive energy waste. The information technology (IT) equipment and cooling systems of a DC are major contributors to power consumption. Additionally, failure of a DC cooling system leads to higher operating temperatures, causing critical electronic devices, such as servers, to fail which leads to significant economic loss. Improvements can be made in two ways, through (1) better design of a DC architecture and (2) optimization of the system for better heat transfer from hot servers. Row-based cooling is a suitable DC configuration that reduces energy costs by improving airflow distribution. Here, the IT equipment is contained within an enclosure that includes a cooling unit which separates cold and back chambers to eliminate hot air recirculation and cold air bypass, both of which produce undesirable airflow distributions. Besides, due to scalability, ease of implementation, and operational cost, row-based systems have gained in popularity for DC computing applications. However, a general thermal model is required to predict spatiotemporal temperature changes inside the DC and properly apply appropriate strategies. As yet, only primitive tools have been developed that are time-consuming and provide unacceptable errors during extrapolative predictions. We address these deficiencies by developing a rapid, adaptive, and accurate hybrid model by combining a DDM and the thermofluid transport relations to predict temperatures in a DC. Our hybrid model has low interpolative prediction errors below 0.7 oC and extrapolative errors less than one half of black-box models. Additionally, by changing the studied DC configuration such as cooling unit fans and severs locations, there are a few zones with prediction error more than 2 oC. Existing methods for cooling unit fault detection and diagnosis (FDD) are designed to successfully overcome individually occurring faults but have difficulty handling simultaneous faults. We apply a gray-box model involves a case study to detect and diagnose cooling unit fan and pump failure in a row-based DC cooling system. Fast detection of anomalous behavior saves energy and reduces operational costs by initiating remedial actions. Cooling unit fans and pumps are relatively low-reliability components, where the failure of one or more components can cause the entire system to overheat. Therefore, appropriate energy-saving strategies depend largely on the accuracy and timeliness of temperature prediction models. We used our gray-box model to produce thermal maps of the DC airspace for single as well as simultaneous failure conditions, which are fed as inputs for two different data-driven classifiers, CNN and RNN, to rapidly predict multiple simultaneous failures. Our FDD strategy can detect and diagnose multiple faults with accuracy as high as 100% while requiring relatively few simultaneous fault training data samples. / Thesis / Candidate in Philosophy

Battery electric buses (BEBs) bring several advantages to public transportation systems. With fixed routes and scheduled trips, the implementation of BEBs in the transit context is considered a seamless transition towards a zero greenhouse gases transit system. However, energy consumption uncertainty is a significant deterrent for mainstream implementation of BEBs. Demonstration and trial projects are often conducted to better understand the uncertainty in energy consumption (EC). However, the BEB's energy consumption varies due to uncertainty in operational, topological, and environmental attributes. This thesis aims at developing simulation, data-driven, and low-resolution models using big data to quantify the EC of BEBs, with the overarching goal of developing a comprehensive planning framework for BEB implementation in bus transit networks. This aim is achieved through four interwind objectives. 1) Quantify the operational and topological characteristics of bus transit networks using complex network theory. This objective provides a fundamental base to understanding the behaviour of bus transit networks under disruptive events. 2) Investigate the impacts of the vehicular, operational, topological, and external parameters on the EC of BEBs. 3) Develop and evaluate the feasibility of big-data analytics and data-driven models to numerically estimate BEB's EC. 4) Create an open-source low-resolution data-based framework to estimate the EC of BEBs. This framework integrates the modelling efforts in objectives 1-3 and offers practical knowledge for transit providers. Overall, the thesis provides genuine contributions to BEB research and offers a practical framework for addressing the EC uncertainty associated with BEB operation in the transit context. Further, the results offer transit planners the means to set up the optimum transit operations profile that improves BEB energy utilization, and in turn, reduces transit-related greenhouse gases. / Thesis / Doctor of Engineering (DEng)

According to the Centre for Research on the Epidemiology of Disasters, the global average number of CID has tripled in less than four decades (from approximately 1,300 Climate-Induced Disasters (CID) between 1975 and 1984 to around 3,900 between 2005 and 2014). In addition, around 1 million deaths and $1.7 trillion damage costs were attributed to CID since 2000, with around $210 billion incurred only in 2020. Consequently, the World Economic Forum identified extreme weather as the top ranked global risk in terms of likelihood and among the top five risks in terms of impact in the last 4 years. These risks are not expected to diminish as: i) the number of CID is anticipated to double during the next 13 years; ii) the annual fatalities due to CID are expected to increase by 250,000 deaths in the next decade; and iii) the annual CID damage costs are expected to increase by around 20% in 2040 compared to those realized in 2020. Given the anticipated increase in CID frequency, the intensification of CID impacts, the rapid growth in the world’s population, and the fact that two thirds of such population will be officially living in urban areas by 2050, it has recently become extremely crucial to enhance both community and city resilience under CID. Resilience, in that context, refers to the ability of a system to bounce back, recover or adapt in the face of adverse events. This is considered a very farfetched goal given both the extreme unpredictability of the frequency and impacts of CID and the complex behavior of cities that stems from the interconnectivity of their comprising infrastructure systems. With the emergence of data-driven machine learning which assumes that models can be trained using historical data and accordingly, can efficiently learn to predict different complex features, developing robust models that can predict the frequency and impacts of CID became more conceivable. Through employing data analytics and machine learning techniques, this work aims at enhancing city resilience by predicting both the occurrence and expected impacts of climate-induced disasters on urban areas. The first part of this dissertation presents a critical review of the research work pertaining to resilience of critical infrastructure systems. Meta-research is employed through topic modelling, to quantitatively uncover related latent topics in the field. The second part aims at predicting the occurrence of CID by developing a framework that links different climate change indices to historical disaster records. In the third part of this work, a framework is developed for predicting the performance of critical infrastructure systems under CID. Finally, the aim of the fourth part of this dissertation is to develop a systematic data-driven framework for the prediction of CID property damages. This work is expected to aid stakeholders in developing spatio-temporal preparedness plans under CID, which can facilitate mitigating the adverse impacts of CID on infrastructure systems and improve their resilience. / Thesis / Doctor of Philosophy (PhD)

Background: Data-driven decision-making (DDDM) has emerged as a primary approach to decision-making in many organizations. It uses data and analytics to guide decision-making processes and can lead to better business outcomes. Prior research has focused on DDDM in large corporations operating in large economies, and therefore this thesis will examine DDDM in small and medium enterprises in Sweden.  Purpose: The purpose of this research study is to examine the effect DDDM has on the financial performance of Swedish SMEs to investigate if the utilization of DDDM benefits companies financially and to understand the effect of managerial experience, technical skills, information quality, and firm size on the data-driven decision-making process. Method: This study is based on the positivism paradigm, following deductive reasoning and a quantitative approach of gathering data through digital surveys. The sample consisted of 55 Swedish SMEs gathered through simple random sampling. Further, the data was analyzed using Pearson correlation, Spearman rank correlation, and regression analysis to test hypotheses. Findings: The literature review identified a research gap on DDDM, factors that effect DDDM, and Financial Performance. Four hypotheses were developed to answer the research questions. The OLS regression found that DDDM had no significant effect on Financial Performance, the first hypothesis was not supported. The Information Quality variable had a significant positive effect on DDDM resulting in support for the second hypothesis. However, Managerial Experience and Technical Skills did not have a significant effect in the main regression model, hypotheses three and four were not supported. Conclusion: The thesis showed that DDDM did not have a significant effect on financial performance in Swedish SMEs. Additionally, managerial expertise and technical skills did not have an effect on DDDM. However, Information quality did have an effect on the DDDM process and was correlated with technical skills, which is in line with the theories used in the study: Organizational Information Processing Theory (OIPT) and Absorptive Capacity. This further supports that information quality is vital for the DDDM process and can explain why DDDM might not always lead to improvements in financial performance for Swedish SMEs.

With the emphasis on high standards and fiscal accountability, there is a heightened need to inform the research linking student achievement to the allocation of resources. This mixed methods inquiry sought to study how schools utilized Title 1 and Title 1 stimulus funding from 2009-2011 to determine if correlations existed between areas of resource utilization and student achievement by studying both the use of funding and the processes that fifteen elementary and middle Title 1 schools in southern California utilized. The focus was to document resource use of Title 1 and Title 1 stimulus allocations and determine if a correlation existed between expenditures and improved student achievement (quantitative) and to discover themes that existed in student achievement improvement, especially including factors that affect the decision making process at the school (qualitative). Findings suggested that expenditures for professional development and programs for at-risk students played a key role in student achievement growth. The leadership of the school principal was also an indicator of student achievement growth. The use of Title 1 monies, including the increase in Title 1 stimulus monies, were beneficial to schools and positively contributed to the increase in student achievement. Overall, money, when spent well, led to improved student achievement.