The Weapon Combat Effectiveness (WCE) analytics is very expensive, time-consuming, and dangerous in the real world because we have to create data from the real operations with a lot of people and weapons in the actual environment. The Modeling and Simulation (M&S) of many techniques is used for overcoming these limitations. Although the era of big data has emerged and achieved a great deal of success in a variety of fields, most WCE research using the Defense Modeling and Simulation (DM&S) techniques were studied without the help of big data technologies and techniques. The existing research has not considered various factors affecting WCE. This is because current research has been restricted by only using constructive simulation, a single weapon system, and limited scenarios. Therefore, the WCE analytics using existing methodologies have also incorporated the same limitations, and therefore, cannot help but get biased results. To solve the above problem, this dissertation is to initially review and compose the basic knowledge for the new WCE analytics methodology using big data and DM&S to further serve as the stepping-stone of the future research for the interested researchers. Also, this dissertation presents the new methodology on WCE analytics using big data generated by Live, Virtual, or/and Constructive (LVC) simulations. This methodology can increase the fidelity of WCE analytics results by considering various factors. It can give opportunities for application of weapon acquisition, operations analytics and plan, and objective level development on each training factor for the weapon operators according to the selection of Measures of Effectiveness (MOEs) and Measures of Performance (MOPs), or impact factors, based on the analytics goal.

This dissertation focuses on developing models to study the problem of searching and retrieving items in a dense storage environment. We consider a special storage configuration called an inverted T configuration, which has one horizontal and one vertical aisle. Inverted T configurations have fewer aisles than a traditional aisle-based storage environment. This increases the storage density; however, requires that some items to be moved out of the way to gain access to other more deeply stored items. Such movement can result in item location uncertainty. When items are requested for retrieval in a dense storage environment with item location uncertainty, searching is required. Dense storage has a practical importance as it allows for the use of available space efficiently, which is especially important with the scarce and expensive space onboard of US Navy's ships that form a sea base. A sea base acts as a floating distribution center that provides ready issue material to forces ashore participating in various types of missions. The sea basing concept and the importance of a sea base's responsiveness is our main motivation to conduct this research. In chapter 2, we review three major bodies of literature: 1) sea based logistics, 2) dense storage and 3) search theory. Sea based logistics literature mostly focuses on the concept and the architecture of a sea base, with few papers developing mathematical models to solve operational problems of a sea base, including papers handling the logistical and sustainment aspects. Literature related to dense storage can be broken down into work dealing with a dense storage environment with an inverted T configuration and other papers dealing with other dense storage configurations. It was found that some of the dense storage literature was motivated by the same application, i.e. sea based logistics. Finally, we surveyed the vast search theory literature and classification of search environments. This research contributes to the intersection of these three bodies of literature. Specifically, this research, motivated by the application of sea basing, develops search heuristics for dense storage environments that require moving items out of the way during searching. In chapter 3, we present the problem statements. We study two single-searcher search problems. The first problem is searching for a single item in an inverted T dense storage environment. The second one is searching for one or more items in an inverted T storage environment with items stacked over each other in the vertical direction. In chapter 4, we present our first contribution. In this contribution we propose a search plan heuristic to search for a single item in an inverted T, k-deep dense storage system with the objective of decreasing the expected search time in such an environment. In this contribution, we define each storage environment entirely by the accessibility constant and the storeroom length. In addition, equations are derived to calculate each component of the search time equation that we propose: travel, put-back and repositioning. Two repositioning policies are studied. We find that a repositioning policy that uses the open aisle locations as temporary storage locations and requires put-back of these items while searching is recommended. This recommendation is because such a policy results in lower expected search time and lower variability than a policy that uses available space outside the storage area and handles put-back independently of the search process. Statistical analysis is used to analyze the numerical results of the first contribution and to analyze the performances of both repositioning polices. We derive the probability distribution of search times in a storeroom with small configurations in terms of the accessibility constant and length. It was found that this distribution can be approximated using a lognormal probability distribution with a certain mean and standard deviation. Knowing the probability distribution provides the decision makers with the full range of all possible probabilities of search times, which is useful for downstream planning operations. In chapter 5, we present the second contribution, in which we propose a search plan heuristic but for multiple items in an inverted T, k-deep storage system. Additionally, we consider stacking multiple items over each other. Stacking items over each other, increases the number of stored items and allows for the utilization of the vertical space. In this second contribution, we are using the repositioning policy that proved its superiority in the first contribution. This contribution investigates a more general and a much more challenging environment than the one studied in the first contribution. In the second environment, to gain access to some items, not only may other items need to be moved out of the way, but also the overall number of movements for items within the system will be highly affected by the number of items stacked over each other. In addition, the searcher is given a task that includes searching and retrieving a set of items, rather than just one item. For the second contribution, the performance of the search heuristic is analyzed through a Statistical Design of Experiments, and it was found that searching and retrieving multiple items instead of just a single item, would decrease the variability in search times for each storeroom configuration. Finally, in chapter 6, conclusions of this research and suggestions for future research directions are presented.

In this dissertation, various optimization problems from the area of transportation and power systems will be respectively investigated and the uncertainty will be considered in each problem. Specifically, a long-term problem of electricity infrastructure investment is studied to address the planning for capacity expansion in electrical power systems with the integration of short-term operations. The future investment costs and real-time customer demands cannot be perfectly forecasted and thus are considered to be random. Another maintenance scheduling problem is studied for power systems, particularly for natural gas fueled power plants, taking into account gas contracting and the opportunity of purchasing and selling gas in the spot market as well as the maintenance scheduling considering the uncertainty of electricity and gas prices in the spot market. In addition, different vehicle routing problems are researched seeking the route for each vehicle so that the total traveling cost is minimized subject to the constraints and uncertain parameters in corresponding transportation systems. The investigation of each problem in this dissertation mainly consists of two parts, i.e., the formulation of its mathematical model and the development of solution algorithm for solving the model. The stochastic programming is applied as the framework to model each problem and address the uncertainty, while the approach of dealing with the randomness varies in terms of the relationships between the uncertain elements and objective functions or constraints. All the problems will be modeled as stochastic mixed-integer programs, and the huge numbers of involved decision variables and constraints make each problem large-scale and very difficult to manage. In this dissertation, efficient algorithms are developed for these problems in the context of advanced methodologies of optimization and operations research, such as branch and cut, benders decomposition, column generation and Lagrangian method. Computational experiments are implemented for each problem and the results will be present and discussed. The research carried out in this dissertation would be beneficial to both researchers and practitioners seeking to model and solve similar optimization problems in transportation and power systems when uncertainty is involved.

Emergency Departments (EDs) represent a crucial component of any healthcare infrastructure. In today's world, healthcare systems face growing challenges in delivering efficient and time-sensitive emergency care services to communities. Overcrowding within EDs represents one of the most significant challenges for healthcare quality that adversely impacts clinical outcomes, patient safety, and overall satisfaction. Research in this area has resulted in creating several ED crowding indices, such as National Emergency Department Overcrowding Scale (NEDOCS) and Emergency Department Work Index (EDWIN) that have been developed to provide measures aimed at mitigating overcrowding. Recently, efforts made by researchers to examine the validity and reproducibility of these indices have shown that they are not reliable in accurately assessing overcrowding in regions beyond their original design settings. The shortcomings of such indices stem from their reliance upon the perspective and feedback of only clinical staff and the exclusion of other stakeholders. These limited perspectives introduce bias in the results of ED overcrowding indices. This study starts with confirming the inaccuracy of such crowding indices through examining their validity within a new healthcare system. To overcome the shortcomings of previous indices, the study presents a novel framework for quantifying and managing overcrowding based on emulating human reasoning in overcrowding perception. The framework of the proposed study takes into consideration emergency operational and clinical factors such as patient demand, patient complexity, staffing level, clinician workload, and boarding status when defining the crowding level. The hierarchical fuzzy logic approach is utilized to accomplish the goals of this framework by combining a diverse pool of healthcare expert perspectives while addressing the complexity of the overcrowding issue. The innovative design of the developed framework reduces bias in the assessment of ED crowding by developing a knowledge-base from the perspectives of multiple experts, and allows for its implementation in a variety of healthcare settings. Statistical analysis of results indicate that the developed index outperform previous indices in reflecting expert subjective assessments of overcrowding.

Manufacturing is a crucial element in the global economy. During the last decade, the national manufacturing sector loses nearly 30% of its workforce and investments. Consequently, the quality of the domestic goods, global share, and manufacturing capabilities has been declined. Therefore, innovative ways to optimize the usage of the Smart Manufacturing Systems (SMS) are required to form a new manufacturing era. This research is presenting a framework to optimize the design of SMS. This includes the determination of the suitable machines that can perform the job efficiently, the quantity of those machines, and the potential messaging system required for sharing information. Multiple reviews are used to form the framework. Expert machine selection matrix identifies the required machines and machine parameter matrix defines the specifications of those machines. While business process modeling and notation (BPMN) captures the process plan in object-oriented fashion. In addition, to agent unified modeling language (AUML) that guides the application of message sequence diagram and statecharts. Finally, the configuration is obtained from a hybrid simulation model. Agent based-modeling is used to capture the behavior of the machines where discrete event simulation mimics the process flow. A case study of a manufacturing system is used to verify the study. As a result, the framework shows positive outcomes in supporting upper management in the planning phase of establishing a SMS or evaluating an existing one.

In this dissertation, we study the long-term electricity infrastructure investment planning problems in the electrical power system. These long-term capacity expansion planning problems aim at making the most effective and efficient investment decisions on both thermal and wind power generation units. One of our research focuses are uncertainty modeling in these long-term decision-making problems in power systems, because power systems' infrastructures require a large amount of investments, and need to stay in operation for a long time and accommodate many different scenarios in the future. The uncertainties we are addressing in this dissertation mainly include demands, electricity prices, investment and maintenance costs of power generation units. To address these future uncertainties in the decision-making process, this dissertation adopts two different optimization approaches: decision-dependent stochastic programming and adaptive robust optimization. In the decision-dependent stochastic programming approach, we consider the electricity prices and generation units' investment and maintenance costs being endogenous uncertainties, and then design probability distribution functions of decision variables and input parameters based on well-established econometric theories, such as the discrete-choice theory and the economy-of-scale mechanism. In the adaptive robust optimization approach, we focus on finding the multistage adaptive robust solutions using affine policies while considering uncertain intervals of future demands. This dissertation mainly includes three research projects. The study of each project consists of two main parts, the formulation of its mathematical model and the development of solution algorithms for the model. This first problem concerns a large-scale investment problem on both thermal and wind power generation from an integrated angle without modeling all operational details. In this problem, we take a multistage decision-dependent stochastic programming approach while assuming uncertain electricity prices. We use a quasi-exact solution approach to solve this multistage stochastic nonlinear program. Numerical results show both computational efficient of the solutions approach and benefits of using our decision-dependent model over traditional stochastic programming models. The second problem concerns the long-term investment planning with detailed models of real-time operations. We also take a multistage decision-dependent stochastic programming approach to address endogenous uncertainties such as generation units' investment and maintenance costs. However, the detailed modeling of operations makes the problem a bilevel optimization problem. We then transform it to a Mathematic Program with Equilibrium Constraints (MPEC) problem. We design an efficient algorithm based on Dantzig-Wolfe decomposition to solve this multistage stochastic MPEC problem. The last problem concerns a multistage adaptive investment planning problem while considering uncertain future demand at various locations. To solve this multi-level optimization problem, we take advantage of affine policies to transform it to a single-level optimization problem. Our numerical examples show the benefits of using this multistage adaptive robust planning model over both traditional stochastic programming and single-level robust optimization approaches. Based on numerical studies in the three projects, we conclude that our approaches provide effective and efficient modeling and computational tools for advanced power systems' expansion planning.

The unsupervised ensemble learning, or consensus clustering, consists of finding the optimal com- bination strategy of individual partitions that is robust in comparison to the selection of an algorithmic clustering pool. Despite its strong properties, this approach assigns the same weight to the contribution of each clustering to the final solution. We propose a weighting policy for this problem that is based on internal clustering quality measures and compare against other modern approaches. Results on publicly available datasets show that weights can significantly improve the accuracy performance while retaining the robust properties. Since the issue of determining an appropriate number of clusters, which is a primary input for many clustering methods is one of the significant challenges, we have used the same methodology to predict correct or the most suitable number of clusters as well. Among various methods, using internal validity indexes in conjunction with a suitable algorithm is one of the most popular way to determine the appropriate number of cluster. Thus, we use weighted consensus clustering along with four different indexes which are Silhouette (SH), Calinski-Harabasz (CH), Davies-Bouldin (DB), and Consensus (CI) indexes. Our experiment indicates that weighted consensus clustering together with chosen indexes is a useful method to determine right or the most appropriate number of clusters in comparison to individual clustering methods (e.g., k-means) and consensus clustering. Lastly, to decrease the variance of proposed weighted consensus clustering, we borrow the idea of Markowitz portfolio theory and implement its core idea to clustering domain. We aim to optimize the combination of individual clustering methods to minimize the variance of clustering accuracy. This is a new weighting policy to produce partition with a lower variance which might be crucial for a decision maker. Our study shows that using the idea of Markowitz portfolio theory will create a partition with a less variation in comparison to traditional consensus clustering and proposed weighted consensus clustering.

In this paper, a new framework for sharing Combat Damage Assessment(CDA) is proposed to find out the differences of each CDA system between military combat units belonging to their own federate in a HLA/RTI federation. When there are engagements in a battle among combat units belonging to their own federate in the HLA/RTI federation, each result of damage assessments is very different. This affects the HLA/RTI federation’s confidence and needed to be overcome because it is also one of the major issues to generate reliable engagement data. Also, a RTI can generate only qualitative data about combat damage while quantitative data can be useful. Therefore, the new framework for sharing CDA and generating quantitative CDA data is proposed to solve the problems with a CDA Module of one federate which is considered to have a standard engagement logic. The new framework is also tested through two case studies by using two federates of a HLA 1516 / MÄK RTI federation. This new framework will be helpful to increase the interoperability in a HLA/RTI federation, provide an environment in which all developers can reuse the proposed new framework, and generate quantitative engagement data through this new framework.

The healthcare system is always defined as a complex system. At its core, it is a system composed of people and processes and requires performance of different tasks and duties. This complexity means that the healthcare system has many stakeholders with different interests, resulting in the emergence of many problems such as increasing healthcare costs, limited resources and low utilization, limited facilities and workforce, and poor quality of services. The use of simulation techniques to aid in solving healthcare problems is not new, but it has increased in recent years. This application faces many challenges, including a lack of real data, complicated healthcare decision making processes, low stakeholder involvement, and the working environment in the healthcare field. The objective of this research is to study the utilization of case-based reasoning in simulation modeling in the healthcare sector. This utilization would increase the involvement of stakeholders in the analysis process of the simulation modeling. This involvement would help in reducing the time needed to build the simulation model and facilitate the implementation of results and recommendations. The use of case-based reasoning will minimize the required efforts by automating the process of finding solutions. This automation uses the knowledge in the previously solved problems to develop new solutions. Thus, people could utilize the simulation modeling with little knowledge about simulation and the working environment in the healthcare field. In this study, a number of simulation cases from the healthcare field have been collected to develop the case-base. After that, an indexing system was created to store these cases in the case-base. This system defined a set of attributes for each simulation case. After that, two retrieval approaches were used as retrieval engines. These approaches are K nearest neighbors and induction tree. The validation procedure started by selecting a case study from the healthcare literature and implementing the proposed method in this study. Finally, healthcare experts were consulted to validate the results of this study.

Creativity has been widely recognized as critical to the economic success of organizations for over 60 years. Today, it is considered to be the most highly prized "commodity" of businesses. As such, there have been numerous efforts to better understand creativity with the goal of increasing individual creativity and therefore improving the economic success of organizations. An emerging area of research on creativity recognizes creativity as a complex, social process that is dependent upon many factors, including those of an environmental nature. In support of this perspective, a growing amount of research has investigated the effect of social networks on individual creativity. This relationship is based on the premise that an individual's social network affects access to diverse information, which in turn, is critical for creativity. The previous studies on this relationship, however, have been conducted in a limited number of environments, most of which have been knowledge-intensive in nature. As such, this study was conducted in a fast-food restaurant environment to determine whether the relationship between social networks and creativity is the same as in other, previously studied environments. Data was collected for a sample of 247 employees of an organization consisting of seven fast-food franchise restaurants of a popular fast-food restaurant chain in the northeast region of the United States. An ordinary least squares regression model was developed to investigate the relationship between creativity and the commonly studied social network variables: number of weak ties, number of strong ties, clustering, and centrality. The social network variables accounted for 17.3% of the overall variance in creativity, establishing that a relationship does exist between social networks and creativity in the fast-food restaurant environment. This relationship, however, was not as expected. In contrast to expectations, weak ties were not found to be a significant, positive predictor of creativity. Also, strong ties were found to be a significant, positive predictor of creativity, where it was expected that this relationship would be in the negative direction. Centrality, however, was found to be a significant, positive predictor of creativity, as expected, while the results for clustering were inconclusive due to its high correlation with the other social network variables in the study. As such, it appears that the relationship between social networks and creativity may be different in the fast-food restaurant environment when compared to environments previously studied. It is possible that this difference is a result of the differences between high and low knowledge-intensive working environments. The lack of support for weak ties as a significant positive predictor of creativity in conjunction with limited opportunities for significant creative achievement suggests that access to diverse information may be less important for creativity in the fast-food restaurant environment than in other environments. The findings that strong ties and centrality are significant, positive predictors of creativity, however, appear to indicate that the ability to implement a creative idea, however minor it may be, is more important in the fast-food restaurant environment than the generation of that idea in the first place. Due to the limitations of this study, however, it is not possible to definitively conclude this notion without efforts to determine which factor afforded by positions rich in strong ties or high in centrality, the informational benefits or the organizational influence, is more important for creativity.