Stochastic optimization is an optimization method that solves stochastic problems for minimizing or maximizing an objective function when there is randomness in the optimization process. In this dissertation, various stochastic optimization problems from the areas of Manufacturing, Health care, and Information Cascade are investigated in networks systems. These stochastic optimization problems aim to make plan for using existing resources to improve production efficiency, customer satisfaction, and information influence within limitation. Since the strategies are made for future planning, there are environmental uncertainties in the network systems. Sometimes, the environment may be changed due to the action of the decision maker. To handle this decision-dependent situation, the discrete choice model is applied to estimate the dynamic environment in the stochastic programming model. In the manufacturing project, production planning of lot allocation is performed to maximize the expected output within a limited time horizon. In the health care project, physician is allocated to different local clinics to maximize the patient utilization. In the information cascade project, seed selection of the source user helps the information holder to diffuse the message to target users using the independent cascade model to reach influence maximization. The computation complexities of the three projects mentioned above grow exponentially by the network size. To solve the stochastic optimization problems of large-scale networks within a reasonable time, several problem-specific algorithms are designed for each project. In the manufacturing project, the sampling average approximation method is applied to reduce the scenario size. In the health care project, both the guided local search with gradient ascent and large neighborhood search with Tabu search are developed to approach the optimal solution. In the information cascade project, the myopic policy is used to separate stochastic programming by discrete time, and the Markov decision process is implemented in policy evaluation and updating.

We study multi-level optimization problem on energy system, transportation system and information network. We use the concept of boundedly rational user equilibrium (BRUE) to predict the behaviour of users in systems. By using multi-level optimization method with BRUE, we can help to operate the system work in a more efficient way. Based on the introducing of model with BRUE constraints, it will lead to the uncertainty to the optimization model. We generate the robust optimization as the multi-level optimization model to consider for the pessimistic condition with uncertainty. This dissertation mainly includes four projects. Three of them use the pricing strategy as the first level optimization decision variable. In general, our models' first level's decision variables are the measures that we can control, but the second level's decision variables are users behaviours that can only be restricted within BRUE with uncertainty.

High dimensional data and the presence of outliers in data each pose a serious challenge in supervised learning. Datasets with significantly larger number of features compared to samples arise in various areas, including business analytics and biomedical applications. Such datasets pose a serious challenge to standard statistical methods and render many existing classification techniques impractical. The generalization ability of many classification algorithms is compromised due to the so-called curse of dimensionality. A new binary classification method called constrained subspace classifier (CSC) is proposed for such high dimensional datasets. CSC improves on an earlier proposed classification method called local subspace classifier (LSC) by accounting for the relative angle between subspaces while approximating the classes with individual subspaces. CSC is formulated as an optimization problem and can be solved by an efficient alternating optimization technique. Classification performance is tested in publicly available datasets. The improvement in classification accuracy over LSC shows the importance of considering the relative angle between the subspaces while approximating the classes. Additionally, CSC appears to be a robust classifier, compared to traditional two step methods that perform feature selection and classification in two distinct steps. Outliers can be present in real world datasets due to noise or measurement errors. The presence of outliers can affect the training phase of machine learning algorithms, leading to over-fitting which results in poor generalization ability. A new regression method called relaxed support vector regression (RSVR) is proposed for such datasets. RSVR is based on the concept of constraint relaxation which leads to increased robustness in datasets with outliers. RSVR is formulated using both linear and quadratic loss functions. Numerical experiments on benchmark datasets and computational comparisons with other popular regression methods depict the behavior of our proposed method. RSVR achieves better overall performance than support vector regression (SVR) in measures such as RMSE and R2 adj while being on par with other state-of-the-art regression methods such as robust regression (RR). Additionally, RSVR provides robustness for higher dimensional datasets which is a limitation of RR, the robust equivalent of ordinary least squares regression. Moreover, RSVR can be used on datasets that contain varying levels of noise. Lastly, we present a new novelty detection model called relaxed one-class support vector machines (ROSVMs) that deals with the problem of one-class classification in the presence of outliers.

Online learning is a growing branch of machine learning which allows all traditional data mining techniques to be applied on an online stream of data in real-time. In this dissertation, we present three efficient algorithms for feature ranking in online classification problems. Each of the methods are tailored to work well with different types of classification tasks and have different advantages. The reason for this variety of algorithms is that like other machine learning solutions, there is usually no algorithm which works well for all types of tasks. The first method, is an online sensitivity based feature ranking (SFR) which is updated incrementally, and is designed for classification tasks with continuous features. We take advantage of the concept of global sensitivity and rank features based on their impact on the outcome of the classification model. In the feature selection part, we use a two-stage filtering method in order to first eliminate highly correlated and redundant features and then eliminate irrelevant features in the second stage. One important advantage of our algorithm is its generality, which means the method works for correlated feature spaces without preprocessing. It can be implemented along with any single-pass online classification method with separating hyperplane such as SVMs. In the second method, with help of probability theory we propose an algorithm which measures the importance of the features by observing the changes in label prediction in case of feature substitution. A non-parametric version of the proposed method is presented to eliminate the distribution type assumptions. These methods are application to all data types including mixed feature spaces. At last, we present a class-based feature importance ranking method which evaluates the importance of each feature for each class, these sub-rankings are further exploited to train an ensemble of classifiers. The proposed methods will be thoroughly tested using benchmark datasets and the results will be discussed in the last chapter.

Women labor force participation plays an important role in economic. The developing in economy in Saudi Arabia depends on men rather than women, more than 50 years the Saudi women participation in the labor force extremely is low, this dissertation seeks to identify the challenges and barriers to employment for women in Riyadh and Tabuk. This study examines three research questions. The first question explored the difference between the rate of women unemployment in Tabuk and the rate of women unemployment in Riyadh. The second question investigated ways in which a logistic regression using demographics data could be used to predict the women unemployment rates in two cities. The third question investigated the challenges faced by unemployed women in two cites. An online survey was administrated to both groups. The survey included demographic information and Women Labor Force Participation Instrument. A Chi-Square test was developed from the data to test the differences of the unemployed women in two cites. In order to analyze the second question, the researcher utilized two statistical analysis tests. A logistic regression equation was developed from the data to predict unemployment rates in two cites. Additionally, Partial least squares structural equation modeling were used to analyze the exploratory research question. Content analysis was also used to analyze the challenges faced by unemployed women.

No description available.

The Critical Path and Critical Chain project management methodologies have different and unique characteristics; both methods are used by project managers to manage projects. This research develops and implements two methodologies, milestones (critical path) versus duration (modified critical chain), in a university environment. The implementation of the two methods allows for the comparison of the effectiveness of both styles. The intent is to determine if there is a significant difference in the project management methodologies in a university setting. The design of both projects allows for easy implementation and the ability to easily record determined measureable statistics.

The human-machine interface (HMI) of a Nuclear Power Plant (NPP) Main Control Room (MCR) is complex. Understanding HMI factors that influence Reactor Operator (RO) performance and workload when controlling an NPP is important. The Nuclear Regulatory Commission (NRC) began a program of research known as the Human Performance Test Facility (HPTF) with the goal of collecting human performance data to better understand cognitive and physical elements that support safe control room operation. The HPTF team developed an experimental methodology to evaluate workload using perceived ratings, performance measures, and physiological correlates. This methodology focuses on tasks commonly performed during operations in an NPP. These tasks include monitoring plant parameters, following defined procedures, and manipulating controls to change the state of the NPP. O'Hara and colleagues developed a framework for task classification. Reinerman-Jones and colleagues modified this framework such that monitoring and detection are separate task types. The task types (i.e., checking, detection, and response implementation) selected for experimentation are composed of steps within defined operating procedures that are rule-based. Testing workload using sufficient numbers of ROs is impractical due to limited availability. The HPTF has developed the "equal but different" principle. This principle attempts to simplify complex tasks, such that novices can perform them and experience equivalent workload trends as an expert would when performing the original task. The validity of using the "equal but different" principle with novices in place of experts is uncertain. This research addresses this uncertainty by comparing novices and experts using the "equal but different" principle. Novices performed four tasks within each of the three task types using a simplified Instrument and Control (I&C) panel and a reduced 3-way communication instruction set. Experts performed the same four tasks within each task type with a fully configured I&C panel and a complete 3-way instruction set. Overall, the experts across the three task types tended to rate level of perceived workload lower than novices. However, experts also rated themselves as performing worse for the three task types than novices. Experts performed better than novices when it came to identifying correct I&C; however, their 3-way communication performance was worse. Physiological measures from EEG between the two groups were not statistically different. ECG findings did show a slight difference. The methodology and associated findings has applicability for MCR designs and regulation recommendations. Novice populations are easier to access than experts and the present research shows that when properly designed, novices can serve in complex operator positions.

The growing digital economy has imposed greater demand on the electricity supply's reliability in the past decades, with more consumers and electric vehicles (EV) becoming connected to the electric grid. While the uncertainty of the outcomes is unavoidable, there is a need for more accurate forecasts, modeling tools, and detailed roadmaps that can support the reliable transitioning of power systems. Predicting the electricity demand growth will allow energy managers to understand consumer demand in the near future better. However, there are challenges for forecasting the peak demand growth since it is very difficult to model the various complex features that affect it (i.e., weather patterns, economic growth, etc.). This dissertation contributes to the body of knowledge by proposing an integrated forecasting framework that can help decision-makers deal with the day-to-day outcomes while closely monitoring the monthly peak demand growth. This data-driven framework brings together recent trends in the machine learning and system dynamics field to support forecasting. This hybrid modeling could help deliver more accurate forecasts that will help decision-makers test and adjust their strategies according to critical changes in supply and demand. An actual case study on the Republic of Panama was used to understand the challenges for managing large-scale power systems and the recent impacts that the COVID-19 pandemic had on the energy sector. The case study shows the potential to leverage technological trends (big data, internet of things) and state of the methods such as Convolutional Neural Networks for predicting electricity demand.

No description available.