Browsing by Subject "Parametric modeling"
Now showing 1 - 5 of 5
- Results Per Page
- Sort Options
Item A framework for optimized and automated tower crane planning in preconstruction phase(2018-05) Ji, Yuanshen; Leite, Fernanda L.; Alves, Thais; Borcherding, John; Caldas, Carlos; Machemehl, RandyThe construction industry has only been using tower cranes to assist lift tasks for less than eight decades. In this time, various types of tower cranes have been developed to suit specific requirements of different construction job sites. However, tower cranes are less broadly used in North America, especially in the United States, than Europe and Asia. This is partially attributed to ineffective tower crane planning in the preconstruction phase, which lacks a formalized planning process and a commonly agreed upon planning target. In the state-of-practice, Engineers conduct iterative planning in a manual approach using mostly 2D representations of project data and equipment specifications. The chance of identifying an optimal solution is small and site-specific constraints, such as spatial-temporal conflicts and the ever-changing clearance when collaborating with other machinery equipment are subject to being overlooked. A formalized planning process, along with advanced modeling, simulation, and optimization techniques, provides intuitive observation of the multiple dimensional solution space of the tower crane planning problem. It has the potential to automate and advance the planning of machinery equipment used on building construction projects. This dissertation research enhances tower crane planning in the preconstruction phase using parametric modeling, visualization, four-dimensional simulation, rule-based checking, and optimization algorithms. The overall goal of this study is to create a framework that allows engineers and researchers to formalize site-specific constraints of tower crane plans, and to automate the analysis, evaluation, and visualization of multiple alternative plans. In order to identify the optimal solution, an optimization formulation using mixed-integer programming was introduced. An application of such framework is presented in real-world scenarios and the results demonstrate that the effectiveness and efficiency of tower crane planning in the preconstruction phase can be improved and a broad range of alternative plans can be quantitatively assessed for communication, training, or continuous improvement. The main contribution of this study includes the introduction of a machinery equipment plan assessment framework. This framework incorporates project data (e.g., schedule, site-layout, lift demands) and a specialized tower crane planning model to visualize, identify, and analyze the obstructions in alternative plans with respect to spatial, capacity, and safety constraints.Item Automated estimation of time and cost for determining optimal machining plans(2012-05) Van Blarigan, Benjamin; Campbell, Matthew I.; Li, WeiThe process of taking a solid model and producing a machined part requires the time and skillset of a range of professionals, and several hours of part review, process planning, and production. Much of this time is spent creating a methodical step-by-step process plan for creating the part from stock. The work presented here is part of a software package that performs automated process planning for a solid model. This software is capable of not only greatly decreasing the planning time for part production, but also give valuable feedback about the part to the designer, as a time and cost associated with manufacturing the part. In order to generate these parameters, we must simulate all aspects of creating the part. Presented here are models that replicate these aspects. For milling, an automatic tool selection method is presented. Given this tooling, another model uses specific information about the part to generate a tool path length. A machining simulation model calculates relevant parameters, and estimates a time for machining given the tool and tool path determined previously. This time value, along with the machining parameters, is used to estimate the wear to the tooling used in the process. Using the machining time and the tool wear a cost for the process can be determined. Other models capture the time of non-machining production times, and all times are combined with billing rates of machines and operators to present an overall cost for machining a feature on a part. If several such features are required to create the part, these models are applied to each feature, until a complete process plan has been created. Further post processing of the process plan is required. Using a list of available machines, this work considers creating the part on all machines, or any combination of these machines. Candidates for creating the part on specific machines are generated and filtered based on time and cost to keep only the best candidates. These candidates can be returned to the user, who can evaluate, and choose, one candidate. Results are presented for several example parts.Item Parametric energy modeling tool for climate dependent guidelines(2013-05) Morales, Cristian Enrique; Garrison, MichaelThe purpose of this thesis is to develop a simple tool that can help designers and researchers obtain general guidelines for buildings in terms of energy usage and LCC. Another objective of this thesis is to apply this tool to residential buildings in order to understand which variables are relevant in terms of energy consumption and LCC costs. A one-story rectangular house was parameterized in terms of five variables: total glazing area; south window-to-wall ratio (WWR); east and west WWR (which are symmetrical for these two facades); insulation width; and window type (ranging from a single clear window to a double low e-clear argon filled window). A high average glazing area (30-40% of floor area) was applied in order to increase energy loads and to augment the importance of the window properties. Simulation was performed through Energy-plus (in conjunction with a code developed especially for this project) for three cities: Austin, Boston, and Seattle. A total of 1055 simulations were run for each city. The experiment showed that only the total glazing area, the E-W WWR and the window types were relevant variables. The former variable is highly correlated with total energy consumption across all cities. Another important conclusion was that each variable's effect on energy consumption worked independently of each other, as there were no considerable differences when analyzing variables individually, as opposed to analyzing them holistically. Results showed that, for Austin and Boston, it was possible to reduce energy loads by 35% and 27% respectively with a double low-e green window (as compared to a single clear window). Similarly, Seattle showed a reduction of 29% for a double low e-clear argon filled window. Nevertheless, the simplest type of window (type 1) presented the best results in terms of LCC. Therefore, we can conclude that only under a high-energy demand situation, such as with office buildings, would it be possible to obtain positive LCC results for double glazed windows. Consequently, double glazed windows will not present positive economical results in typical residential buildings. A second simulation was performed under a tighter HVAC schedule and higher internal loads. In this new scenario, the best windows were the same as with the first simulation, but maximum energy savings were higher: 50%, 34% and 35% for Austin, Boston, and Seattle, respectively. Nevertheless, when considering LCC, a double-clear window presented the best results for Austin, Boston, and Seattle, with 17%, 11%, and 5% reductions in costs respectively compared to the type 1 window. Therefore, if designers are only concerned with costs, the problem of what window to choose becomes non-trivial only for high-energy demand cases.Item Parametric modeling and design synthesis for electromechanical actuators(2004-05-22) Gloria, Christopher Elmose; Tesar, DelbertMechanical systems are becoming increasingly complex, containing hundreds of design parameters where the imperative is to increase performance while reducing costs. To meet these demands will require a paradigm shift away from trial and error procedures for monolithic structures and one-off designs towards an open architecture where global technology diffusion and systems on demand can become the accepted basis for significant economic return on investment. To achieve this goal, the design process must be more completely understood for the designer to handle increasingly complex systems in a manner that improves design. It is the goal of this work to convey the need for improved design science and to lay a foundation for the design of electromechanical actuators using fundamental principles and parametric formulations. Overall, this research provides the electromechanical actuator designer with a framework and some initial tools to aid in the design synthesis process with the ultimate goal of improving actuator design. This entails the development of a step by step process beginning with the selection of the actuator architecture based on specified design priorities. This is followed by the development of a parametric model of the chosen actuator configuration based on the design priorities and including the most important design parameters. Given this model, a method for extracting useful information in the form of design tradeoffs is developed. This work uses a method called algebraic reduction to manipulate the parametric model and gain design insights. An example demonstrating this design synthesis framework is given and includes the application of algebraic reduction to polynomial systems.Item Statistically driven decision making in football through the use of reinforcement learning, random utility models, and parametric modeling(2022-12-01) Biro, Preston; Walker, Stephen G., 1945-; Calder, Catherine; Murray, Jared S; Fink, JoshDecision making under conditions of uncertainty is inherently a difficult task. The use of data can alleviate this difficulty by informing the decision maker of past results, but often raw data can still mislead if not properly put into context. Additionally, long-term optimal behavior does not always align with short-term needs, and thus even a well-tailored algorithm can provide undesirable results. The game of football provides a unique avenue for application due to the structure of the game and increasing levels of data availability. Through the use of reinforcement learning and random utility models, statistically optimal decisions can be identified under a variety of utility mindsets. Parametric models properly representing the data generating process can also provide insight on the underlying information. Using football play-by-play data from the NFL and college level (specifically for the Presbyterian College Fall 2021 team), a system of algorithms is designed to assist with the task of play calling. These algorithms rely on the uncovering of the underlying utility of states in the game, which themselves can provide additional information on how to increase efficiency.