Thesis (S.M. in Engineering and Management)--Massachusetts Institute of Technology, System Design and Management Program, 2009. / Cataloged from PDF version of thesis. / Includes bibliographical references (p. 100-102). / The S&P 500 represents a diverse pool of securities in addition to Large Caps. A range of audiences are interested in the S&P 500 forecasts including investors, speculators, economists, government and researchers. The primary objective is to attempt to provide an accurate 3 month and 12 month forecast using the recent credit crisis data, specifically during the time range of 10/2008 - 09/2009. Several methods were used for prediction fit including: Linear Regression, Time Series Models: Autoregressive Integrated Moving Averages (ARIMA), Double Exponential Smoothing, Neural Networks, GARCH, and Bootstrapping Simulations. The criteria to evaluate forecasts were the following metrics for the evaluation range: Root Mean Square Error (RMSE), Absolute Error (MAE), Akaike information criterion (AIC) and Schwartz Bayesian criterion (SBC). But most importantly, the primary forecasting measure includes MAE and Mean Absolute Percentage Error (MAPE), which uses the forecasted value and the actual S&P 500 level as input parameters. S&P 500 empirical results indicate that the Hybrid Linear Regression outperformed all other models for 3 month forecasts with the explanatory variables: GDP, credit default rates, and VIX volatility conditioned on credit crisis data ranges, but performed poorly during speculation periods such as the Tech Bubble. The Average of Averages Bootstrapping Simulation had the most consistent historical forecasts forl2 month levels, and by using log returns from the Great Depression, Tech Bubble, and Oil Crisis the simulation indicates an expected value -2%, valid up to 12 months. / (cont.) ARIMA and Double Exponential smoothing models underperformed in comparison. ARIMA model does not adjust well in the "beginning" of a downward/upward pattern, and should be used when a clear trend is shown. However, the Double Exponential Smoothing is a good model if a steep incline/decline is expected. ARMAX + ARCH/EGARCH performed below average and is best used for volatility forecasts instead of mean returns. Lastly, Neural Network residual models indicate mixed results, but on average outperformed traditional time series models (ARIMA/Double Exponential Smoothing). Additional research includes forecasting the S&P 500 with other nontraditional time series methods such as VARFIMA (vector autoregressive fractionally integrated moving averages) and ARFIMA models. Other Neural Network techniques include Higher Order Neural Networks (HONN), Psi Sigma network (PSN), and a Recurrent Neural Network (RNN) for additional forecasting comparisons. / by Eric Glenn Chan. / S.M.in Engineering and Management

Thesis (S.M.)--Massachusetts Institute of Technology, System Design and Management Program, 2008. / Includes bibliographical references (leaves 116-118). / The partitioning of a system can and will dictate the creative space for a designer or engineer. This thesis will analyze how using a new automotive architecture known as a Front End Module (FEM) can affect a limited specific subset of stakeholders. Through the use of interviews of subject matter experts, literature research and the use of System Design Management tools, an in depth analysis will be done on the FEM and how it affects the craftsmanship, damageability and assembly attributes. It will be shown how the craftsmanship attribute can be improved through the strategic use of FEM's to allow for a feed-forward system where build data are incorporated into upcoming FEM builds. Even with this advantage, the FEM architecture will not negatively impact the damageability attribute or assembly attribute if the proper design cues and strategies are followed. The FEM will also be intensely analyzed using the tools from the MIT SDM program where it will be evaluated as an architecture itself through the specific and targeted intent and beneficiary breakdown. The analysis will also include an Object/Process Mapping analysis where it will be proposed that the true customer of the automotive front end is not the individual that purchased the vehicle but rather the visual society as a whole. Finally, a managerial approach will be taken for the analysis of the inherent and inevitable supplier relationship that is required with using this FEM architecture. Interviews were conducted with two suppliers of OEM's and their common road blocks will be analyzed such as lack of holistic thinking or failure to understand the role of the system integrator. Proposed next steps will be laid out to address these barriers in order to open the communication channels between the supply base and the Original Equipment Manufacturers. / by Vincent R. Mahé. / S.M.

Thesis (S.M.)--Massachusetts Institute of Technology, System Design and Management Program, 2007. / Includes bibliographical references (p. 70-72). / The purpose of this thesis to evaluate the opportunity for service provider entry and of the airborne internet, to analyze the disruptive impact technology used by AirCell and AeroSat has had on the development of an airborne internet, and to identify various stake holders and their value propitiation. The airborne internet has the potential to change the way we fly and spend time when sitting in the plane. In the last fifty years, there has not been much technological advancement in the air traffic control system. Airplane operation still depends on current ground control and radar systems that are very expensive and very difficult to scale. These technologies are also heavily dependant on humans. There have been many technological advancements out side of the aviation industry. Establishing an airborne internet is a tremendous opportunity for everyone. With the help of an airborne Internet, each plane can transmit its identity, location, and also direct video footage that will help Homeland security fight against terrorism. The airborne internet has the ability to connect airplanes not just via a computer on the ground (or via satellite) but directly with each other, relaying information from other planes in an Internet-like fashion. The airborne internet is strongly supported by the Pentagon, FAA and NASA. The U.S. Air Force and FAA are working on defining the architecture of an airborne network and hope to begin actively developing and testing the network itself between 2008 and 2012. According to the FAA, in 2005 there were 10 million flights carrying a total of 660 million passengers in the United States. For the FAA there are a number of merits to working with an airborne internet service provider to continue tests and validate the technical and economic feasibility of an airborne internet. / (cont.) First, there appears to be a substantial market -- in the range of $1b -- for services that require internet connectivity on the air for the commercial airline, air cargo, business jet, and general aviation sector. Second, current alternatives such as satellite solutions and existing air-to-ground solutions fail to meet all the needs of the mass market. Satellite solutions provided by companies such as Inmarsat, Iridium, and Globalstar are priced at a premium and carry an expensive cost structure from the maintenance and investment in orbiting satellites. Airborne Internet service can be offered through three different technologies first, a satellite solution offered by Boeing; second, air-to-ground systems provided by companies such as AirCell; and third, a network of airplane ground -to - air system like AeroSat, all of which are compatible with the planned FAA architecture. Boeing's model is prohibitively expensive; a business model for an airborne internet solution based on a South West Airlines type low cost approach may make an airbome internet more feasible The model would rely on low service fees to promote greater consumer usage, high capacity utilization of ground stations to promote margins, low aircraft equipment costs to help cash flows, and risk/reward sharing with airlines to promote aircraft operator adoption. Assuming that a service provider relied on revenue from non-FAA related services, it could still generate ample margins to support other general FAA applications behind the scenes. The FAA can demonstrate overall support for an airborne internet vision, help attract key players to the ecosystem needed to implement the system, promote usage, and drive required airline ROI. The FAA could also drive the implementation of industry standards required to eventually ensure globally consistent services. / (cont.) However, even with these clear benefits, there are a few key risks that need to be considered and further evaluated. First, this analysis evaluated the economic feasibility of an airborne internet. It does not take into consideration testing or validating the potential network performance from AeroSat's innovative mesh approach in an actual pilot test. Second, more extensive demonstrations will be required to further validate performance and the related cost for the supporting infrastructure. Some key economics like the number of antennae required on aircraft as the network grows should be explored in greater detail after initial simulations. Finally, uncertainty over potential developments of spectrum-free solutions, evolutes of ultra-wideband with potentially disruptive cost structures, could slow the market from adopting a spectrum-based solution. Although this is unlikely given the FAA's current stance on the use of UWB, the issue is worth further research and conversations with the FAA. Accordingly, continued testing, development, and analysis to test feasibility and clarify the key unknowns is recommended. There are a few areas that deserve special attention. First, the target customer composition required to drive the business model should be finalized. The reliability and performance of the mesh-approach is partly dependent on the density of airtraffic in relation to the location of installed ground stations. Second, spectrum requirement issues, including the cost of acquisition and regulatory compliance, need clarification as they strongly impact the business model. Third, the potential magnitude and variability of assumed revenue sources, as well as the timing of cash collections across key customer segments, should be explored. / (cont.) Both of these impact the assumed free-cash-flows generated by the potential business model. Finally the potential terms of airline risk/reward sharing contracts required to equip aircraft with different quantities and types of antennae, need further exploration. Air carriers seem to be moving away from models where they absorb all of the equipment/certification costs - the economic feasiblity of a potential service provider depend on the service provider's ability to offer airlines this service at a reasonably good rate. / by Anand Bhadouria. / S.M.

Thesis (S.M.)--Massachusetts Institute of Technology, System Design and Management Program, 2009. / Cataloged from PDF version of thesis. / Includes bibliographical references (p. 77-79). / This thesis considers the case of Ubiquisys Ltd., a femtocell company, and the ability for its products to provide added services to users in the home. The findings include recommendations for software applications, application delivery, timing, and pricing. The actions that should be taken as a result of this study include a few things in both application architecture as well as long-term strategy. First, the femtocell needs to allow communication between the smartphone, the home network, and other smart devices. Secondly, the femtocell has no self-contained interface to the user, and thus will require platform specific applications sold through the smartphone OS or handset vendor. Lastly, Ubiquisys needs to decide on a strategy, including both place and timing, for rolling out services. Femtocell applications require femtocells, which in the next few years will only reach an installed base several orders of magnitude smaller than the smartphone. It is recommended that mobile advertising based free applications should be provided initially while slowly phasing in paid applications in 2011 and 2012. Through application of system architecture analysis and Design Structure Matrices (DSM) to the current and emerging architectures, this paper provides a template for analyzing ecosystems for home cloud services and content delivery. This is achieved through an in depth analysis of two current product architectures. Information drawn from analysis of these systems is then used to make conclusions and recommendations about how a femtocell can provide the greatest value in the home. / by Kelly Yedinak. / S.M.

Thesis (S.M.)--Massachusetts Institute of Technology, System Design & Management Program, 2000. / Includes bibliographical references (p. 77). / As the diffusion of the Internet reaches the early majority phase, customers, both retail consumers and businesses, are utilizing the Internet in ever increasing numbers for the information search and decision making portions of the sales process. Consumers are online actively searching, researching, selecting, and purchasing products. The most direct output of this process is the end product selection, but many opportunities present themselves for the improvement of market research and corporate product development through the utilization and analysis of the online purchase process. This opportunity is especially applicable to a purchase process in which the customer is not just searching for products by selecting hard constraints around product attributes, but is instead engaged in an online dialogue with a trusted virtual advisor. During this dialogue, the customer reveals their product needs and preferences and the trusted advisor recommends products based upon a maximization of customer utility. If a "perfect" product is available for each customer, then the maximum utility value across all available products will be constantly rising during the online dialogue and all customer needs will be met. Given the reality of product development limitations, ever-changing customer needs, and the introduction of new products and features, consumers are routinely left with unmet needs. This thesis focuses on the development and implementation of a virtual engineer who, after the customer has explicitly given permission, listens to the dialogue between the customer and the trusted virtual advisor. When an unmet customer need is identified through a drop in the maximum calculated utility, the virtual engineer joins the online dialogue. Through the automated generation of contextual questions around the unmet need, the engineer gathers detailed information about the exact nature of the customer's need and translates that need into useful engineering terms. After a large number of customer needs have been collected, the Listening In! process estimates the market share opportunity currently available through the introduction of products that meet the identified needs. The Listening In! process represents a new opportunity in the utilization of the Internet for the improvement of both market research and the traditional product development process. / by Christopher Cyril Mann. / S.M.

Thesis (S.M.)--Massachusetts Institute of Technology, System Design & Management Program, 1999. / "September 1999." / Includes bibliographical references (leaves 30-31). / by Kimberly J. Ryan. / S.M.

Thesis (S.M.)--Massachusetts Institute of Technology, System Design & Management Program, 2002. / Includes bibliographical references (p. 184-187). / The case for proactive market interventions to stimulate accelerated development and adoption of cleaner advanced fuel/vehicle systems continues to gain momentum globally. World population growth, rising national wealth, persistence of auto-centric land development patterns, and the growing popularity of truck-based platforms have reversed the trend toward lighter vehicles, higher fuel efficiencies, and cleaner emissions that began in the 1970's. Since 1984, fuel efficiency has stagnated and even declined as oil prices moved to historically low levels. Lack of consumer demand for fuel economy served to shift investment toward other measures of value such as performance and utility. The convergence of car and truck markets has further impeded progress on the fuel economy front. Industry continues to come under intensifying pressure from international and domestic concerns regarding adverse vehicle emissions impacts on public health, environmental degradation, global climate change, and national security vulnerabilities stemming from dependence on foreign oil. In response, the world's major auto conglomerates have embarked on a variety of strategies to deliver cleaner vehicles to market. Strategies span in-house and strategic partnering efforts across a range of both available and developing technologies in fuels, batteries, fuel cells, electric hybrids, and improvements to current internal combustion engine (ICE) designs. This thesis intends to examine a set of representative technological solution pathways that address two key questions for decision-makers: (1) whether market adoption of advanced fuel/vehicle systems can occur under plausible conditions, and (2) what industry strategies and public interventions can best leverage innovation to achieve the accelerated adoption of technologies beneficial to sustainability goals? To answer these questions, the work employs a system dynamics-based model in order to simulate the complex dynamics surrounding this issue. The model provides a useful framework for comprehending the relative directional impacts of varying industry strategies, public interventions, and external market and cultural forces that affect potential outcomes. The work suggests that plausible adoption scenarios are realizable within a thirty-year time horizon, but that forces deleterious to the innovative capacity of established domestic firms may significantly impede progress. I outline these forces, explain their origins, and recommend industry strategies and public interventions that appropriately address these obstacles. / by Eric C. Cahill. / S.M.

Thesis (S.M.)--Massachusetts Institute of Technology, System Design & Management Program, 2000. / Includes bibliographical references (p. 133-134). / Lean principles focus on employing value added activities to reduce product development cycle time, increase quality, and reduce cost. Lean originated in the automotive industry and has since been centered in the manufacturing domain. Lessons learned on implementing Lean initiatives have been captured by the in the Lean Aerospace Initiative (LAI) and incorporated into the Lean Enterprise Model (LEM) (http://lean.mit.edu/public/index.html). To the author's knowledge, this is the first research effort specifically designed to apply the Lean principles and the Lean Enterprise Model to the aerospace software requirement derivation process. Data supporting this research is the result of a comprehensive two-year research effort involving three detailed case studies with 45 case study interviews, 125 stakeholder surveys collected from ten aerospace software upgrades, feedback from numerous aerospace industry practitioners and Massachusetts Institute of technology (MIT) faculty. Ten aerospace software upgrades were analyzed at both an enterprise level and an organizational level to identify the presence of Lean practices. At the enterprise level, metrics typically used to measure enterprise performance (Flow Time, Stakeholder Satisfaction, Quality Yield, and Resource Utilization) were found to be appropriate for the software requirement process but not adequately implemented. An organizational analysis observed five of the twelve Lean practices as effectively implemented and identified opportunities to implement four more Lean practices. / by Brian J. Ippolito. / S.M.

Thesis (S.M.)--Massachusetts Institute of Technology, System Design and Management Program, 2006. / Includes bibliographical references (p. 57-58). / The software product business reached the $150 billion mark at the end of 2005. The pricing and licensing of new products, maintenance services, services and service maintenance have become an important strategy to deliver smooth and steady revenue growth. The main objective of this thesis work is to observe the trends among various revenue prospects such as product sales, maintenance sales, service sales and service maintenance sales revenue. Both qualitative and quantitative analyses are pursued to achieve the research objectives. The software industry has gone through several transformations from its debut. Initially, it started as service-oriented industry and then transitioned into a product-oriented industry and is currently transforming into a hybrid of both product and service industries. Just as the industry has undergone from a service to a product to a service (more like a hybrid) oriented industry, so too has the product licensing and maintenance licensing scenario been changing. Perpetual licensing has established its strong presence from the very beginning. While both single user and concurrent user licensing models were standard license offerings, concurrent licensing has been eliminated from the enterprise application segment. A usage-based licensing model is gaining popularity but perpetual licensing continues to be the preferred one. Software firms sell a lot of products early on and fail to keep up the same rate of new sales to new customers. From the regression model, it was observed that the product sales as a percent of total sales decreased by 2.6% each year. As the firm grows older, the firm's rate of growth of product sales decreases. Also, product sales as a percent of total sales increased by 0.88 % for each percentage increase in growth of maintenance sales. / (cont.) Software firms are providing various contents that are included in the maintenance service. Maintenance discounting occurs but not comparable to new product discounting. Fix-time promise in maintenance rather than just the response time commitment is an emerging trend. Software firms are pushing premium maintenance services along with subscription-based licensing offerings. It was also observed that maintenance sales as a percent of total sales increased by 2.1% each year. Also, the maintenance sales as a percent of total sales increased by 0.42 % for each percent decrease in the growth of product sales. At the very early stages of the firm, rate of growth of maintenance sales is low and increases as the firm gets older. / by Shivashis Nayak. / S.M.

Thesis (S.M.)--Massachusetts Institute of Technology, System Design & Management Program, 2003. / Includes bibliographical references (p. 202-203). / The automobile embodies complex dynamic system architecture with thousands of components and as many interconnections. The modern day vehicle architecture attempts to balance significant tradeoffs and constraints to achieve the system goals. There are innumerable combinations, which may or may not achieve success. This work proposes a new method for evaluation of complex dynamic system architecture through a hierarchical synthesis of specific qualitative and quantitative tools and methods within a system architecture framework. The proposed methodology is applied to key subsystems of a specific high performance car to assess primarily the merits of the process. Current methods for system architecture definition at the automobile manufacturer utilized for analysis rely primarily on experience-based intuition within an architecting framework. Current system architecture frameworks and the manufacturer's process utilized appear insufficient, as significant issues (often dynamics related) arise in the verification and validation phase of their product development process, requiring change to vehicle architecture. Changes in architecture at this phase of the manufacturer's product development process have significant cost, timing and perhaps functional performance implications. Many system architecture and engineering tools exist to aid architecture definition, but a hierarchy in usage and the interrelationships of the tools are not clearly defined. The proposed solution for rigorous complex dynamic system architecture evaluation includes a four phase hierarchical synthesis of known qualitative and quantitative tools and methods within a holistic system architecture framework. For purposes of this thesis, the proposed evaluation methodology is labeled "CD-SAAM" for Complex Dynamic System Architecture Assessment Methodology. The proposed methodology is a rigorous complement, superimposed on the concept development phase, to the standard product development design process. CD-SAAM mainly combines known system architecting and system engineering framework, principles and tools. Application of CD-SAAM to a high performance car's powertrain and chassis system architecture's second level form and function decomposition, serve to demonstrate many high level conclusions. The hierarchy and synthesis of framework, principles and tools in CD-SAAM provided a valuable and rigorous method to evaluate complex dynamic system architecture. While certain aspects of the proposed methodology appear time-consuming, each step and the overall process serve to greatly improve consistent success with respect to achievement of a system's goals within its constraints. Application of CD-SAAM also underscores the importance and need for explicit design parameter identification and analysis in complex dynamic system architecture assessment. The performance car application also provides insight into the value of DOE RSE methods in architecture assessment, as opposed to its typical region of use in detailed design analysis. Finally, a positive by-product of the analysis includes CD-SAAM's ability to evaluate the consistency and attainability of goals within the given constraints. / by Scott M. Ahlman. / S.M.