Institute for Systems Research Technical Reports
Permanent URI for this collectionhttp://hdl.handle.net/1903/4376
This archive contains a collection of reports generated by the faculty and students of the Institute for Systems Research (ISR), a permanent, interdisciplinary research unit in the A. James Clark School of Engineering at the University of Maryland. ISR-based projects are conducted through partnerships with industry and government, bringing together faculty and students from multiple academic departments and colleges across the university.
Browse
Search Results
Item Sensitivity Analysis and Discrete Stochastic Optimization for Semiconductor Manufacturing Systems(2000) Mellacheruvu, Praveen V.; Herrmann, Jeffrey W.; Fu, Michael C.; ISRThe semiconductor industry is a capital-intensive industry with rapid time-to-market, short product development cycles, complex product flows and other characteristics. These factors make it necessary to utilize equipment efficiently and reduce cycle times. Further, the complexity and highly stochastic nature of these manufacturing systems make it difficult to study their characteristics through analytical models. Hence we resort to simulation-based methodologies to model these systems.This research aims at developing and implementing simulation-based operations research techniques to facilitate System Control (through sensitivity analysis) and System Design (through optimization) for semiconductor manufacturing systems.
Sensitivity analysis for small changes in input parameters is performed using gradient estimation techniques. Gradient estimation methods are evaluated by studying the state of the art and comparing the finite difference method and simultaneous perturbation method by applying them to a stochastic manufacturing system. The results are compared with the gradients obtained through analytical queueing models. The finite difference method is implemented in a heterogeneous simulation environment (HSE)-based decision support tool for process engineers. This tool performs heterogeneous simulations and sensitivity analyses.
The gradient-based techniques used for sensitivity analysis form the building blocks for a gradient-based discrete stochastic optimization procedure. This procedure is applied to the problem of allocating a limited budget to machine purchases to achieve throughput requirements and minimize cycle time. The performance of the algorithm is evaluated by applying the algorithm on a wide range of problem instances.
Item Comparing Gradient Estimation Methods Applied to Stochastic Manufacturing Systems(2000) Mellacheruvu, Praveen V.; Fu, Michael C.; Herrmann, Jeffrey W.; ISRThis paper compares two gradient estimation methods that can be usedfor estimating the sensitivities of output metrics with respectto the input parameters of a stochastic manufacturing system.A brief description of the methods used currently is followedby a description of the two methods: the finite difference methodand the simultaneous perturbation method. While the finitedifference method has been in use for a long time, simultaneousperturbation is a relatively new method which has beenapplied with stochastic approximation for optimizationwith good results. The methods described are used to analyzea stochastic manufacturing system and estimate gradients.The results are compared to the gradients calculated fromanalytical queueing system models.These gradient methods are of significant use in complex manufacturingsystems like semiconductor manufacturing systems where we havea large number of input parameters which affect the average total cycle time.These gradient estimation methods can estimate the impact thatthese input parameters have and identify theparameters that have the maximum impact on system performance.
Item Integrated Manufacturing Facility Design(1995) Ioannou, George; Minis, I.; ISRThis dissertation addresses for the first time the integrated problem of designing the manufacturing shop layout concurrently with its material handling system. Specifically, this study provides a method to derive shop designs that are economic to set up and efficient to operate. In doing so, it considers the following highly interrelated issues: i) The topology of the material flow network, ii) the transporter fleet size and routing, and iii) the layout of the resource groups on the shop floor. The design problem is modeled by a comprehensive mathematical program which captures critical practical concerns such as investment and operational costs, traffic congestion, and transporter capacities. The model is decomposed into three NP- hard subproblems by fixing and/or aggregating variables and constraints. The first subproblem is the generic multi-commodity fixed charge capacitated network design, for which an improved lower bound is derived based on a dual ascent method. This problem is solved by three heuristics that provide near-optimal network designs. The second subproblem concerns the transporter routing, which is a special case of the distance-constrained vehicle routing problem. For the transporter routing problem near-optimal solutions are derived in polynomial time by two efficient heuristics with bounded worst-case performance. Tight lower bounds are provided by solutions to the assignment problem. An integrated method combines the most effective heuristics for the material handling system design subproblems with a simulated annealing scheme to solve the global shop design problem. Our novel approach addresses simultaneously most major decisions involved in manufacturing shop design, and provides globally near optimal solutions. The method is applied to the redesign of the shop of a large manufacturer, and generates a particularly attractive production system design reducing significantly both investment and operational costs, while providing for smooth system operation.Item An Integrated Model for Manufacturing Shop Design(1995) Ioannou, George; Minis, Ioannis; ISRThis paper presents an integer programming formulation for the manufacturing shop design problem, which integrates decisions concerning the layout of the resource groups on the shop floor with the design of the material handling system. The model reflects critical practical design concerns including the capacity of the flow network and of the transporters, and the tradeoff between fixed (construction and acquisition) and variable (operational) costs. For realistic industrial cases, the size of the problem prevents the solution through explicit or implicit enumeration schemes. The paper addresses this limitation by decomposing the global model into its natural components. The resulting submodels are shown to be standard problems of operations research. The decomposition approach provides ways to solve the integrated shop design problem in an effective manner.Item Current Research on Manufacturing Shop and Material Handling System Design(1995) Ioannou, George; Minis, Ioannis; ISRThe importance of the manufacturing shop design in the successful operation of a production system is well known and as a result, significant research has been devoted to this area. This paper reviews important literature in various aspects of manufacturing shop design including layout, material flow path design, and transporter fleet sizing and routing. In addition, the paper focuses on contributions to integration issues such as the design for operation of material handling systems, and the concurrent design of the shop layout and the transportation system. Research studies in these areas are critically examined, and emerging opportunities for research are identified.Item Diamond-Tree: An Index Structure for High-Dimensionality Approximate Searching(1992) Faloutsos, Christos; Jagadish, H.V.; ISRA selection query applied to a database often has the selection predicate imperfectly specified. We present a technique, called the Diamond-tree, for indexing fields to perform similarity-based retrieval, given some applicable measures of approximation. Typically, the number of features (or dimensions of similarity) is large, so that the search space has a high-dimensionality, and most traditional methods perform poorly. As a test case, we show how the Diamond-tree technique can be used to perform retrievals based on incorrectly or approximately specified values for string fields. Experimental results show that our method can respond to approximately match queries by examining a small portion (1% - 5%) of the database.