Thesis: S.M. in Engineering and Management, Massachusetts Institute of Technology, System Design and Management Program, 2018. / This electronic version was submitted by the student author. The certified thesis is available in the Institute Archives and Special Collections. / Cataloged from student-submitted PDF version of thesis. / Includes bibliographical references (pages 121-123). / The purpose of this work is to examine the "The Cluster Approach" -- the humanitarian response coordination strategy adopted by the Inter-Agency Standing Committee (IASC) following the 2005 'Humanitarian Response Review' -- through the lens of systems thinking and develop potential system architecture representations to explore how the coordination mechanism can enhance complementarity, partnerships, and collaboration among humanitarian actors. The qualitative analysis of "The Cluster Approach" through system architecture principles strongly suggests, that indeed, the framework -- as currently envisioned by the IASC and the humanitarian community -- can be described and illustrated as a structured and architected system. In addition, the analysis demonstrates that the system architecture visualization can help (1) validate the existing framework and (2) design new variants to improve and strengthen the formal and functional relationships while leveraging the underlying organizational platform of the IASC's constituent membership. The analysis also suggests that visualizing the elements of the system as well as the interrelationships among response organizations, actors, and the transactions between these through system architecture principles -- reasoned and guided by holistic thinking -- can be useful and consequential to manage complexity and reduce ambiguity of the IASC's humanitarian system. Finally, extensions of this research to (1) design critical coordination priorities, (2) incorporate more architectural flexibility to manage exceptions, and (3) improve situational awareness of actors to adjust behaviors can hopefully lead to more effective and socially meaningful humanitarian response efforts. / by John F. Barresi. / 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 119-120). / This thesis presents an approach for the identification of opportunities to improve the value of new infrastructures through flexibility. This approach applies to the very early design phase of a new system, where architectural decisions have to be taken under the highest amount of uncertainty. Because the value of optionality increases with uncertainty, it is in this phase that flexibility has the highest potential to positively impact the value of a project. The proposed approach is centered around a list of decisions, common to almost every infrastructure, that can lead to flexible or inflexible systems, and a set of criteria that allows us to make an informed guess of which flexible design opportunities are likely to be valuable by looking at characteristics of the uncertainties. The identified flexible design opportunities are quantified using spreadsheet-based Monte Carlo simulations and optimization. Two case studies demonstrate by example this approach: a European high-latitude space launch complex for satellite constellations in polar orbits, and the Italian strategy to provide natural gas to Sardinia via Small-Scale Liquefied Natural Gas (LNG) infrastructure. The space launch complex case shows that, in presence of market uncertainty, a flexible infrastructure that can support the implementation of different launchers (solid, liquid, or hybrid-motor rockets) lead to a project with higher Expected Net Present Value (ENPV) than an inflexible infrastructure committing upfront to one launcher technology, with the additional benefit of aligning the interests of a hypothetical public-private partnership. The LNG for Sardinia case demonstrates how the combination of the flexibility of capacity expansion in small increments and the flexibility of networking the island with the mainland using a gas power plant leads to a higher ENPV and better Value at Risk than an optimized inflexible infrastructure. This case also introduces a view of the flexibility of networking systems (or sites within a system) to divert excess capacity as an alternative to a reversible capacity expansion, which is rarely available for infrastructures. Both the approach for the identification of flexible design opportunities and the new perspective offered here on the flexibility of networking should be investigated further in a promising domain excluded from the scope of this work: decentralized infrastructures. / by Davide Lasi. / S.M. in Engineering and Management

Thesis: S.M. in Engineering and Management, Massachusetts Institute of Technology, School of Engineering, System Design and Management Program, 2017. / Cataloged from PDF version of thesis. / Includes bibliographical references (pages 60-65). / Although the use of anti-neoplastic chemotherapy provides benefit to patients with both malignant and non-malignant diseases, the use of these agents can be at times associated with safety concerns for both patients and the healthcare workers that administer the medication. In order to mitigate the risks or hazards that are identified there are several potential tools to consider. The tool considered for this thesis will be applying a System Theoretic Accident Model and Processes (STAMP). STAMP is used to investigate the safety of complex systems involving humans, organizations, computers, and other equipment. STAMP has the advantage of facilitating the understanding of highly complicated environments where traditional safety techniques become too costly and cumbersome and hence less efficient. "In the traditional causality models, accidents are considered to be caused by chains of failure events, each failure directly causing the next one in the chain" (Leveson, Engineering a Safer World, 2011). This view is rather different from the perspective taken by STAMP. In STAMP, accidents arise from complex processes involving, not just component failures and faults, but also system design errors, unintended component interactions, human errors, management oversight inadequacies, and more (Leveson, 2011). This thesis presents the "control structure" component of STPA as derived from inputs from healthcare workers particular to the Dana-Farber Cancer Institute. The suggested control structure will ultimately lay the groundwork for future work on a detailed Systems-Theoretic Process Analysis (STPA) and generate specific recommendations to help address the identified risks and hazards in addressing patient safety issues. / by Harding J. Hall. / S.M. in Engineering and Management

Thesis: S.M. in Engineering and Management, Massachusetts Institute of Technology, School of Engineering, System Design and Management Program, 2017. / Cataloged from PDF version of thesis. / Includes bibliographical references (pages 76-80). / Rapid disease diagnosis is critical during infectious disease outbreaks to enable early intervention measures and minimize risk of disease transmission. Recent outbreaks in low-resource settings have highlighted challenges with traditional laboratory-based diagnostic approaches including a dependence on supporting infrastructure and highly trained users. Limitations of laboratory-based devices often result in geographical separation of labs from cases creating delays and barriers for diagnosis. There is increasing interest in the use of point-of-care diagnostics during outbreaks to enable more dispersed field diagnostic approaches and improve accessibility of testing. Point-of-care diagnostics, however, are often less accurate than laboratory-based tests, which can make them a less trusted option. This thesis explores the possibility that accessible, less accurate point-of-care devices could enable more efficient containment of disease outbreaks compared to current practices that employ expensive, and often distant laboratory-based tests. Although the benefit of point-of-care devices has been discussed anecdotally, little work has been done to quantify the relative impact of point-of-care diagnostics on transmission characteristics during an outbreak. This thesis aims to establish a basic cross-domain simulation model that considers medical, engineering, and societal/cultural factors that contribute to disease outbreak outcomes. The simulation approach is used to assess the trade-off between diagnostic access and accuracy during the 2014 West Africa Ebola outbreak to determine if point-of-care devices could have offered a benefit. A sensitivity analysis is also conducted to assess the potential impact of diagnostics on future outbreaks. Simulation results support the hypothesis that deployment of point-of-care devices to increase accessibility of testing could significantly reduce the number of secondary infections during an outbreak. This finding is shown to be true across outbreaks of varying sizes and transmission characteristics and for devices with varying accuracy performance. / by Ashley L. Whitney. / S.M. in Engineering and Management

Thesis: S.M. in Engineering and Management, Massachusetts Institute of Technology, System Design and Management Program, 2018. / This electronic version was submitted by the student author. The certified thesis is available in the Institute Archives and Special Collections. / Cataloged from student-submitted PDF version of thesis. / Includes bibliographical references (pages 117-123). / The UN predicts that by the year 2030, the global water demand will outstrip supply by 40%. In face of the global water crisis, it is crucial to increase irrigation efficiency in agriculture, which currently consumes 70% of the global freshwater supply. Studies have shown that using precision agricultural technology to control irrigation can reduce water consumption by as much as 20% and increase crop yield by up to 30% in developing countries. Such technologies, however, are inaccessible to millions of small-scale farmers who need them the most because of their prohibitive costs and design intended for large-scale farming businesses. To address this technological gap, social enterprise SoilSense delivers affordable and robust IoT soil sensor systems to small-scale farmers, empowering them to irrigate more efficiently by providing data on when and where to irrigate based on soil measurements. This study analyzes existing literature on irrigation and soil sensor technology and applies a human-centered design approach to understand the needs of an underserved user group: smallscale and medium-scale avocado farmers. By engaging these farmers and subject matter experts in the field, key insights are drawn on the nuances of avocado cultivation, challenges in irrigation and water management, and the use of technology and data analytics in farming. This user research highlights the small-scale and medium-scale farmers' pain points and their vision for how technology could improve their operations. In addition to informing the iterative design of the SoilSense system prototype and business model, this study also endeavors to help address the global water crisis through continuous innovation and advancement in IoT agricultural technology. / by Julia C. Wong. / S.M. in Engineering and Management

Thesis: S.M. in Engineering and Management, Massachusetts Institute of Technology, System Design and Management Program, 2019 / Cataloged from PDF version of thesis. / Includes bibliographical references (pages 108-114). / Many companies are increasing their focus on Artificial Intelligence as they incorporate Machine Learning and Cognitive technologies into their current offerings. Industries ranging from healthcare, pharmaceuticals, finance, automotive, retail, manufacturing and so many others are all trying to deploy and scale enterprise Al systems while reducing their risk. Companies regularly struggle with finding appropriate and applicable use cases around Artificial Intelligence and Machine Learning projects. The field of Artificial Intelligence has a rich set of literature for modeling of technical systems that implement Machine Learning and Deep Learning methods. This thesis attempts to connect the literature for business and technology and for evolution and adoption of technology to the emergent properties of Artificial Intelligence systems. The aim of this research is to identify high and low value market segments and use cases within the industries, prognosticate the evolution of different Al technologies and begin to outline the implications of commercialization of such technologies for various stakeholders. This thesis also provides a framework to better prepare business owners to commercialize Artificial Intelligence technologies to satisfy their strategic goals. / by Siddhartha Ray Barua. / S.M. in Engineering and Management / S.M.inEngineeringandManagement Massachusetts Institute of Technology, System Design and Management Program

Thesis: S.M. in Engineering and Management, Massachusetts Institute of Technology, System Design and Management Program, September, 2020 / Cataloged from the official version of thesis. / Includes bibliographical references (pages 65-67). / Calibration of Magnetic Flux Leakage (MFL) In-line Inspection (ILI) tools is an important part of the overall pipeline integrity management process. Over-called or under-called corrosion features can have significant impacts on safety and resource management. This thesis examines methods for improving the Validation and Calibration processes using Bayesian Inference. The focus is on improving the tolerance that is applied to undug features to optimize the execution of risk-based repairs. A simulated data set was generated, with two separate categories, one which represents tool performance on basic features and another for challenging features. The calculated parameters of [alpha], [beta], and [sigma], were calculated using a Bayesian model leveraging a Markov Chain Monte Carlo simulator. The [sigma] parameter is used to determine the appropriate tolerance to apply and was compared with a [sigma] calculated via the method recommended by API 1163. Results from the example data set show that in challenged situations, the Confidence Level of the tool performance can be increased from 89% to 95% and the mean average error can be decreased using the Bayesian Inference model. Opportunities to use the methods outlined to improve other processes in ILI validation are discussed. By appropriately updating the likelihood used in the Bayesian model with dig data, the tolerance can more accurately represent the undug features and risk management decisions can be conducted accordingly. / by Jeffrey Liang Lee. / S.M. in Engineering and Management / S.M.inEngineeringandManagement Massachusetts Institute of Technology, System Design and Management Program

Thesis: S.M. in Engineering and Management, Massachusetts Institute of Technology, System Design and Management Program, 2019 / Cataloged from PDF version of thesis. / Includes bibliographical references. / Space and water heating account for nearly two-thirds of energy consumption in U.S. homes, and a large contributor to energy costs of end-use residential dwellings. Most home heating systems in the United States are fueled by fossil fuels - natural gas and fuel oil (heating oil) - representing more than 50 percent of all U.S. homes' heating. These heating systems result in higher greenhouse gas emissions than electric heating systems now, and the emissions difference will increase as the grid trends toward lower carbon intensity in the decades ahead. Electrification of residential heating systems, by eliminating site fossil fuel use for heating, provides an important element of ultimately achieving carbon-free buildings. The objective of this research is to analyze the heating load of end-use residential dwellings. The research for this thesis achieves this by first conducting a survey of energy usage profile of some residents in Boston, Massachusetts and Houston, Texas. / It then applies a thermal model to simulate building heat load, which was used in developing an electrification cost model to verify and validate the case for electrification of residential dwellings. Thermal models were developed for two cities, Boston and Houston, having contrasting winter weather and electricity rates. The model simulated heat load demand and energy outputs from heat pumps in both cities and analyzed resulting data and potential tradeoffs compared with electric resistance and gas furnace heating systems. Results show that heat in residential dwellings using electric air-source heat pumps (ASHPs) is more cost-effective and energy efficient compared with other heating systems. Model analyses indicate that heat demand in residential dwellings, which increase as outside temperature decreases due to heat loss, is disproportionately higher at low temperatures because the performance of ASHPs drops with outside temperature. / However, ASHP performance is higher in Houston compared to Boston due to milder winter temperatures in the former. And the "balance point" between heat load and energy output decreases as capacity of ASHP increases. / by Tolu O. Sodeinde. / S.M. in Engineering and Management / S.M.inEngineeringandManagement Massachusetts Institute of Technology, System Design and Management Program

Thesis: S.M. in Engineering and Management, Massachusetts Institute of Technology, System Design and Management Program, 2019 / Cataloged from PDF version of thesis. / Includes bibliographical references (pages 89-92). / Pharmaceutical R&D involves significant uncertainty, including high attrition rate and long time for a project to progress from a target identification phase to commercial launch. Despite this uncertainty, senior leaders must make decisions about R&D portfolio, the impact of which may not be observable for many years. Hence the purpose of this thesis is to understand the current state of Pharmaceutical R&D portfolio management and identify the gaps in R&D portfolio management research. The literature survey revealed that though there are many qualitative and quantitative approaches for the portfolio management of projects in the development phase (i.e. from pre-clinical to phase 3.), the topic of portfolio management in the drug discovery phase (i.e. target validation to lead optimization) have not been well covered in the literature. Hence the problem statement of this thesis is to develop a portfolio management approach for drug discovery. / Portfolio management in pharmaceutical drug discovery space is not only a mathematics problem but also a representation problem in terms of activities, resources, decisions, dependencies, and uncertainties. There is something about the nature of the scope of early research in pharma which makes it different from downstream phases and respective parallels in other sectors. Improved representation can lead to improved prediction in drug discovery phase. Hence as the first step, structured survey was conducted to listen to insights from an experienced professional in the drug discovery domain at NIBR (Novartis Institute of Biomedical Research) to build the required understanding about discovery phase. The survey results helped in identification of the biology, chemistry, medical, marketing, and strategy factors generally taken into consideration during drug discovery project prioritization. / Resource allocation is not considered during project prioritization - even though resource allocation determines the cycle-time that in turn influences the probability of project to reach pre-clinical phase. Proposed semi-quantitative portfolio management approach, which is based on the survey results, incorporates three key aspects of drug discovery project - scope feasibility (science), desirability (market and strategy), and time feasibility (resource allocation and cycle time). Proposed criterion based model for computing scope feasibility and desirability can be uniformly and transparently applied to all the projects across different disease areas and requires discussion between concerned teams to generate required scores. Also, proposed resource allocation model will enable portfolio management teams to generate multiple scenarios (trade spaces) on scope feasibility, time feasibility, and desirability dimensions. / Based on the thresholds, which can be calculated from past data, portfolio team and management in conjunction with other teams such as disease area representatives, chemistry team, marketing etc. can decide the best scenario. The future work needs to focus on validating the proposed portfolio management approaches and models with the real data from past projects in the drug discovery phase in order to enable to organization-wide implementation. / by Ravi Badwe. / S.M. in Engineering and Management / S.M.inEngineeringandManagement Massachusetts Institute of Technology, System Design and Management Program

Thesis: S.M. in Engineering and Management, Massachusetts Institute of Technology, System Design and Management Program, 2019 / Cataloged from PDF version of thesis. / Includes bibliographical references (pages 62-64). / Advances in data science and technology promise to help clinicians diagnose and treat certain conditions. But there are other complex and poorly characterized illnesses for which the drivers and dependent variables are not understood well enough to take full advantage of the copious patient data that may exist. For these diseases new techniques need to be explored to gain better understanding of the nature of the disease, its subtypes, cause, consequence, and presentation. Modern genetics have shown that these diseases often have multiple subtypes, as well as multiple phenotypes as indicated by the new laboratory data. Examples of such diseases include common and important illness such as Type 2 diabetes (T2D) - affecting approximately 30 million Americans, Crohn's Disease - 1 million USA suffers, epilepsy - 3.4 million Americans, and migraines - another 3.2 million in the United States. / Our research explores how machine learning (ML) can be applied to these less well understood complex diseases to improve clinical translation and management. This thesis will discuss how unsupervised machine learning techniques can be used for complex phenotype clustering to identify sub-types of T2D for better clinical management and treatment. T2D is a complex heterogenous disease affecting the world's population at rapidly increasing rates. While clinicians now better understand the heterogeneity of the disease, T2D treatment strategies still remain largely based on populations rather than on a specific patient's subtype. This thesis explores the concept of using data analytics and ML to identify sub-types of T2D as the first step in moving towards precision medicine & treatments. / This thesis includes (a) characterization of T2D as a heterogenous disease, (b) existing research attempts to dissect the disease into sub-types based on phenotypes and gene expressions, and their limitations, (c) phenotype clustering analysis on T2D patients using unsupervised machine learning techniques and MIMIC III database, and (d) analysis of the clusters/subgroups in different ways to understand their clinical significance. With multiple iterations of the clustering experiment, this thesis, (a) provides a good test of concept for sub-classification of T2D patients using unsupervised machine learning techniques such as, clustering and dimension reduction, (b) establishes a data pipeline and clustering model framework to be applied to richer datasets, (c) suggests various experiment design options for further analysis, and (d) establishes a direction for future work including advanced modelling techniques and predictive analytics for complex diseases. / by Priyasha Mehta. / S.M. in Engineering and Management / S.M.inEngineeringandManagement Massachusetts Institute of Technology, System Design and Management Program