WSC 2001 Final Abstracts

Manufacturing Applications Track

Monday 10:30:00 AM 12:00:00 PM
Role of Simulation in Industries

Chair: Chen Zhou (Georgia Institute of Technology)

The Definition and Potential Role of Simulation within an Aerospace Company
Craig A. Murphy and Terrence D. Perera (Sheffield Hallam University)

Simulation software has reached a technological level that provides high flexibility and integration capabilities necessary for product design, development and manufacturing efficiency. Within the manufacturing industry, this simulation potential has not been fully recognized, although it is now becoming a matter of interest through the documented benefits it has provided. This paper discusses the issues of simulation definition, selection and integration with both business systems and each other. This also discusses the practical difficulties a business would encounter in the development of a fully digital environment through simulation integration, and data management.

Biotech Industry: Simulation and Beyond
Prasad V. Saraph (Bayer Corporation)

The Biotech Industry is relatively new to the use of simulation techniques. This paper discusses an application of discrete event simulation in a continuous process Biotech manufacturing facility of Bayer Corporation at Berkeley. The SIGMA® simulation model imitating demand and supply of a critical utility (WFI - Water For Injection) was used to analyze the WFI shortage. The model has been in use for the last year and it has effectively eliminated WFI shortages. Based on this analysis, a set of guidelines was designed to ensure better availability of this critical utility. The model initiated a project to reduce the consumption of WFI. The model was also used for strategic capacity analysis and to assess the impact of capital projects on future budgetary plans. This whole project was completed in two months and resulted in direct benefits worth $ 1,100,000.

A Simulation Case Study of Production Planning and Control in Printed Wiring Board Manufacturing
Heidi M. E. Korhonen, Jussi Heikkilä, and Jon M. Törnwall (TAI Research Centre, Helsinki University of Technology)

Production planning and control in printed wiring board (PWB) manufacturing is becoming more difficult as PWB's technology is developing and the production routings become more complex. Simultaneously, the strategic importance of delivery accuracy, short delivery times, and production flexibility is increasing with the highly fluctuating demand and short product life cycles of end products. New principles, that minimize throughput time while guaranteeing excellent customer service and adequate capacity utilization, are needed for production planning and control. Simulation is needed in order to develop the new principles and test their superiority. This paper presents an ongoing simulation project that aims at developing the production planning and control of a PWB manufacturer. In the project, a discrete event simulation model is built of a pilot case factory. The model is used for comparing the effect of scheduling, queuing rules, buffer policies, and lot sizes on customer service and cost efficiency.

Monday 1:30:00 PM 3:00:00 PM
Enterprise-wide Modeling

Chair: Jeffrey W. Hermann (University of Maryland)

A Taxonomy of a Living Model of the Enterprise
Larry Whitman, Kartik Ramachandran, and Vikram Ketkar (Wichita State University)

A designer has a choice of many models, methods, frameworks, and architectures. There is little consistency between these terms among researchers. Some of the most widely used architectures and frameworks are described with definitions and concepts that distinguish them clearly. This paper proposes a clear definition of these terms, a clear distinction between these and a methodology that will significantly aid in the comparison and evaluation of various enterprise models. A direct benefit of this research is a more clear presentation of how the enterprise modeling community uses enterprise models.

Distributed Simulation: An Enabling Technology for the Evaluation of Virtual Enterprises
Jayendran Venkateswaran, Mohammed Yaseen Kalachikan Jafferali, and Young-Jun Son (The University of Arizona)

This paper presents an application distributed simulation to the evaluation of virtual enterprises. Each company or candidate can use a simulation of its facilities to determine if it has the capability to perform its individual function in the virtual enterprise. Then, these simulations can be integrated into a distributed simulation of the complete enterprise, and used to predict the viability and profitability of the proposed product collaboration. In this paper, a prototype distributed simulation for such a purpose is presented. First, information flows as well as material flows among members in a virtual enterprise are identified using IDEF? a formal function modeling method. Sequences of the identified functions are then presented using the finite state automata formalism. These interactions are then implemented for a commercial simulation package. Finally, a distributed simulation composed of three individual simulations is successfully tested across platforms over both the internet and the local area network.

Ford's Power Train Operations – Changing the Simulation Environment
John Ladbrook (Ford Motor Company Limited ) and Annette Januszczak (Ford Motor Company Limited)

This paper discusses the changes required to Ford's Power Train Operations (PTO) simulation environment to ensure the maximum benefit from the investment in simulation. Three key elements were identified as essential to maximizing use. These were Availability, Support, and the right Tools for the Job. The background driving the change was that Simulation had been a key tool in the planning and process improvement of PTO Manufacturing Engineering facilities since the early 80's. The original deployment allowed users to be responsible for the selection, purchase and maintenance of their own systems. This resulted in low utilization, high unit cost and a diversity of products. The achievement was to transform an isolated approach taken on two continents into a single one across 5 continents, while significantly reducing the unit cost. The method was to select a single software solution that could be distributed across the Ford Intranet to anyone in PTO.

Monday 3:30:00 PM 5:00:00 PM
Simulation in Shipyards

Chair: Young-Jun Son (University of Arizona)

Simulation of Shipbuilding Operations
Charles McLean and Guodong Shao (National Institute of Standards and Technology )

This paper discusses the objectives and requirements for a shipbuilding simulation. It presents an overview of a generic simulation of shipbuilding operations. The shipbuilding simulation model can be used as a tool to analyze the schedule impact of new workload, evaluate production scenarios, and identify resource problems. The simulation helps identify resource constraints and conflicts between competing jobs. The simulation can be used to show expected results of inserting new technologies or equipment into the shipyard, particularly with respect to operating costs and schedule impact. The use of DOD High Level Architecture (HLA) and Run Time Infrastructure (RTI) as an integration mechanism for distributed simulation is also discussed briefly.

Hierarchical Modeling of a Shipyard Integrated with an External Scheduling Application
Ali S. Kiran, Tekin Cetinkaya, and Juan Cabrera (Kiran Consulting Group)

This paper presents a hierarchical approach on the simulation of large-scale discrete event systems used recently by Kiran Consulting Group (KCG) to model shipyard operations. Because of the dynamic, stochastic and complex nature of the shipbuilding processes, bottleneck identification and estimation of the impact of new technology implementation is extremely difficult to derive via analytical methods. The simulation model of a large-scale discrete event system can be considered as a collection of sub-systems, which are represented by the simulation models that are independently created, modified, and saved. This approach also includes methods that integrate these submodels into an overall model in order to run different scenarios and identify global performance measures.

Discrete Simulation Development for a Proposed Shipyard Steel Processing Facility
Daniel L. Williams (Electric Boat Corporation), Daniel A. Finke (The Pennsylvannia State University), D. J. Medeiros (Department of Industrial & Manufacturing Engineering) and Mark T. Traband (Applied Research Laboratories)

This paper describes the efforts required to convert conceptual designs and undefined processes for a proposed advanced steel processing shipyard facility into a discrete event simulation. Modeling of a completely non-existent entity poses many difficulties, yet the results can still be beneficial. The lack of actual production data and corresponding business rules, causes an in-depth review of all available information combined with that which can be extrapolated from vendor specification sheets or human experience. Most of the equipment required for this advanced processing facility will be custom built to suit the needs of this highly technical complex. This facility which will ultimately support construction of vessels, was driven by high expectations of improved production efficiencies. The model is expected to support not only the pre-construction design phases of the building, but also to serve as a post-construction production planning tool.

Tuesday 8:30:00 AM 10:00:00 AM
Process Control and Improvement

Chair: Farhad Azadivar (University of Massachusetts Dartmouth)

Prediction of Process Parameters for Intelligent Control of Freezing Tunnels Using Simulation
Sreeram Ramakrishnan, Richard A. Wysk, and Vittaldas V. Prabhu (Pennsylvania State University)

Various analytical and empirical methods assuming the existence of steady state and requiring homogenous properties of the product have been used with limited success in estimating freezing times in the food processing industry. Irrespective of the method adopted for estimating freezing time requirements, a critical process issue that needs to be considered is that of system control. Simulation models suggest that a feed-forward control strategy, as discussed in this paper, can be used to control a freezing tunnel and obtain considerable energy savings while ensuring ‘appropriate’ freezing of all products. The control strategy discussed in this paper, involves the continuous monitoring of product input and controlling either or both of the refrigerant flow and conveyor speed. The primary objective of this paper is to demonstrate the use of simulation to predict process parameters for ‘intelligent control’ of freezing tunnels, and provide an estimate of potential energy savings.

Quantifying Simulation Output Variability Using Confidence Intervals and Statistical Process Control
Amy Jo Naylor (Corning Inc.)

Two types of variability can occur in model output: variability between replications and variability within each replication. The objective of the model combined with the type of output variability determines which tool is more appropriate for output analysis. Many output analysis techniques are used to translate simulation model results into a format that answers the model objective. This paper compares two tools for output analysis: confidence intervals and statistical process control. Each tool quantifies a different type of variaiton from the model results. As such, statistical process control is applied beyond monitoring the consistency of run data. A supply chain example with one factory, multiple parts, and multiple distribution centers is used throughout the paper to illustrate these concepts.

Plate/Sheet Nest Release and Throughput Simulation for WSC ’01
Leland D. Weed (The Raymond Corporation)

The BT/Raymond Corporation is a manufacturer of narrow aisle electric fork-trucks and uses two Delmia simulation software packages: UltraArc® and Quest®. In the Greene NY facility, one of the Quest® simulations shows the start of the fabrication process. The plate/sheet line is a group of machines that punch, machine, profile, and form steel material ranging in thickness from 0.030” to 1.250”. Since each product is built to customer order, the mix of parts to produce on the line is continually changing. The simulation of this process reads the data that schedules the work for the various machines, then runs the line showing capacity and throughput issues a day ahead of the factory floor run. The data that the model reads can also be changed to experiment with different product build quantities.

Tuesday 10:30:00 AM 12:00:00 PM
Decision Making using Simulation

Chair: Durk-Jouke van der Zee (University of Groningen)

Solving Sequential Decision-Making Problems Under Virtual Reality Simulation System
Yang Xianglong, Feng Yuncheng, and Li Tao (Beijing University of Aeronautics & Astronautics) and Wang Fei (Institute of International Economy,State Development Planning Commission of China)

A large class of problems of sequential decision-making can be modeled as Markov or Semi-Markov Decision Problems, which can be solved by classical methods of dynamic programming. However, the computational complexity of the classical MDP algorithms, such as value iteration and policy iteration, is prohibitive and will grow intractably with the size of problems. Furthermore, they require for each action the one step transition probability and reward matrices, which is often unrealistic to obtain for large and complex systems. Here, we provide the decision-maker a sequential decision-making enviroment by establishing a virtual reality simulation system, where the uncertainty property of system can also be shown. In order to obtain the optimal or near optimal policy of sequential decision problem, simulation optimization algorithms as infinitesimal perturbation analysis are applied to complex queuing systems. We present a detailed study of this method on the sequential decision-making problem in Boeing-737 assembling process.

Modelling and Improving Human Decision Making with Simulation
Stewart Robinson, Thanos Alifantis, and Robert Hurrion (Warwick Business School), John Edwards (Aston Business School), John Ladbrook (Ford Motor Company) and Tony Waller (Lanner Group)

Modelling human interaction and decision-making within a simulation presents a particular challenge. This paper describes a methodology that is being developed known as 'knowledge based improvement'. The purpose of this methodology is to elicit decision-making strategies via a simulation model and to represent them using artificial intelligence techniques. Further to this, having identified an individual's decision-making strategy, the methodology aims to look for improvements in decision-making. The methodology is being tested on unplanned maintenance operations at a Ford engine assembly plant.

Tuesday 1:30:00 PM 3:00:00 PM
Manufacturing Controls

Chair: Amarnath Banerjee (Texas A&M University)

Understanding the Fundamentals of Kanban and CONWIP Pull Systems Using Simulation
Richard P. Marek (Ford Motor Company), Debra A. Elkins (General Motors) and Donald R. Smith (Texas A&M University)

This paper presents an introductory overview and tutorial in simulation modeling and control of serial Kanban and CONWIP (CONstant Work In Process) pull systems using ARENA/SIMAN 3.5/4.0. Card level estimation is discussed for both types of pull systems, and a heuristic method to adjust card levels controlling system WIP (Work In Process) is provided. The objective is to present a tutorial for students and practicing engineers familiar with the basics of simulation, but unfamiliar with pull system fundamentals.

Real-Time Adaptive Control of Multi-Product Multi-Server Bulk Service Processes
Durk-Jouke van der Zee (University of Groningen)

Batching jobs in a manufacturing system is a very common policy in most industries. Main reasons for batching are avoidance of setups and/or facilitation of material handling. Batch processing systems often consist of multiple machines of different types for the range and volumes of products that have to be handled. Building on earlier research in aircraft industry, where the process of hardening synthetic aircraft parts was studied, we discuss a new heuristic for the dynamic scheduling of these types of systems. It is shown by an extensive series of simulation experiments that the new heuristic outperforms existing heuristics for most system configurations.

Improving Simulation Model Adaptability with a Production Control Framework
Sean M. Gahagan and Jeffrey W. Herrmann (Institute for Systems Research)

Simulation models provide a powerful tool for the analysis of manufacturing systems, but their utility beyond the design stage of the system life cycle is hampered by the high cost of model maintenance. To reduce this cost, models must be made more adaptable. We believe that adaptability can be increased by separating the flow of material from the flow of information through a model system, especially with respect to changes related to production control. Coordination of these flows, however, requires a production control framework. In this paper, we propose a three-level, hierarchical production control framework and define the elements necessary to implement it in a simulation model. We demonstrate the use of this approach by considering a simple flow shop undergoing production control changes. We define the parameters of the shop using the framework and implement the changes with little effort.

Tuesday 3:30:00 PM 5:00:00 PM
Analysis of Manufacturing Systems

Chair: Chen Zhou (Georgia Institute of Technology)

Computer Simulation Analysis of Electricity Rationing Effects on Steel Mill Rolling Operations
Thomas F. Brady (Purdue University North Central)

This paper presents an application of computer simulation as a policy analysis tool for the electric utility industry. In the last decade, the amount of electricity generation capacity has remained constant while demand for electricity has been increasing. This situation puts industrial electricity users, those who use large highly varying quantities of electricity in potentially risky production and financial situations. In this paper, we describe a computer simulation model that examines the electricity requirements of a steel mill in a constrained electricity supply environment. By using simulation, we develop and analyze policies that quantify the costs and benefits of collaborative strategies for efficient electricity usage from both perspectives.

A Practical Bottleneck Detection Method
Christoph Roser, Masaru Nakano, and Minoru Tanaka (Toyota Central Research and Development Laboratories)

This paper describes a novel method for detecting the bottleneck in a discrete event system by examining the average duration of a machine being active for all machines. The machine with the longest average uninterrupted active period is considered the bottleneck. The method is widely applicable and also capable of analyzing complex and sophisticated systems. The results are highly accurate, distinguishing between bottleneck machines and non-bottleneck machines with a high level of confidence. This approach is very easy to use and can be implemented into existing simulation tools with little effort, requiring only an analysis of the log file which is readily available by almost all simulation tools. This method satisfies not only academic requirements with respect to accuracy but also industry requirements with respect to usability.

Using Simulation and Neural Networks to Develop a Scheduling Advisor
Thanos Alifantis and Stewart Robinson (University of Warwick)

The research using artificial intelligence and computer simulation introduces a new approach for solving the job shop scheduling problem. The new approach is based on the development of a neural network-scheduling advisor, which is trained using optimal scheduling decisions. The data set, which is used to train the neural network, is obtained from simulation experiments with small-scale job shop scheduling problems. The paper formulates the problem and after a review of the current solution methods it describes the steps of a new methodology for developing the neural network-scheduling advisor and collecting the data required for its training. The paper concludes by mentioning the expected findings that can be used to evaluate the degree of success of the new methodology.

Wednesday 8:30:00 AM 10:00:00 AM
Automation in Modeling

Chair: Chen Zhou (Georgia Institute of Technology)

Using Automation for Finishing Room Capacity Planning
Ryan Heath Melton (CMD Systems) and C. Thomas Culbreth, Stephen D. Roberts, and Jeffrey A. Joines (North Carolina State University)

Capacity planning of a furniture finishing system using both deterministic analysis and stochastic simulation is conveniently performed with the aid of ActiveX Automation Users interactively build a complete model of a finishing system with an Excel interface, which creates a deterministic model. The spreadsheet decouples data input from the simulation model construction and execution, and provides a user-friendly tool for analyzing a finishing system. Using the spreadsheet, simulation data is provided to the deterministic model, and an Arena simulation model and animation of individual finishing line operations is constructed through ActiveX automation. A manufacturing manager unfamiliar with modeling techniques can use the interface to plan the finishing system and conduct simulation experiments with various input parameters such as line loading techniques, operations balancing, and line speeds. Through the interface, results from the simulation can be used in an iterative process to analyze and refine design parameters of the finishing line.

Computer-Aided Manufacturing Simulation (CAMS) Generation for Interactive Analysis – Concepts, Techniques, and Issues
Boonserm Kulvatunyou and Richard A. Wysk (Pennsylvania State University)

Simulation model is usually developed as a one-time use analytical model by a system analyst (usually from external firm) rather than for a routine and interactive use by a shop floor engineer. This is because it usually takes longer time to generate a result from the simulation, and the simulation model of manufacturing system is usually too sophisticated and time consuming to use as an interactive tool by the manufacturing/production engineer. A CAMS reduces this complication by encapsulating the ‘complicated-logic’ and automating the ‘tedious data-acquisition’ with a more user-friendly interface like a spreadsheet or database input form. This paper describes how CAMS can automatically generate a simulation model; specifically, techniques and issues to structure the model to hide those tasks, so that it is a user-friendly interactive decision support with minimal amount of automation code. The paper concludes with a capacity analysis example from the real industry.

Database Driven Factory Simulation: A Proof-of-Concept Demonstrator
Lars G. Randell and Gunnar S. Bolmsjö (Lund University)

The paper presents a database-based method to reduce the development time and project lead-time for large discrete-event simulation models of entire factories. The database used to automatically generate and drive the simulation model is a copy of the production planning database. A set of proof-of-concept tools and a database have been generated to verify the method and it has been shown that it is feasible to run a simulation using the production planning data as the only information source. The software developed is modular and designed to work in heterogeneous environments. The method is expected to reduce the modeling and maintenance effort considerably when modeling entire factories. The method will result in a holistic and fairly accurate assessment of performance measures for an entire factory.

Wednesday 10:30:00 AM 12:00:00 PM
General Manufacturing Applications

Chair: Larry E. Whitman (Wichita State University)

Feasibility for Automatic Data Collection
Neil H. Robertson and Terrence Perera (Sheffield Hallam University)

It is argued that the data collection process is the most crucial and time consuming stage in the model building process. This is primarily due to the influence that data has in providing accurate simulation results. Data collection is an extremely time consuming process predominantly because the task is manually orientated. Hence, automating this process of data collection would be extremely advantageous. This paper presents how simulation tools could utilize the Corporate Business Systems as the potential source for simulation data. Subsequently a unique interface could be developed and implemented to provide this data directly to the simulation tool. Such an interface would prove to be an invaluable tool for users of simulation.

A Virtual Environment for Simulating Manufacturing Operations in 3D
Ravi Chawla and Amarnath Banerjee (Texas A&M University)

This paper presents a method for simulating basic manufacturing operations (unload, load, process, move, and store) in a 3D virtual environment. The virtual environment provides a framework for representing a facility layout in 3D, which encapsulates the static and the dynamic behavior of the manufacturing system. The 3D manufacturing objects in the facility are mapped with the nodes in the framework. The framework, a modified scenegraph structure, is a tree structure, which can be manipulated by updating the parent-child relationships and the transformation matrix to simulate the basic manufacturing operations. The method can be easily extended to represent more specific manufacturing operations.

[ Return to Top | Return to WSC '01 Program ]