Thesis (S.M.)--Massachusetts Institute of Technology, Engineering Systems Division, 2012. / Cataloged from PDF version of thesis. / Includes bibliographical references (p. 91-96). / Mobile traffic has grown substantially over the last few years; a trend which is expected to continue. The chief reasons behind this phenomenon are the availability of better handsets, faster cellular networks and the variety of content available on the internet suitable for consumption on mobile devices. The nature of the traffic is also changing from pure web browsing with latency-tolerant traffic to video, which is becoming the major class of content consumed on mobile devices. This trend, combined with the trend of decreasing prices per GB of data, which constrains the amount of money an operator can spend upgrading its network and they see increasing value in alternative solutions to address this data deluge while managing costs and maintaining customer service. A variety of solutions have been tried by operators based on enhanced charging, traffic engineering and backhaul infrastructure upgrades. Wi-Fi offload is one such promising solution as it addresses the congestion problem where it is most severe because of data consumption by users using streaming video. Cellular spectrum is a scarce and expensive resource for operators, and by allowing them to offload traffic to Wi-Fi networks in unlicensed spectrum they can free cellular spectrum for more valuable applications. Wi-Fi offload has, however, suffered from the incapability to manage seamless handovers and the required interaction of the user to select a Wi-Fi network. This made the process of attaching to a Wi-Fi network very complicated. These limitations have been addressed in recent standards and make the case for Wi-Fi offload more viable and attractive than earlier. At the same time new video optimization techniques such as H.264/SVC which allow the use of multiple streams and channel will allow content providers or distributors to use multiple networks and to scale video seamlessly according to handset capabilities and network conditions. The thesis proposes a solution, based on a set of new Wi-Fi standards and the new H.264/AVC codecs, which leverages a combination of low cost Wi-Fi and high reliability cellular networks to reduce the cost of video transmission while maintaining a comparable QOE for nomadic users. The thesis also enumerates some of the basic procedures that can be supported using the proposed architecture. This new architecture opens new opportunities for existing players in the mobile content ecosystem and adds new players to the ecosystem. The thesis identifies the needs and opportunities for each of the new player and also develops a cost model for streaming video using this solution. / by Amit Limaye. / S.M.

Thesis (M. Eng. in Logistics)--Massachusetts Institute of Technology, Engineering Systems Division, 2004. / Includes bibliographical references (leaves 59-60). / This research examines how an athletic footwear company should establish its new product development and launch process to eliminate wastes in the processes and improve the time to market. Currently, it typically takes an athletic footwear company twelve months to introduce new product samples. Retailers place orders after they see samples, however they will not receive and sell the shoes in their retail stores until six months later. The total process from an idea generated to the time when the final products launch takes eighteen months. While this system is set up due to historical reasons, forward looking management teams in the industry see a lot of inefficiencies in it, especially when athletic footwear becomes more and more fashion driven. Why should retailers stick to this advance buying pattern where they take big risks predicting the market six months ahead of time? What if this advance buying pattern is eliminated for whatever reasons? How companies can improve their new products launch process to make them prepared for the possible new challenges in the future? This research studies the new product development process in a large athletic footwear company (Hereinafter US-Footwear). Recommendations include adopting a systematic new products development framework to shorten the time to market. Specifically, this systematic roadmap will force companies to redefine milestones and key activities; this approach will also form a "funnel" screening and informed decision making mechanism. Consequently, companies would be able to eliminate non-value added activities and focus their valuable resources only on the most winning products. It will thus provide companies huge potential to shorten the time to market by doing fewer activities, fewer products and by greatly / (cont.) reducing iterative design changes. Lastly, the author believes that fashion business in general could benefit by adopting the similar approach. / by Xiaoqin Dong. / M.Eng.in Logistics

Thesis: Ph. D., Massachusetts Institute of Technology, Engineering Systems Division, 2015. / 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 235-240). / Scarce pipeline capacity in regions that rely on natural gas technologies for electricity generation has created volatile prices and reliability concerns. Gas-fired generation firms uniquely operate as large consumers in the gas market and large producers in the electricity market. To explore the effects of this coupling, this dissertation investigates decisions for firms that own gas-fired power plants by proposing a mixed-integer linear programming model that explicitly represents multi-year pipeline capacity commit- ments and service agreements, annual forward capacity offers, annual maintenance schedules, and daily fuel purchases and electricity generation. This dissertation's primary contributions consist of a detailed representation of a gas-fired power-plant owner's planning problem; a hierarchical application of a state-based dimensionality reduction technique to solve the hourly unit commitment problem over different tem- poral scales; a technique to evaluate a firm's forward capacity market offer, including a probabilistic approach to evaluate the risk of forced outages; a case study of New England's gas-electricity system; and an exploration of the applicability of forward capacity markets to reliability problems for other basic goods. / by Tommy Leung. / Ph. D.

Thesis (M. Eng. in Logistics)--Massachusetts Institute of Technology, Engineering Systems Division, 2004. / "June 2004." / Includes bibliographical references (leaves 112-115). / This paper assesses the value of Radio Frequency Identification (RFID) in the transportation forecasting, planning, and execution processes for truckload (TL) and less than truckload (LTL) services. The results show that the value of RFID in transportation highly varies according to the sub-process and the player which are considered. The value in forecasting and planning is high for consignees and TL carriers, but low for shippers. For the execution process, the value is high for consignees, moderate for shippers, low for TL carriers and potentially high for LTL carriers. Fundamentally, the common value that RFID can bring to each player is labor cost reduction, obtained by automation of dock and clerical activities, and better service in tracking and reconciling, thanks to accurate and up-to-date information. The potential for dramatic improvement in transportation through RFID exists. Achieving such improvement will be possible only if information sharing rules, such as those espoused by Collaborative Transportation Management (CTM), and well formalized procedures are implemented. RFID can give to CTM invaluable "raw material", while CTM can enable the transportation players to fully take advantage of this material. / by Antoine Guitton. / M.Eng.in Logistics

Thesis (M. Eng. in Logistics)--Massachusetts Institute of Technology, Engineering Systems Division, 2004. / Includes bibliographical references (p. 47-48). / According to financial theory, open markets efficiently and effectively aggregate all available information about future events into their prices. Recent empirical evidence has shown that speculative markets, from gambling to web-games, are better at predicting the future than more commonly used statistical or survey-based forecasting methods. As a result, a number of companies have conducted experiments to evaluate the use of prediction markets as an alternative forecasting methodology. This paper offers a comprehensive framework for determining when and how prediction markets should be employed in a business context. / by Jared M. Schrieber. / M.Eng.in Logistics

Thesis (M. Eng. in Logistics)--Massachusetts Institute of Technology, Engineering Systems Division, 2005. / Includes bibliographical references (leaf 57). / Targus, a consumer products distributor of laptop cases and accessories, lacks formal processes for deciding on initial purchases of short-term life cycle products with uncertain demand. This thesis reviews the current processes for the initial purchases and determines where the Company can reduce inventory risks related to the initial purchase. The research and analysis has three aspects: interviews of eight Targus managers was used to understand current initial purchases practice, a sample of initial purchase data was selected to analyze forecast errors and life cycle management, and the newsvendor problem was applied to the data sample to determine the optimal purchase which was then compared to Targus' purchases. It was found that Targus has a tendency to under-purchase items for the initial purchase and does not establish potential profitability of products prior to introduction. This thesis recommends the Company to incorporate a newsvendor approach as a basis to benchmark its initial purchases for the initial product introduction process. / by Eugene T. Takenaga. / M.Eng.in Logistics

Thesis (M. Eng. in Logistics)--Massachusetts Institute of Technology, Engineering Systems Division, 2012. / Cataloged from PDF version of thesis. / The purpose of this research was to evaluate whether a shipper's fuel surcharge (FSC) program affected its per-load transportation costs in the United States full-truckload (TL) transportation industry. In this study, we restricted transportation costs to line-haul charges and fuel surcharge premiums. Using ordinary least squares (OLS) regression, we examined the effect of a shipper's FSC program on its line-haul charges. We controlled for well-established transportation cost drivers, including distance and geography. We found that carriers discounted their line-haul rates according to a shipper's FSC program. The more a shipper paid in FSC premiums, the less it paid in line-haul charges. For fuel prices above $2.08 per gallon, however, the fuel surcharge premiums dwarfed the line-haul discount. This effect was most pronounced for shippers with low efficiency values. Shippers with lower efficiency values paid higher per-load transportation costs than shippers with higher efficiency values. / by Molly Abramson and Ajit Sawant. / M.Eng.in Logistics

Thesis (S.M. in Engineering and Management)--Massachusetts Institute of Technology, Engineering Systems Division, 2013. / Cataloged from PDF version of thesis. / Includes bibliographical references (pages 63-64). / Since the advent of Web 2.0, crowd funding has played an increasingly important role as a means of financing for startup companies. Crowd funding is a particular means of financing where money is obtained from the public in exchange for equity or rewards. Currently, only accredited investors, including investment firms, pension funds and individuals with personal net worth of at least $1 million or earning at least $200,000 a year (US Securities and Exchange Commission, 2012) can invest in private companies and get equity in return. With the passing of the JOBS Act by President Barack Obama in April 2012, making investments in exchange for equity in private companies will soon be available to small investors. This thesis examines the phenomenon of crowd funding through estimating the cost of capital for the crowd funded projects and the factors influencing their success. Data is obtained from a popular fund raising website, Kickstarter and analysis is carried out using regression. The results show that the probability of a successful fund raising campaign is rather low, at 43%. Setting a low funding target, entering a market that has fewer competing products and building up popular support through captivating design or meeting latent user needs is associated with a higher probability of fund raising success. In addition, though the median cost of capital is negative, the cost of capital exhibits a wide range and it may be more expensive to fund projects through crowd funding compared to debt financing, which has a much smaller spread of its cost of capital. As such, the results show that a lower the cost of capital is associated with a lower cost of goods sold, faster delivery of rewards and a high proportion of free capital, which may be obtained by encouraging donations or asking for a higher price premium for its products relative to retail price. / by Lingkai Terence Teo. / S.M.in Engineering and Management

Thesis: M. Eng. in Logistics, Massachusetts Institute of Technology, Engineering Systems Division, 2014. / Cataloged from PDF version of thesis. / Includes bibliographical references (pages 81-82). / This thesis aims to quantitatively estimate the potential impact of 3D Printing on global supply chains. Industrial adoption of 3D Printing has been increasing gradually from prototyping to manufacturing of low volume customized parts. The need for customized implants like tooth crowns, hearing aids, and orthopedic-replacement parts has made the Life Sciences industry an early adopter of 3D Printing. Demand for low volume spare parts of vintage cars and older models makes 3D Printing very useful in the Automotive industry. Using data collected from expert interviews, site visits, and online sources, and making assumptions where necessary, we developed our model by comparing the current supply chain processes and cost with the future supply chain processes and cost after 3D Printing was adopted. We also developed models to show future trends in 3D Printing adoption and costs. There were several challenges and limitations in this process due to limited availability of primary data, which led us to use secondary sources like the internet and make assumptions. One of the key features of our thesis is that we explicitly state all our assumptions, and present a model that is amenable to what-if analysis. Our analyses suggest that 3D Printing will change future supply chains significantly as production will move from make-to-stock in offshore/low-cost locations to make-on-demand closer to the final customer. This will significantly reduce transportation and inventory costs. The model shows that this will be especially true for low volume products. The models also show us the sensitivity analysis around the change in supply chain costs with the projected decrease in the cost and an increase in adoption of 3D Printing. The other major impact will be the reduction in lost sales due to unavailability of products and increase in customer satisfaction with almost 100% product availability. Finally, our analyses also indicate that 3D Printing could change the dynamics of the logistics industry: there may be reduction in the volume of freight business with an opportunity for 3PL companies to provide 3D Printing services in warehouses. / by Varun Bhasin and Muhammad Raheel Bodla. / M. Eng. in Logistics

Thesis (Ph. D.)--Massachusetts Institute of Technology, Engineering Systems Division, 2010. / Cataloged from PDF version of thesis. / Includes bibliographical references. / The Clean Development Mechanism (CDM) has evolved at a surprising speed since 2003 and is considered to have made positive contributions to the development of greenhouse-gas-reducing projects in developing countries. Taking into account its historical significance as the first effort of its kind and its current success, a thorough evaluation of its system and its effectiveness is of critical importance. Against this backdrop, this study closely investigates each stage of the CDM project cycle from development and registration of projects to issuance of certified emission reductions and identifies influential factors for the successful CDM implementation. For the analysis, we performed an extensive quantitative analysis augmented by a descriptive study, based on information of approximately 5000 CDM project. Our findings suggest that the development of CDM projects is stimulated by favorable economic, social and technical environments in host countries as well as supportive CDM administration. This explains why projects are currently concentrated in certain countries such as China and India. Once projects are developed and submitted for validation, the success of the CDM projects at the next stages of project cycle related to registration and Certified Emission Reduction (CER) issuance is influenced by their types and a choice of Designated Operational Entities and project consultants. In particular, significant difference in registration success exists across project types, which calls for special attention of both the CDM authority and project participants to projects with high risks like energy efficiency, fossil fuel switch and biomass projects. Lastly, we found that performance of projects is affected by very project-specific conditions. For many of the most poorly performing projects, failure is attributable to technical and operational problems at the initial stage of project implementation, which highlights the importance of well-prepared PDDs. Based on the findings, the thesis concludes with policy recommendations to enhance the capacities and improve the performance of the major players under the CDM. / by Jaemin Song. / Ph.D.