<p>Technological systems contain complex elements and processes with a diverse set of agents and problem-solving arrangements. They often interact with and influence multi-lateral stakeholders with varying interests and incentives. Recent technological developments and engineering advancements such as digital marketplaces and high-tech networks create both new challenges and opportunities to understand further about effective mechanism designs. This dissertation attempts to answer corporate-level mechanism design issues in two different technological systems: high-tech biopharmaceutical networks and the online peer-to-peer lending industry.</p><p><br></p><p>The first part of the dissertation focuses on identifying the emergence and evolution of near decomposable systems in interorganizational relationships. To do so, first I conceptually discuss how near decomposable systems can emerge in interfirm relationships. Second, leveraging advancements in network science, I empirically analyze a detailed biopharmaceutical alliance data set and find that strategic alliance networks of biopharmaceutical firms exhibit near decomposable characteristics. I identify an emerging evolutionary pattern with smaller networks of subcommunities organizing hierarchically over time into a larger network structure, with the subcommunities generally exhibiting local clustering. A salient finding, compared to previous studies in the field of strategic management, is the identification of nested clusters formed in hierarchical fashion within this interfirm network. I find the potential for simultaneous evolutionary processes to be in play in various subnetworks within the overall industry-level network. The accrual of local changes impacting the structural processes of the subnetworks slowly diffuses to the larger, less integrated modules of the network. Finally, with the help of a simulation model, I identify how fitness heterogeneity among firms, fitness heterogeneity among partnerships and the rate of growth of partnerships impact the emergence of near decomposability in varying degrees.</p><p><br></p><p>The second study focuses on understanding an important market access control mechanism: platform owners granting priority access to a subset of supply-side complementors to grow the marketplace and remove potential demand-side bottlenecks. Platform governance mechanisms, such as market access control, help to align all market players towards a specific value proposition. I study the interplay between priority access and the variation in expertise of the complementors. Leveraging a randomized priority access given to expert institutional investors in the online peer-to-peer lending industry, I show that it creates negative spillover effects on the performance of crowd retail investors. I provide evidence in support of two mechanisms in driving the impact of priority access, the intensity of priority access and cream skimming by institutional complementors, on the retail crowd market. Again using simulation to extend the analysis, I find that the brunt of negative impacts is likely borne by more risk-averse retail investors.</p>

<p> People in developing countries have limited access to electricity, especially in rural and remote areas. As electricity consumption is correlated with economic development, the lack of access to electricity is a key obstacle to achieving economic growth. Techniques for improving access to electricity include grid extension and distributed energy resources (DER), but analyzing the tradeoff between grid extension and distributed generation requires a better understanding of the impacts of grid unreliability. In this dissertation, a new method for simulating unreliable electric grids is presented. The method is then used to determine the cost of reliable electricity in areas where the grid is unreliable. The method is extended in order to calculate the distance at which grid extension of an unreliable grid and DER have the same cost, a point known as the economic distance limit (EDL). Finally, the method is applied to analyze the impact of grid sell-back prices on electricity cost and EDL. The methods are demonstrated for a village in Uganda, but hold universally. Results indicate that demand for increased availability increases cost, but now the cost per unit of availability can be calculated and used in decision making. Similarly, with fixed demand availability, we see increasing costs as grid availability decreases. This also results in EDL decreasing as grid availability decreases, as there is little value in extending a grid that functions poorly. From the simulation results, linear approximations of some of the key outputs are developed and are demonstrated to be consistent with results. These provide a method for rapidly calculating electricity costs and EDL without the need to perform numerous simulations. Simple calculations for cost of highly available electricity will enable more informed choices for grid-tied and stand-alone electricity generation for system operators and for policy makers.</p>

<p> The objective of this study was to explore if New Hampshire is experiencing &lsquo;Brain Drain&rsquo; phenomenon. For this study, Brain Drain is defined as a metaphor denoting &ldquo;difficulty finding skilled and talented workers to work in New Hampshire&rdquo;. This study explored the lived experiences and perceptions of two groups over many years, including the recent downturn in the economy and through its ongoing recovery.</p><p> Group 1 were participants from New Hampshire State Level Offices and State Organizations whose purpose is to assist New Hampshire businesses and their workforces. They were asked if they believed New Hampshire was experiencing brain drain phenomenon, and if so how were the New Hampshire State Offices and State Organizations trying to solve it. Group 2 were participants from New Hampshire Businesses, residing in all 10 New Hampshire counties, as well as across different Industry Sectors. They were asked if they believed if their New Hampshire Businesses were experiencing brain drain phenomenon, and if so how were their New Hampshire Businesses trying to solve it. The researcher compared the results of Group 1 and Group 2, to see if they believed New Hampshire and the New Hampshire Businesses were experiencing brain drain phenomenon, and if so were both groups working together to help solve it?</p><p> The study revealed three major findings. First, both Groups believed New Hampshire was experiencing brain drain phenomenon, however there was a disconnect on how they were both trying to solve it. Second, the results demonstrated that the New Hampshire Businesses were experiencing two types of brain drain phenomenon. They were finding difficulty filling Knowledge and Information Age jobs&mdash;mainly in the STEM (Science, Technology, Engineering and Mathematic) fields. However, even more evident, they were finding much more difficulty filling Industrial Age jobs, mainly in the Manufacturing industry sector. Third, the two types of brain drain phenomenon were being compounded with a decline in New Hampshire workforce population, as demographics show the state&rsquo;s population is aging.</p><p> Leadership examples from both groups were cited. Future leadership opportunities were presented. Conclusion: Strive to match business opportunities, with the right skills, and available population.</p>

In this dissertation, we investigate three topics: predictive models for disease diagnosis and patient behavior, optimization for cancer treatment planning, and public health decision making for infectious disease prevention. In the first topic, we propose a multi-stage classification framework that incorporates Particle Swarm Optimization (PSO) for feature selection and discriminant analysis via mixed integer programming (DAMIP) for classification. By utilizing the reserved judgment region, it allows the classifier to delay making decisions on ‘difficult-to-classify’ observations and develop new classification rules in later stage. We apply the framework to four real-life medical problems: 1) Patient readmissions: identifies the patients in emergency department who return within 72 hours using patient’s demographic information, complaints, diagnosis, tests, and hospital real-time utility. 2) Flu vaccine responder: predicts high/low responders of flu vaccine on subjects in 5 years using gene signatures. 3) Knee reinjection: predicts whether a patient needs to take a second surgery within 3 years of his/her first knee injection and tackles with missing data. 4) Alzheimer’s disease: distinguishes subjects in normal, mild cognitive impairment (MCI), and Alzheimer’s disease (AD) groups using neuropsychological tests. In the second topic, we first investigate multi-objective optimization approaches to determine the optimal dose configuration and radiation seed locations in brachytherapy treatment planning. Tumor dose escalation and dose-volume constraints on critical organs are incorporated to kill the tumor while preserving the functionality of organs. Based on the optimization framework, we propose a non-linear optimization model that optimizes the tumor control probability (TCP). The model is solved by a solution strategy that incorporates piecewise linear approximation and local search. In the third topic, we study optimal strategies for public health emergencies under limited resources. First we investigate the vaccination strategies against a pandemic flu to find the optimal strategy when limited vaccines are available by constructing a mathematical model for the course of the 2009 H1N1 pandemic flu and the process of the vaccination. Second, we analyze the cost-effectiveness of emergency response strategies again a large-scale anthrax attack to protect the entire regional population.

<p> Complexity of the capital intensive oil and gas portfolio investments is continuously growing. It is manifested in the constant increase in the type, number and degree of risks and uncertainties, which consequently lead to more challenging decision making problems. A typical complex decision making problem in petroleum exploration and production (E&amp;P) is the selection and prioritization of oilfields/projects in a portfolio investment. Prioritizing oilfields maybe required for different purposes, including the achievement of a targeted production and allocation of limited available development resources. These resources cannot be distributed evenly nor can they be allocated based on the oilfield size or production capacity alone since various other factors need to be considered simultaneously. These factors may include subsurface complexity, size of reservoir, plateau production and needed infrastructure in addition to other issues of strategic concern, such as socio-economic, environmental and fiscal policies, particularly when the decision making involves governments or national oil companies. Therefore, it would be imperative to employ decision aiding tools that not only address these factors, but also incorporate the decision makers&rsquo; preferences clearly and accurately. However, the tools commonly used in project portfolio selection and optimization, including intuitive approaches, vary in their focus and strength in addressing the different criteria involved in such decision problems. They are also disadvantaged by a number of drawbacks, which may include lacking the capacity to address multiple and interrelated criteria, uncertainty and risk, project relationship with regard to value contribution and optimum resource utilization, non-monetary attributes, decision maker&rsquo;s knowledge and expertise, in addition to varying levels of ease of use and other practical and theoretical drawbacks. These drawbacks have motivated researchers to investigate other tools and techniques that can provide more flexibility and inclusiveness in the decision making process, such as Multi-Criteria Decision Making (MCDM) methods. However, it can be observed that the MCDM literature: 1) is primarily focused on suggesting certain MCDM techniques to specific problems without providing sufficient evidence for their selection, 2) is inadequate in addressing MCDM in E&amp;P portfolio selection and prioritization compared with other fields, and 3) does not address prioritizing brownfields (i.e., developed oilfields). This research study aims at addressing the above drawbacks through combining three MCDM methods (i.e., AHP, PROMETHEE and TOPSIS) into a single decision making tool that can support optimal oilfield portfolio investment decisions by helping determine the share of each oilfield of the total development resources allocated. Selecting these methods is reinforced by a pre-deployment and post-deployment validation framework. In addition, this study proposes a two-dimensional consistency test to verify the output coherence or prioritization stability of the MCDM methods in comparison with an intuitive approach. Nine scenarios representing all possible outcomes of the internal and external consistency tests are further proposed to reach a conclusion. The methodology is applied to a case study of six major oilfields in Iraq to generate percentage shares of each oilfield of a total production target that is in line with Iraq&rsquo;s aspiration to increase oil production. However, the methodology is intended to be applicable to other E&amp;P portfolio investment prioritization scenarios by taking the specific contextual characteristics into consideration.</p>

published_or_final_version / Industrial and Manufacturing Systems Engineering / Doctoral / Doctor of Philosophy

No description available.

CIVINS / This project is concerned with evaluation of the merits of competing technological concepts and managing the R&D investments needed to bring them to fruition. A weakness of the current R&D process is that arguments regarding who should receive funding come from a concept's proponents, who are usually interested and biased. More objective evaluation methods are needed. As part of this work it is important to understand better how qualified experts evaluate technological concepts. Can a probabilistically formulated method of integrating knowledge of various performance attributes provide better understanding of the likely performance of a technological concept? This is the question of interest. / CIVINS

Following the 9/11 terrorist events in the United States, the Mexican Navy developed strategies designed to prevent similar attacks on the strategic facilities located in the Campeche Sound in the Gulf of Mexico. The Sound is of great economic importance because more than 83 percent of the petroleum produced in Mexico is extracted from that area. This also makes it a key potential target for international terrorists. This research analyzed and evaluated the Mexican Navy's allocation of surveillance and interdiction resources assigned to the Campeche Sound. The data was obtained via an agent-based simulation, implemented through the use of the software program Map Aware Non-uniform Automata (MANA). The simulation model includes the presence of terrorist boats attacking oil platforms, the Navy resources in the area, service-provider ships in the Sound, and fishing boats that often penetrate into the Sound's exclusion and prevention zones. From the study is concluded that: the most important threat factor in the scenarios is the speed of the enemy boats; and, with its broad surveillance and communication capabilities, the HAWKEYE is the most important navy resource in the area. The results also provide an operational guide to allocate the Navy units in the Campeche Sound.

Unmanned aerial vehicles (UAVs) will be a critical part of the U.S. Army's Future Force. The Future Force will be a highly mobile, network enabled family of systems with integrated sensors and precision munitions. The Future Force will rely heavily on UAVs to provide eyes on the battlefield. These eyes will trigger the deployment of precision munitions by other platforms, and possibly by UAVs themselves. To provide insight into how the numbers and capabilities of UAVs affect a Future Force Combined Arms Battalionâ s (CABâ s) ability to secure a Northeast Asia urban objective, a simulation was built and analyzed. 46,440 computational experiments were conducted to assess how varying the opposing force and the numbers, tactics, and capabilities of UAVs affects the CABâ s ability to secure the objective with minimal losses. The primary findings, over the factors and ranges examined, are: UAVs significantly enhance the CABâ s performance; UAV capabilities and their tactics outweigh the number of UAVs flying; battalion level UAVs, especially when armed, are critical in the opening phases of the battle, as they facilitate the rapid attrition of enemy High Pay-off Targets; and, at least one company level and a platoon level UAV enhances dismounts survivability later in the battle.