Operations Research, Systems Engineering and Industrial Engineering Commons^™

Optimal And Heuristic Resource Allocation Policies In Serial Production Systems, Ramesh Arumugam

All Theses

We have studied the optimal server allocation policies for a tandem queueing system under different system settings. Motivated by an industry project, we have studied a two stage tandem queueing system with arrival to the system and having two flexible servers capable of working at either of the stations. In our research, we studied the system under two different circumstances: modeling the system to maximize throughput without cost considerations, modeling the system to include switching and holding costs along with revenue for finished goods. In the maximizing throughput scenario, we considered two different types of server allocations: collaborative and non-collaborative. …

Effective Vr: Interplay Of Presence, Perception, Fidelity And Transfer Effects In The Development Of Inspection Training Simulators, Sajay Sadasivan

All Dissertations

Increased emphasis on aircraft inspection to ensure aviation safety has resulted in the need for a better trained workforce of aircraft maintenance personnel. Aircraft inspection is mostly visual in nature. Thus it has led to the development of computer simulators to train human inspectors performing the inspection task. However, the lack of immersion and interaction in such simulators has led to limited effectiveness and adoption. Using advances in graphics and virtual reality technology, there is an increase the sense of involvement in using these simulators. Though technology is available to provide higher levels of immersion, it is expensive. It is …

On Step Sizes, Stochastic Shortest Paths, And Survival Probabilities In Reinforcement Learning, Abhijit Gosavi

Engineering Management and Systems Engineering Faculty Research & Creative Works

Reinforcement learning (RL) is a simulation-based technique useful in solving Markov decision processes if their transition probabilities are not easily obtainable or if the problems have a very large number of states. We present an empirical study of (i) the effect of step-sizes (learning rules) in the convergence of RL algorithms, (ii) stochastic shortest paths in solving average reward problems via RL, and (iii) the notion of survival probabilities (downside risk) in RL. We also study the impact of step sizes when function approximation is combined with RL. Our experiments yield some interesting insights that will be useful in practice …

A Model Based Fault Detection And Prognostic Scheme For Uncertain Nonlinear Discrete-Time Systems, Balaje T. Thumati, Jagannathan Sarangapani

Electrical and Computer Engineering Faculty Research & Creative Works

A new fault detection and prognostics (FDP) framework is introduced for uncertain nonlinear discrete time system by using a discrete-time nonlinear estimator which consists of an online approximator. A fault is detected by monitoring the deviation of the system output with that of the estimator output. Prior to the occurrence of the fault, this online approximator learns the system uncertainty. In the event of a fault, the online approximator learns both the system uncertainty and the fault dynamics. A stable parameter update law in discrete-time is developed to tune the parameters of the online approximator. This update law is also …

Distributing Complementary Resources Across Multiple Periods With Stochastic Demand, Shih-Fen Cheng, John Tajan, Hoong Chuin Lau

Research Collection School Of Computing and Information Systems

In this paper, we evaluate whether the robustness of a market mechanism that allocates complementary resources could be improved through the aggregation of time periods in which resources are consumed. In particular, we study a multi-round combinatorial auction that is built on a general equilibrium framework. We adopt the general equilibrium framework and the particular combinatorial auction design from the literature, and we investigate the benefits and the limitation of time-period aggregation when demand-side uncertainties are introduced. By using simulation experiments, we show that under stochastic conditions the performance variation of the process decreases as the time frame length (time …

Neural Network Output Feedback Control Of A Quadrotor Uav, Jagannathan Sarangapani, Travis Alan Dierks

Electrical and Computer Engineering Faculty Research & Creative Works

A neural network (NN) based output feedback controller for a quadrotor unmanned aerial vehicle (UAV) is proposed. The NNs are utilized in the observer and for generating virtual and actual control inputs, respectively, where the NNs learn the nonlinear dynamics of the UAV online including uncertain nonlinear terms like aerodynamic friction and blade flapping. It is shown using Lyapunov theory that the position, orientation, and velocity tracking errors, the virtual control and observer estimation errors, and the NN weight estimation errors for each NN are all semi-globally uniformly ultimately bounded (SGUUB) in the presence of bounded disturbances and NN functional …

Neural-Network-Based State Feedback Control Of A Nonlinear Discrete-Time System In Nonstrict Feedback Form, Pingan He, Jagannathan Sarangapani

Electrical and Computer Engineering Faculty Research & Creative Works

In this paper, a suite of adaptive neural network (NN) controllers is designed to deliver a desired tracking performance for the control of an unknown, second-order, nonlinear discrete-time system expressed in nonstrict feedback form. In the first approach, two feedforward NNs are employed in the controller with tracking error as the feedback variable whereas in the adaptive critic NN architecture, three feedforward NNs are used. In the adaptive critic architecture, two action NNs produce virtual and actual control inputs, respectively, whereas the third critic NN approximates certain strategic utility function and its output is employed for tuning action NN weights …

Long-Term Prediction Of Time Series Using Nne-Based Projection And Op-Elm, Antti Sorjamaa, Yoan Miche, Robert Weiss, Amaury Lendasse

Engineering Management and Systems Engineering Faculty Research & Creative Works

This Paper Proposes a Combination of Methodologies based on a Recent Development -Called Extreme Learning Machine (Elm)- Decreasing Drastically the Training Time of Nonlinear Models. Variable Selection is Beforehand Performed on the Original Dataset, using the Partial Least Squares (Pls) and a Projection based on Nonparametric Noise Estimation (NNE), to Ensure Proper Results by the Elm Method. Then, after the Network is First Created using the Original Elm, the Selection of the Most Relevant Nodes is Performed by using a Least Angle Regression (Lars) Ranking of the Nodes and a Leave-One-Out Estimation of the Performances, Leading to an Optimally Pruned …

A Computer-Based Economic Analysis For Manufacturing Process Selection, Manocher Djassemi

Industrial Technology and Packaging

An important part of financial planning in product development is considering whether the capital expenditures meet volume and cost goals. A good business plan should provide investors with the implications of process selection on the company's bottom line. It is estimated that there are least 1000 manufacturing processes and sub-processes. Considering the number of process choices and quantity of cost data, an economic analysis for process selection may pose a challenge for decision makers. This paper provides an insight to Ashby’s cost modeling method for generating an estimate of unit product cost. The cost model provides a broad indicator for …

Optimal Pruned K-Nearest Neighbors: Op-Knn - Application To Financial Modeling, Q. Yu, A. Sorjamaa, Y. Miche, Amaury Lendasse, Eric Séverin, A. Guillen, F. Mateo

Engineering Management and Systems Engineering Faculty Research & Creative Works

The Paper Proposes a Methodology Called OO-KNN, Which Builds a One Hidden-Layer Feedforward Neural Network, using Nearest Neighbors Neurons with Extremely Small Computational Time. the Main Strategy is to Select the Most Relevant Variables Beforehand, Then to Build the Model using KNN Kernels. Multi response Sparse Regression (MRSR) is Used as the Second Step in Order to Rank Each Kth Nearest Neighbor and Finally as a Third Step Leave-One-Out Estimation is Used to Select the Number of Neighbors and to Estimate the Generalization Performances. This New Methodology is Tested on a Toy Example and is Applied to Financial Modeling. © …

Fuzzy Inference Based Autoregressors For Time Series Prediction Using Nonparametric Residual Variance Estimation, Federico Montesino Pouzols, Amaury Lendasse, Angel Barriga

Engineering Management and Systems Engineering Faculty Research & Creative Works

We Apply Fuzzy Techniques for System Identification and Supervised Learning in Order to Develop Fuzzy Inference based Auto regressors for Time Series Prediction. an Automatic Methodology Framework that Combines Fuzzy Techniques and Statistical Techniques for Nonparametric Residual Variance Estimation is Proposed. Identification is Performed through the Learn from Examples Method Introduced by Wang and Mendel, While the Marquard-Levenberg Supervised Learning Algorithm is Then Applied for Tuning. Delta Test Residual Noise Estimation is Used in Order to Select the Best Subset of Inputs as Well as the Number of Linguistic Labels for the Inputs. Experimental Results for Three Time Series Prediction …

Drop Impact Dynamic Response Study Of Jedec Jesd22-B111 Test Board, Michael Krist, Jianbiao Pan, Andrew Farris, Nicolas Vickers

Industrial and Manufacturing Engineering

Mobile and handheld electronic devices are prone to being dropped. This drop event may result in failure of solder joints inside these devices. The need for RoHS compliant boards coupled with the demand for reliable electronics has resulted in the development of the JEDEC Standard JESD22-B111 to standardize the method of drop testing surface mount electronic components. However, there has been little study on the effects of additional mass on the board and rigidity of the board on drop test reliability. This paper examines the drop impact dynamic responses of the JEDEC JESD22-B111 board. Of interest are the effects of …

The Effect Of Ultrasonic Frequency On Gold Wire Bondability And Reliability, Jianbiao Pan, Minh-Nhat Le, Cuong Van Pham

Industrial and Manufacturing Engineering

This paper presents a systematic study on the effect of 120KHz ultrasonic frequency on the bondability and reliability of fine pitch gold wire bonding. The study was carried out on a thermosonic ball bonder that is allowed to easily switch between ultrasonic frequency of 60KHz and 120 KHz by changing the ultrasonic transducer and the ultrasonic generator. Bonding parameters were optimized through design of experiment methodology for four different cases: 25.4 μm wire at 60 kHz, 25.4 μm wire at 120 kHz, 17.8 μm wire at 60 kHz, and 17.8 μm wire at 120 kHz. The integrity of wire bonds …

Board Level Failure Analysis Of Chip Scale Package Drop Test Assemblies, Nicholas Vickers, Kyle Rauen, Andrew Farris, Jianbiao Pan

Industrial and Manufacturing Engineering

This paper presents the failure analysis results of board level drop tests. In this study, the test vehicle was designed according to the requirements of the Joint Electron Device Engineering Council (JEDEC) drop test board. The test vehicle was assembled with 15 chip scale packages (CSPs) each having 228 daisy-chained 0.5 mm pitch solder joints using Sn-3.0 wt% Ag-0.5 wt% Cu (SAC305) lead free solder. Assemblies were drop tested using three different peak accelerations of 900 G, 1500 G, 2900 G, with 0.7 ms, 0.5 ms, and 0.3 ms pulse durations, respectively. Scanning electron microscopy (SEM) with energy dispersive spectroscopy …

Acuta Enews November 2008, Vol. 37, No. 11

ACUTA Newsletters

In This lssue

From ACUTA Headquarters: FCC lssues Rules to lmplement NET 91 1 ACT........ Jeri Semer, CAE, Exec. Dir

Tech Talk: For Most, the Big Change on Campus ls Wireless.............Kevin Tanzillo, Dux PR

DC Update.............. Dove Ostrom, Washlngton State University

Nominate Now for ACUTA Board of Directors

Be a State/Province Coordinator

Teflon:A Dangerous Part of a High-Risk World.................Frank Bisbee, Wireville USA

Thanks to Exhibitors for 2008

lnfo Links ............. Randy Hayes, Univ. of Northern lowa

Board Report, October....................... Riny Ledgenuood, San Diego State Univ.

Welcome New Members

Check lt Out: RFls/RFPs, Job Postings, Press Releases

Summer Sessions Breakdown Of Course Dates And Time, Paul Savory

Department of Industrial and Management Systems Engineering: Faculty Publications

This report highlights an exhaustive analysis of the hundreds of Summer Session courses to develop an understanding of when they occur during the day and during which of the four academic terms. The analysis shows that the most popular time for scheduling courses is between 9:30 am and noon for each of the four summer session terms. Given the limited number of general purpose classrooms on campus, accommodating faculty preference for rooms, buildings, and computer equipment is sometimes not possible during the most in demand course times. The results also showcase the potential for having summer courses four-day per week …

Navigating From The Local To The Cosmopolitan: Expanding The Scholarship Of Teaching And Learning Commons, Paul Savory, Amy M. Goodburn

Department of Industrial and Management Systems Engineering: Faculty Publications

This collaborative poster highlights how "Expanding‐the‐SOTL‐Commons" is composed of nine collaborating institutions from the US and Canada. Situated within the Carnegie Academy for Scholarship of Teaching and Learning’s Intuitional Leadership Program, our initiative seeks innovative ways to sustain and advance the teaching commons both on our own campuses and across the broader ISSOTL community. Our diverse group of institutions of higher education acknowledges the multiple dimensions of work that communities and individuals contribute to the scholarship of teaching and learning and recognizes that common interest reside in how individuals and institutions enter into this area of scholarship, and how they …

Work In Progress - Enhancing Student-Learning Through State-Of-The-Art Systems Level Design And Implementation, Albert A. Liddicoat, Jianbiao Pan, James G. Harris, Dominic J. Dal Bello, Lynne A. Slivovsky

Industrial and Manufacturing Engineering

The curriculum for undergraduate engineering programs is often partitioned into several courses that are taught in isolation followed by a single culminating senior design or capstone project experience. In the senior design class students being to synthesize the knowledge and skills that they acquired through the engineering curriculum. This paper presents lower and upper division course and curricular changes made to accommodate learning objectives that better prepare students for project-based learning. These learning experiences and skills include: systems level design, experience with state-of-the-art Computer Aided Design (CAD) tools, printed circuit board (PBC) design, design for manufacturability, electronics assembly, project management, …

Alternative Binder Carbide Tools For Machining Superalloys, Daniel J. Waldorf, Michael Stender, Scott Liu, Daniel Norgan

Industrial and Manufacturing Engineering

This study examines the performance of a new class of wear-resistant but economical cutting tools produced by varying the binder composition of standard cemented carbide composites. By replacing some or all of the cobalt binder with rhenium and nickel-based superalloy, a stronger composite tool results, potentially capable of machining heat-resistant superalloys at significantly higher cutting speeds. Sample tools with alternative binder were produced and compared to standard tools bound with cobalt only. Turning experiments on Inconel 718 were run to evaluate wear resistance and tool life for several grades. The experimentation also examined the effects of varying the relative proportions …

Acuta Enews October 2008, Vol. 37, No. 10

ACUTA Newsletters

In This Issue

Nominate Now for 2009-2008 Board of Directors

From the President................. Corinne Hoch, Columbia Univ., ACUTA President

Tech Talk: Why Not Share That Extra Computing Power?................ Kevin Tanzillo, Dux PR

DC Update................. Dave Ostrom, Washington State University

Clemson Researchers' EarthBottle Made fro Plants Will Soon Be in Stores

Info Links...................... Randy Hayes, Univ. of Northern Iowa

The ICT Regulations toolkit: A Resource Worth Checking Out

Loans to Cover Disaster Losses

Board Report................ Riny Ledgerwood, San Diego State Univ., ACUTA Secretary/Treasurer

Check It Out: Press Releases... Job Postings... RFIs/RFPs

OCS Federation Submarien Cable Completed

Welcome New Members

Function-Based Failure Propagation For Conceptual Design, Daniel A. Krus, Katie Grantham

Engineering Management and Systems Engineering Faculty Research & Creative Works

When designing a product, the earlier the potential risks can be identified, the more costs can be saved, as it is easier to modify a design in its early stages. Several methods exist to analyze the risk in a system, but all require a mature design. However, by applying the concept of “common interfaces” to a functional model and utilizing a historical knowledge base, it is possible to analyze chains of failures during the conceptual phase of product design. This paper presents a method based on these common interfaces to be used in conjunction with other methods such as risk …

Determining The Number Of Bp Neural Network Hidden Layer Units, Huayu Shen, Zhaoxia Wang, Chengyao Gao, Juan Qin, Fubin Yao, Wei Xu

Research Collection School Of Computing and Information Systems

This paper proposed an improved method to contrapose the problem which is difficult to determine the number of BP neural network hidden layer units. it is proved that the method is efficeient in reducing the frequency of the test through experients, and improves the efficiency of determining the best number of hidden units, which is more valuable in the applications.

Spreadsheet Data Resampling For Monte-Carlo Simulation, Thin Yin Leong, Wee Leong Lee

Research Collection School Of Computing and Information Systems

The pervasiveness of spreadsheets software resulted in its increased application as a simulation tool for business analysis. Random values generation supporting such evaluations using spreadsheets are simple and yet powerful. However, the typical approach to Monte-Carlo simulations, which is what simulations with stochasticity are called, requires significant amount of time to be spent on data collection, data collation, and distribution function fitting. In fact, the latter can be overwhelming for undergraduate students to learn and do properly in a short time. Resampling eliminates both the need to fit distributions to the sample data, and to perform the ensuing tests of …

Engineering Management Framework In Support Of Modeling & Simulation Application For Domain Specific Procurement, Thomas Guy Litwin

Engineering Management & Systems Engineering Theses & Dissertations

A strategic process is desirable for project-based organizations in order for them to be efficient and effective when developing Modeling & Simulation (M&S) systems. This thesis proposes an overarching process that combines traditional M&S and Engineering Management methodologies in a new framework to support M&S organizations during the procurement process.

This thesis proposes both a Strategic Project Management Process (SPMP) and a systems engineering process for M&S federation development projects. The systems engineering process utilizes the artifacts of Model Driven Architecture (MDA) to support building M&S federations driven by operational requirements. Detailed research of this systems engineering process revealed a …

A Model Based Fault Detection Scheme For Nonlinear Multivariable Discrete-Time Systems, Balaje T. Thumati, Jagannathan Sarangapani

Electrical and Computer Engineering Faculty Research & Creative Works

In this paper, a novel robust scheme is developed for detecting faults in nonlinear discrete time multi-input and multi-output systems in contrast with the available schemes that are developed in continuous-time. Both state and output faults are addressed by considering separate time profiles. The faults, which could be incipient or abrupt, are modeled using input and output signals of the system. By using nonlinear estimation techniques, the discrete-time system is monitored online. Once a fault is detected, its dynamics are characterized using an online approximator. A stable parameter update law is developed for the online approximator scheme in discrete-time. The …

Acuta Enews September 2008, Vol. 37, No. 9

ACUTA Newsletters

In This lssue

From ACUTA Headquarters: Ramped-up Schedule................Jeri A. Semer, CAE, Executive Director

New Section on ACUTA Website

Audio Seminar: Technology-Related lmptications of the 2008 Higher Ed Opportunities Act

Tech Talk: ACUTA Member Schools Are Mostly in the Green ........ Kevin Tanzillo, Dux PR

DC Update ......... Dave Ostrom, Washington State University

lP PBX Witt Reign: Converged PBX Will Slowly Fade Away........................Gary Audin, Delphi, lnc.

Get Your ACUTA Logo here

Look on the Light Side

lnfo Links ............. Randy Hayes, Univ. of Northern lowa

Welcome New Members

Check lt Out: RFls/RFPs, Job Postings, Press Releases

Research On Application Of Rfid In Smart Gate Of Container Terminal, Jiying Song

Faculty Publications - George Fox Digital

With the rapid development of logistics, the operation and management of logistics enterprise becomes more complex and changeable. Under the background, it has become the urgent problem for the policy-maker to resolve that how to construct the highly effective information system, how to apply the information technology, how to carry on the all-the-way tracking, how to enhance the collection, processing and serviceability of information and how to shorten the exchange of information and the operation time. Along with the rapidly growth of domestic as well as domestic and international volume of trade and the fast development of container fleet, that …

A Confidence Paradigm For Classification Systems, Nathan J. Leap

Theses and Dissertations

There is no universally accepted methodology to determine how much confidence one should have in a classifier output. This research proposes a framework to determine the level of confidence in an indication from a classifier system where the output is or can be transformed into a posterior probability estimate. This is a theoretical framework that attempts to unite the viewpoints of the classification system developer (or engineer) and the classification system user (or war-fighter). The paradigm is based on the assumptions that the system confidence acts like, or can be modeled as a value and that indication confidence can be …

Network-Centric Localization In Manets Based On Particle Swarm Optimization, Raghavendra V. Kulkarni, Ganesh K. Venayagamoorthy, Ann K. Miller, Cihan H. Dagli

Engineering Management and Systems Engineering Faculty Research & Creative Works

There exist several application scenarios of mobile ad hoc networks (MANET) in which the nodes need to locate a target or surround it. Severe resource constraints in MANETs call for energy efficient target localization and collaborative navigation. Centralized control of MANET nodes is not an attractive solution due to its high network utilization that can result in congestions and delays. In nature, many colonies of biological species (such as a flock of birds) can achieve effective collaborative navigation without any centralized control. Particle swarm optimization (PSO), a popular swarm intelligence approach that models social dynamics of a biological swarm is …

Sr-2: A Hybrid Algorithm For The Capacitated Vehicle Routing Problem, Angel A. Juan, Javier Faulin, Josep Jorba, Barry Barrios, Scott Erwin Grasman

Engineering Management and Systems Engineering Faculty Research & Creative Works

During the last decades a lot of work has been devoted to develop algorithms that can provide near-optimal solutions for the capacitated vehicle routing problem (CVRP). Most of these algorithms are designed to minimize an objective function, subject to a set of constraints, which typically represents aprioristic costs. This approach provides adequate theoretical solutions, but they do not always fit real-life needs since there are some important costs and some routing constraints or desirable properties that cannot be easily modeled. In this paper, we present a new approach which combines the use of Monte Carlo simulation and parallel and grid …