Thesis (Ph. D.)--Massachusetts Institute of Technology, Engineering Systems Division, 2005. / Includes bibliographical references (v. 2, p. 311-316). / The electric power system in the US developed with the assumption of exogenous, inelastic demand. The resulting evolution of the power system reinforced this assumption as nearly all controls, monitors, and feedbacks were implemented on the supply side. Time invariant, averaged retail pricing was a natural extension of the assumption of exogenous demand and also reinforced this condition. As a result, the market designs and physical control of the system exclude active participation by consumers. Advances in information and communications technologies enable cost effective integration of demand response. Integrating demand into the US electricity system will allow the development of a more complete market and has the potential for large efficiency gains. Without feedbacks between supply and demand, attempts to develop competitive markets for electricity will suffer from a greater potential for market power and system failure. This thesis provides an analysis of the technical, regulatory, and market issues to determine a system structure that provides incentives for demand response. An integrated, dynamic simulation model is utilized to demonstrate the effects of large scale adoption of demand response technologies. The model includes distributed decision making by both consumers and investors in generation capacity, the effects of their decisions on market prices, and the feedbacks between them. Large scale adoption of demand response technology is simulated to quantify the potential benefits of responsive demand. The effects of technology improvement via learning, long term demand elasticity, and policies to promote adoption are considered. / (cont.) The simulations show that diminishing returns for adopters and free rider effects limit the attractiveness of individual adoption. A subsidy to alleviate the costs to adopters can be justified by the significant system level savings from widespread participation. Several pernicious effects can emerge from large scale demand response, however, including increased price volatility due to reductions in generation capacity reserve margin, increases in long term demand, and increased emissions from the substitution of peak generation capacity, such as natural gas and renewables by intermediate capacity. Significant rent transfers will also occur, and stakeholder analysis is conducted to determine interests and distributional effects of large scale demand response. / by Jason W. Black. / Ph.D.

Thesis (M. Eng. in Logistics)--Massachusetts Institute of Technology, Engineering Systems Division, 2011. / Cataloged from PDF version of thesis. / Includes bibliographical references (p. 73). / A large U.S. based retailer underwent a large, complex multi-year supply chain network transformation. This transformation resulted in significant savings in logistics costs. Additionally, the regional distribution center that was introduced as part of this transformation as a new node between the supplier and the store became the decision making center for placing purchase orders with suppliers and for receiving and shipping the purchase order to individual stores. This resulted in longer lead times causing a change in the in-store units held and therefore, directly impacting the net sales. This thesis focuses on establishing the relationship between the stores performance and lead-time, review-time combinations in both supply chain networks, the original direct to store and the new regional distribution based networks. / by Sunil Anand and Xiaobei Song. / M.Eng.in Logistics

Thesis (M. Eng. in Logistics)--Massachusetts Institute of Technology, Engineering Systems Division, 2012. / Cataloged from PDF version of thesis. / Includes bibliographical references (p. 85-89). / The Internet has changed the way people purchase goods in the 21st century: many types of goods and services have become available online. Luxury goods followed this trend after an initial delay, primarily due to the nature of these goods. At the time of the preparation of this document, there were no proven guidelines for building the most successful e-retail website for luxury goods from the brand management and profitability perspectives. Ralph Lauren (RL) is an established American brand, well known for quality and consistent style in the following categories: clothing for women, men, and children; home goods, accessories; and fragrances. RL Corporation houses many labels that constitute premium and luxury offerings. RL currently sells through the Internet in many countries, in addition to countless company owned stores, 9 flagship stores, department stores and boutiques distributed around the world. To continue growth, RL wants to launch an e-retail website for Canada. This thesis aims to provide supply chain network considerations for the successful operation of the Canadian e-retail website for RL. These considerations stem from a careful look into potential luxury website characteristics that would meet the company objective of elevating the brand towards the luxury category. It is recommended that RL secure expansion capacity that will likely be necessary for B2C operation at its Toronto distribution center (DC). In addition, material handling equipment that will process a high volume of small orders should be placed in this DC. The Vancouver cross-docking facility could be expanded in the future as prompted by sales volume and coupled with a DC to cater to the West Coast of Canada. Also, it is recommended that advanced customer tracking systems and databases be employed, especially to determine high value customers for tailored offerings in the luxury segment. / by Dilek Tansoy and Yi Linn Teo. / M.Eng.in Logistics

Thesis (S.M. in Technology and Policy)--Massachusetts Institute of Technology, Engineering Systems Division, 2013. / Cataloged from PDF version of thesis. / Includes bibliographical references (p. 76-82). / The author proposes that personalized learning can be brought to traditional and nontraditional learners through a new type of asynchronous learning platform called Guided Learning Pathways (GLP). The GLP platform allows learners to intelligently traverse a vast field of learning resources, emphasizing content only of direct relevance to the learner and presenting it in a way that matches the learner's pedagogical preference and contextual interests. GLP allows learners to advance towards individual learning goals at their own pace, with learning materials catered to each learner's interests and motivations. Learning communities would support learners moving through similar topics. This thesis describes the software system design and architecture required to support Guided Learning Pathways. The author provides detailed information on eight software applications within GLP, including specific learning benefits and features of each. These applications include content maps, learning nuggets, and nugget recommendation algorithms. A learner scenario helps readers visualize the functionality of the platform. To describe the platform's software architecture, the author provides conceptual data models, process flow models, and service group definitions. This thesis also provides a discussion on the potential social impact of GLP in two areas: higher education institutions and the broader economy. / by Cole J. Shaw. / S.M.in Technology and Policy

Thesis (M. Eng. in Logistics)--Massachusetts Institute of Technology, Engineering Systems Division, 2004. / Includes bibliographical references (p. 87). / Radio frequency identification (RFID) has an enormous potential impact within the semiconductor supply chain, especially within semiconductor manufacturing. The end benefit of RFID will be in the mass serialization, and the subsequent tracking and tracing, of individual semiconductors, or what is referred to as Unit Level Traceability (ULT). Before all of the technical hurdles of ULT are overcome, however, there exists a host of other applications for RFID within semiconductor manufacturing. The identification of what can and what should be RFID-tagged and read, the analysis of how to collect this information and what to do with the data, and the implementation of some targeted opportunities will provides valuable information with regards to the technical and logistical hurdles of RFID within semiconductor manufacturing far before ULT becomes a reality. / by David Ian Cassett [and] Christopher William Chiu Hopeman. / M.Eng.in Logistics

Thesis (M. Eng. in Logistics)--Massachusetts Institute of Technology, Engineering Systems Division, 2009. / Includes bibliographical references (leaves 161-164). / Our research demonstrates that supply chain management techniques can improve the incremental gross profits of wind plant and storage operations by up to five times. Using Monte-Carlo simulation we create and test scenarios that achieve incremental operating profits of up to 15 percent of base case revenue, and show pre-tax profit. We show that energy storage-specifically in the form of utility-scale batteries - can become economically-viable today when using supply chain management strategies under certain scenarios. To achieve these results we have built a simulation model with three data inputs. First, we synthesized the output of a 120 MW wind plant in Maine for both summer and winter seasons. Second, we simulated New England ISO market pricing data for both the Day-Ahead and Real-Time markets in summer and winter seasons using Monte Carlo simulations. Third, using actual data from two existing battery companies, we incorporated the technical and cost specifications for two energy storage facilities. All of these data inputs feature adjustable parameters so we can test various plant configurations, market volatilities and storage capabilities, among other inputs. Using our model, we then employed supply chain management network design strategies and daily operating policies to test profitability improvements on our wind plant-plus-storage operation. For example, we ran simulations for scenarios where our storage facility is either located in Maine next to our wind plant, or located in another state. / (cont.) Also, since storage can make wind generation a predictable capacity resource, we ran simulations to test results in both the Day-Ahead and Real-Time markets. In addition we developed four (4) inventory management policies with dynamic input (charge) and output (discharge) strategies for our storage units. For each policy, we had to conceptualize the policy - while considering planning horizon, lead time, holding costs, shortage costs, market pricing and storage capabilities - and then build functionality in our model to execute those strategies in dynamic pricing and wind plant output environments. The outcomes of our simulation model include incremental gross profit, operating profit and pre-tax profit for each of 54 scenarios, as well as 11 management insights for wind plant and storage operators, storage technology manufacturers and New England ISO leadership. / by Prashant Saran and Clayton W. Siegert. / M.Eng.in Logistics

Thesis (Ph. D.)--Massachusetts Institute of Technology, Engineering Systems Division, 2012. / Cataloged from PDF version of thesis. / Includes bibliographical references (p. 209-222). / This dissertation presents empirical findings related to two aspects of long-range planning: scenario planning as a planning method and cognition of planners. Long-range planning situations are encountered when designing public infrastructures (e.g. transportation systems) as well as developing strategies for corporate enterprises (e.g. firms' supply chains), due to the long implementation times and/or long lives of the invested assets. Such investments tend to have high stakes, face extreme uncertainty about the future environment they encounter, and have an open-systems nature as the implementation and operation of assets affects and is affected by the actions of many and diverse stakeholders. Three research questions pertaining to these aspects are answered in this work as three stand-alone studies. The first study (Chapter 2) examines the effects of scenario planning on long-range investment decisions made by field experts. The results of three field experiments show that experts systematically change their investment decisions and/or their confidence in them after evaluating the investments in a scenario. Field experts are also more likely to invest in flexible strategies after being exposed to multiple scenarios. The second study (Chapter 3) presents an extensive and an abridged version of the scenario creation process. Instead of seeing scenario-creation as an art, this research provides two versions of a more engineered scenario-creation process, and demonstrates their application in two separate field studies. Both versions of the process are presented with detailed instructions and rationale for performing each step. This study also provides clear definitions of the terms used in the process description and grounds them in the organizations literature. The third study (Chapter 4) explores the relationship between a manager's perceptions and beliefs about future environment, and the strategies s/he recommends. Using a prospective research design, this study first tests three hypotheses about general characteristics of managerial cognition. A closer look at different cognitive types in this data reveals unsuspected patterns in strategic thinking of managers of different types. A typology of managerial cognition is built using this inherent variation. Inductive analysis shows that managers of different cognitive types envision strikingly different types of strategies. / by Shardul Sharad Phadnis. / Ph.D.

Thesis (S.M. in Technology and Policy)--Massachusetts Institute of Technology, Engineering Systems Division, 2013. / 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. Page 75 blank. / Includes bibliographical references (p. 67-74). / This thesis develops a framework for a flexible design approach to support decision-making in water supply infrastructure planning. It contrasts with a conventional, deterministic planning approach that uses past data or forecasts to anticipate future needs. This thesis surveys current approaches that attempt to consider uncertainty, including scenario planning, decision analysis, sensitivity analysis, real options, dynamic strategic planning, and adaptive management. A flexible design approach builds on current approaches and explores flexibility through infrastructure size and function. The approach intends to be applicable across various water infrastructure systems. This thesis describes real world and theoretical applications of flexible design, including climate change adaptation planning for water utilities, flexible planning for water infrastructure investments, and flexibility in urban drainage systems. The proposed flexible design approach employs probabilistic and simulation methods to anticipate a range of future circumstances and identify top-performing strategies. The engine of the framework is a time-series stochastic analysis that uses simulation in a discounted cash flow Excel model. First, it identifies key inputs and performance metrics, characterizes uncertainty distributions, and defines strategies of varying flexibility. Next, it employs Monte Carlo simulation and compares strategy performance through target curves and multiple criteria analyses. Singapore's water resources system inspires the characteristics of the model. The best-performing flexible approach introduces a cost savings of 15% over a 50-year timespan. To successfully implement a flexible design approach, leaders in the profession must guide the shift to planning methods that explicitly recognize the role of uncertainty in the planning process. While some implementation barriers present difficulties, the proposed flexible design approach enables substantial cost savings and fosters a deeper understanding of a water resources system in the face of future uncertainty. / by Melanie Kathleen Wong. / S.M.in Technology and Policy

Thesis (Ph. D.)--Massachusetts Institute of Technology, Engineering Systems Division, 2011. / Cataloged from PDF version of thesis. / Includes bibliographical references (p. 248-253). / Platforming has become an important means of cost-sharing among industrial products. However, recent research has shown that many firms face systemic downward pressure on commonality, with the result that many platforms realize less commonality than intended. This research was chartered to evaluate the costing of commonality benefits, the associated returns from commonality investments, and the potential impact of divergence on commonality benefits. This dissertation used a tiered approach to the research questions. A statistical study of commonality returns was conducted, finding evidence for a potential link between divergence and cost growth. Broad practice surveys of 16 firms revealed cost allocation practices and internal funding strategies as potential determinants of the commonality-cost relationships. Three detailed case studies were conducted to trace benefit trajectories through platformed products in the presence of commonality changes. We find support for the hypothesis that divergence has cost consequences, notably reducing inventory benefits, creating higher quality expenses and requiring additional manufacturing coordination. Additionally, we show that lead variants bearing platform costs achieved weaker investment returns and re-captured few benefits from later variants. We find also find evidence to refute the notion that representation at design reviews ensures downstream benefits are represented. Several management practices for making commonality decisions are identified. We propose a framework for commonality cost decisions, which explicitly captures the impact of individual variant decisions on the platform's cost structure. We identify a commonality cycle, a progression of commonality strategies seen by firms, driven by growing benefit analogies among platforms enabling larger investments, premature investment evaluation, and unrealized returns on commonality investments. / by Bruce G. Cameron. / Ph.D.

Thesis (S.M.)--Massachusetts Institute of Technology, Engineering Systems Division, 2005. / Includes bibliographical references (p. 147-151). / The research contained in this thesis explores design attributes of the enterprise performance measurement system required for the transformation to the lean enterprise and its management. Arguments are made from the literature that successful deployment of the lean practices, across three different stages of the evolution of lean thinking, requires a supporting performance measurement system. The increase in scope of lean practices at each stage of the evolution increases the complexity in achieving synchronization across the enterprise subsystems. The research presents various attributes of the performance measurement system required at each stage and further derives the three key attributes for the design of the lean enterprise performance measurement system. These three attributes are: enterprise level stakeholder value measures, the causal relationships across performance measures at each level, and Uniform and consistent set of performance measures. A detailed case study of an aerospace and defense business of a multi-industry corporation which has embarked on a journey towards creating a lean enterprise is presented. / (cont.) It highlights several challenges in the transformation from the perspective of performance measurement. The key challenges identified are: First, disconnect between the performance measurement for the lean practices and regular business practices hinder the adoption of lean practices. This disconnect exists due to the existence of both legacy performance measures and the new measures. Second, lack of understanding of the cause-effect relationship between performance measures across different enterprise levels poses difficulty evaluating the impact of lean related efforts. Third, use of non-uniform performance measures across various enterprise subsystems leads to non-lean behavior. The theory underlying performance measurement is reviewed including the widely-accepted performance measurement frameworks suggested for the design of enterprise performance measurement system. Analysis of these frameworks reveals that none of the existing frameworks completely capture the desired attributes for the lean enterprise performance measurement system. / (cont.) To design the lean enterprise performance measurement system, this research suggests a conceptual design that explicates the use of various tools and techniques to address the critical attributes. To identify stakeholder value measures this design demonstrates the use of stakeholder value analysis. Use of system dynamics modeling and structural equation modeling is suggested to establish, validate and evolve the cause-effect relationships between performance measures. And, to maintain the uniform set of measures the creation of measures dictionary is explained. Further, research is needed to empirically validate the model as a means for successful transformation and management of the lean enterprise. / by Vikram Mahidhar. / S.M.