Thesis: S.M. in Engineering and Management, Massachusetts Institute of Technology, System Design and Management Program, 2018. / "June 2018." Cataloged from PDF version of thesis. / Includes bibliographical references (pages 57-60). / Gifted and talented students with coexisting learning disabilities, also known as twice-exceptional, are increasingly recognized in U.S. schools. This increasing awareness needs to be met with improved legal protection, better methods for identification and optimized teaching strategies for the unique needs of these students. For this thesis, literature from a range of disciplines including education, cognitive science, and psychology regarding twice-exceptionality is examined, with a specific focus on gifted students with language-based learning disabilities like dyslexia, along with commonly comorbid factors such as ADHD and anxiety. The challenges of these students are also modeled using human centered design tools. Personas of archetypal twice-exceptional students are presented to create empathy for them and awareness of their unmet needs. Design frameworks are examined that aim to improve education universally for all students. Research strongly suggests that twice-exceptional students are under-identified and underserved in our schools and that comprehensive, individualized teaching strategies are necessary in order for them to reach their full potential. Teaching methods are outlined that simultaneously highlight strengths and accommodate the challenges of this important group of gifted learners. / by John Francis Stillman. / S.M. in Engineering and Management

Thesis: S.M. in Engineering and Management, Massachusetts Institute of Technology, System Design and Management Program, 2018. / Cataloged from PDF version of thesis. / Includes bibliographical references (pages 56-58). / This paper focuses on designing novel ways to alleviate human and economic impacts caused by weather and climate disasters such as droughts and cyclones. Natural disasters are becoming apparent and continue to grow in number, intensity, and impact. Authorities, organizations and community groups who focus on rebuilding and relief efforts are constantly facing challenges in redevelopment effort, environmental hazards, health care and funding support to help communities become recover and be more resilient. When dealing with aftermath due to natural disaster the communities do have heightened sense awareness and come together to provide the necessities of rebuilding infrastructure. There are short-term actions, such as an evacuation based on the weather forecasting. Can a system that properly communicates with all affected stakeholders to be prepared for the natural disaster. The implemented system takes the appropriate actions thereby by reducing the human and economic impacts. This precious window of opportunity time between the forecast and actual natural disasters is regularly overlooked which affects the recovery and resilience process. This thesis explains how to design a holistic system that can lessen the risk of natural disaster with a system for forecasting, automatic trigger responses and disburse required funding when certain threshold conditions are met prior to natural disasters. The proposed framework takes into consideration of blockchain technologies that are at the relatively early stage of development. The objectives are to develop novel early funding mechanism and explained using conceptual architecture with private blockchain and smart contracts that can be designed to automatically execute early funding mechanism when the natural hazard thresholds are reached. / by Suresh G. Rajan. / S.M. in Engineering and Management

Thesis: S.M. in Engineering and Management, Massachusetts Institute of Technology, System Design and Management Program, 2018. / Cataloged from PDF version of thesis. / Includes bibliographical references (pages 84-86). / Cybersecurity is a growing research area with direct commercial impact to organizations and companies in every industry. With all other technological advancements in the Internet of Things (IoT), mobile devices, cloud computing, 5G network, and artificial intelligence, the need for cybersecurity is more critical than ever before. These technologies drive the need for tighter cybersecurity implementations, while at the same time act as enablers to provide more advanced security solutions. This paper will discuss a framework that can predict cybersecurity risk by identifying normal network behavior and detect network traffic anomalies. Our research focuses on the analysis of the historical network traffic data to identify network usage trends and security vulnerabilities. Specifically, this thesis will focus on multiple components of the data analytics platform. It explores the big data platform architecture, and data ingestion, analysis, and engineering processes. The experiments were conducted utilizing various time series algorithms (Seasonal ETS, Seasonal ARIMA, TBATS, Double-Seasonal Holt-Winters, and Ensemble methods) and Long Short-Term Memory Recurrent Neural Network algorithm. Upon creating the baselines and forecasting network traffic trends, the anomaly detection algorithm was implemented using specific thresholds to detect network traffic trends that show significant variation from the baseline. Lastly, the network traffic data was analyzed and forecasted in various dimensions: total volume, source vs. destination volume, protocol, port, machine, geography, and network structure and pattern. The experiments were conducted with multiple approaches to get more insights into the network patterns and traffic trends to detect anomalies. / by Lwin P. Moe. / S.M. in Engineering and Management

Thesis: S.M. in Engineering and Management, Massachusetts Institute of Technology, System Design and Management Program, 2018. / Cataloged from PDF version of thesis. / Includes bibliographical references (pages 70-74). / Artificial intelligence - making machines intelligent - is a methodology to build, train, and run machines that are capable of making decisions on its own. Artificial intelligence technologies are gaining significant adoption across a wide range of activities in an organization across different industries. This is fueled by increasing focus on data-driven decision-making methods for all kind of tasks (external or internal) in an organization. Venture capital industry - traditional sub-segment of financial services industry - works heavily on human interactions and relationships. Venture capital investments are considered high-risk, high-return asset class. Venture investment decision-making could be optimized by machine learning applied to previous deals, company data, founder data, and more. It is quite possible that a system could analyze founder personalities, company metrics, and team attributes and improve venture capitalist's decision-making. This thesis is an attempt to analyze and breakdown venture capitalist decisions and understand how Artificial Intelligence tools and techniques could be utilized by VCs to improve decision-making in venture capital. By focusing on the decision-making involved in the following eight value chain areas of a venture capital firm - deal sourcing, deal selection, valuation, deal structure, post-investment value added, exits, internal organization of firms, and external organization of firms, we could discover the extent to which artificial intelligence tools and techniques could be used to improve human decision-making in the venture capital industry. Subsequently, we could also identify how artificial intelligence could be practically used in such decision-making scenarios and also the benefits and associated risks involved in using artificial intelligence system in venture capital decision-making. / by Chahat Jain. / S.M. in Engineering and Management

Thesis: S.M. in Engineering and Management, Massachusetts Institute of Technology, System Design and Management Program, 2018. / Cataloged from PDF version of thesis. / Includes bibliographical references (page 51). / There is an increasing trend in SaaS businesses to offer products and services via subscription licensing. According to a Gartner Study', 80 percent of software vendors will change their business model from perpetual licensing to subscription pricing by 2020. This implies that existing software companies will need to transition to subscription licensing to continue to stay relevant in the market. Strategy change of this nature is not straightforward, and requires a lot of key departments to come together to execute it well. This thesis focuses on understanding the impact of the shift to subscription licensing on external customers and internal departments (sales, marketing, product management & finance) of organizations that make such a change. Reactions of external customers are analyzed via a case study on Tableau, which recently moved to a subscription based model for its Business Intelligence (BI) software. Senior leaders from various SaaS companies were interviewed to understand the organizational structure required to implement a successful change to subscription licensing. Finally, this study provides some actionable insights for organizations that are planning to switch to a subscription based pricing strategy. / by Shweta Jindal. / S.M. in Engineering and Management

Thesis: S.M. in Engineering and Management, Massachusetts Institute of Technology, System Design and Management Program, 2018. / Cataloged from PDF version of thesis. / Includes bibliographical references (pages 65-66). / A key benefit of the DSM representation is that it gives a visual interpretation of relationships between its elements. The array format allows us to sort the elements using clustering algorithms which try to group the relationships into modules which are as independent as possible. There are a number of clustering algorithms available which may each end up sorting the DSMs differently using different objectives, for example, activities in a time-based DSM can be sequenced to reduce iterations or to improve concurrency. However, most of these algorithms take a deductive approach which results in only one 'optimal' output. If an abductive approach is used instead, multiple solutions can be generated for the user to evaluate, some which may provide insight on useful configurations that he or she may have overlooked. In electrical engineering, we often make use of transforms to convert time domain signals into frequency domain signals in order to glean additional information which may not have been initially apparent. In this respect, using a frequency domain transform on a DSM matrix gives us additional insights into the relationships represented. An example of one such insight would be into the sorted-ness of a DSM to which module cuts can be defined. By applying a frequency transform to a pixel representation of the DSM and examining the transform coefficients, we gain an understanding of what image patterns exist in the DSM. Rules pertaining to these coefficients could then be defined which would classify a DSM as well sorted (with the dependencies being grouped up) or being unsorted (with the dependencies being scattered). This thesis demonstrates the above technique to rank each permutation of an 8x8 matrix on their conformance to certain rules or behaviors in order to filter out useful configurations in an abductive approach. When comparing the highest-ranking hypotheses against the optimal result from other clustering and sequencing algorithms, this algorithm performed on par with them to reduce external dependencies and iterations respectively. The frequency based scoring was also shown to be a useful metric when determining the optimal module cut of a system. / by Jonathan Ho Jun. / S.M. in Engineering and Management

Thesis: S.M. in Engineering and Management, Massachusetts Institute of Technology, System Design and Management Program, 2018. / Cataloged from PDF version of thesis. / Includes bibliographical references (page 68). / System dynamics modelling is used to explore varying levels of concurrency in a typical design-build-produce project introducing a new product. Faster product life-cycles and demanding schedules have introduced the importance of beginning downstream work (build/manufacturing) while upstream work (design) is incomplete. Conceivably, this project concurrency improves project schedule and cost by forcing rework to be discovered and completed earlier in the project life. Depending on the type of project, some design errors may only be discoverable once the build phase has begun its work. Namely, systemic errors and assembly errors that cannot be easily discovered within the design phase. Pushing build activity earlier in the project allows the rework to be discovered earlier in the project, shortening the overall effort required to complete the project. A mathematical simulation, created using Vensim@ system modeling software, was created by James Lyneis to simulate two-phase rework cycles. The model was tuned to match data based on a disguised real project. Various start dates (as a function of project percentage complete) for downstream phases were explored to find optimal levels of concurrency. Project types were varied by exploring three levels of "rework discoverable within the design phase" to cover a range of project types. The simulation found that for virtually all project types, significant schedule and effort benefits can be gained by introducing the downstream phase as early as 30% to 40% into the project progress and ramping downstream effort over an extended period of time. / by David Garufi. / S.M. in Engineering and Management

Thesis: S.M. in Engineering and Management, Massachusetts Institute of Technology, System Design and Management Program, 2018. / Cataloged from PDF version of thesis. / Includes bibliographical references (pages 39-40). / There has been significant research concerning the investment case for Environmental, Social and Governance Funds (ESG), however research into how these funds are constructed has been less studied. The purpose of this study is not to investigate the risk return case for ESG funds. Instead, this study will focus on the uniqueness of construction, and underlying assets of ESG specific funds. The majority of ESG classified investing is done through fund firms who self willingly vet their existing funds to ESG guidelines. It is more elocutionary, rather than a focused construction methodology. The hypothesis of this study is that funds created specifically for ESG investing are built on this same methodology, and are adapted from an existing fund very similar to the S&P 500. To test for uniqueness, large cap US equity ESG funds were compared against how many of the underlying assets were shared with the S&P 500. Signals found heavy overlap. However when looking at how the underlying assets are weighted in the fund verse the S&P 500, differences become more pronounced. Interestingly in the aggregate, the portion of the ESG funds dedicated to stocks that are not included in the S&P 500 were not that significant. There are several funds that are constructed with very different underlying assets than the S&P 500 Index, and funds that are very similar. This study then investigated how much of the underlying assets of each fund differed from the S&P 500 by adjusting the weights of just the underlying assets which it shares with each fund to measure the effect of dilution from the removed "non-ESG" compliant stocks. The resulting increase in overlap was significant for several individual funds, but modest for all funds. Then this study sampled to find if there is more overlap with different common index funds. Interestingly, there was often a higher overlap with the S&P 500 than with a fund's stated benchmark such as the Russell 1000 or Russell 1000 Value Index. Finally this study looked for correlations between the 3 month, 1 year, 3 year and Morningstar ESG peer performance percentiles. Modest correlations were found slightly favoring funds which were more similar to the S&P 500. Then correlations between each fund's management fee and similarity in the underlying assets were tested. There is evidence that the more unique the fund is, the higher the management fee. However, there is no evidence of correlation between the fund's management fee and the fund's Morningstar ESG score. The take away from this study is that some funds are very similar to index funds, like the S&P 500, while other funds have very little in common with standard index funds. There was significant overlap in the underlying assets and the S&P 500, however there was also significant differences in how the underlying assets were weighted. There was not a one to one exchange with a non-ESG compliant underlying asset with another asset with similar characteristic but was ESG compliant. / by John Y. Gilmore. / S.M. in Engineering and Management

Thesis: S.M. in Engineering and Management, Massachusetts Institute of Technology, System Design and Management Program, 2018. / Cataloged from PDF version of thesis. / Includes bibliographical references (pages 231-237). / The effects of Sequestration on U.S. Defense spending in the mid-2010s serve as a reminder that complex defense weapons systems are funded by a finite budget. These defense systems have become increasingly complex over time, and with that complexity has come substantial cost. The ability to afford those systems, many of whose burgeoning lifecycle costs have far exceeded initial estimates, has been strained. The Nunn-McCurdy Act is the means of notifying Congress of cost overruns in major defense development programs, and often sets in motion program terminations. But it is, however, an incomplete means of managing a portfolio of systems. It addresses affordability, but does not assess the utility of the subject assets to the stakeholders, either as standalone assets or as part of a synergistic collection. Utility-at-cost provides a more useful figure of merit that can be evaluated objectively and unemotionally, not just in a static context, but over a range of possible future contexts-and not only for a single system, but also a collection of disparate systems and systems of systems. The Portfolio-Level Epoch-Era Analysis for Affordability (PLEEAA) method (Vascik, Ross, and Rhodes, 2015, 2016) builds upon established analytical techniques including Multi-Attribute Tradespace Exploration, Epoch-Era Analysis, and the Responsive Systems Comparison framework. It notably introduced elements of Modern Portfolio Theory, which hitherto were constrained to portfolios of financial assets like stocks and bonds. This research illustrates the general applicability of PLEEAA by exploring two case studies, the U.S. Air Force airborne Intelligence, Surveillance and Reconnaissance portfolio, and the U.S. space-based geospatial intelligence portfolio inclusive of both government-owned and commercial assets. On the whole, these case studies are "top-down" in nature, levying emerging and potentially disruptive technologies on the asset mix. A more rigorous analytical method would be to conduct the "bottom-up" or "as-is" case using only established assets, and compare the two results. Such an approach could illustrate the incremental and potentially synergistic behavior of new assets introduced to the portfolio design problem. / by Jason W. Dieffenbacher. / S.M. in Engineering and Management

Thesis: S.M. in Engineering and Management, Massachusetts Institute of Technology, System Design and Management Program, 2018. / Cataloged from PDF version of thesis. / Includes bibliographical references (page 66). / Corporations are increasingly utilizing corporate venture capital (CVC) as a significant component of their external innovation strategy. Over the past several years, these CVCs have grown to contribute a large percentage of all startup funding in the US. The growing role of CVCs in the innovation ecosystem presents pressing questions around the structures, objectives, and stability of this particular source of funding. After several decades of CVC history, nearly all CVCs have converged onto the dual objective of investing for both strategic and financial returns. It is the existential need to return strategic value back to the parent corporation that separates CVCs as distinct from institutional venture capital (VC) firms. While the survivability and growth of institutional VCs depend solely on financial return performance, the survivability and growth of CVCs depend on demonstrations of both a respectable financial return, as well as relevant and significant strategic returns. This research explores and examines the capture and measure of strategic value in CVC investments through a series of interviews with prominent CVC units representing a cross section of various industries. A framework for characterizing four taxonomies of strategic investment objectives is proposed and used to landscape a sample of CVCs in order to determine whether the capture of strategic value in CVCs is emergent from the system design of a CVC's structure, practices, and organizational linkages. A survey on how CVCs measure direct and indirect strategic value revealed that the vast majority of CVCs were unable to, or do not attempt to measure the performance of this primary investment objective. Both quantitative and qualitative treatments were given to the analysis of the research data on the structures, practices, and strategies related to value capture in strategic VC investments. This research found a wide range of approaches towards capturing strategic value in CVC investments. However, the measurement of such value remains elusive. Very few instances of actual measurement of strategic value were observed, which paints a picture of a significant funding source of US innovation largely unjustified by the lack of performance measurements on existential investment objectives. / by Timothy Chiang. / S.M. in Engineering and Management