INFORMS Atlanta 1996 Sponsor: INFORMS Section on Manufacturing & Service Operations Management

SC18.1 Effects of Combined Production & Maintenance Scheduling on Meeting Customer Due Dates Thomas W. Sloan, George Shanthikumar --- Univ. of CA, Haas Sch. of Bus., S545 Stud. Serv. Bldg. 1900, Berkeley, CA 94720-1900, (tsloan@haas.berkeley.edu)

We address the problem of scheduling production and maintenance for¨ a single-machine manufacturing system producing multiple products¨ with stochastically ordered due dates. We find conditions under¨ which intuitively appealing, monotone policies are optimal.¨ Numerical examples illustrate how different parameters of the model¨ affect the optimal policy.

SC18.2 A Model for a Multi-Echelon Service Parts Logistics System with Unscheduled & Scheduled Repair Customers Morris A. Cohen, Yu-Sheng Zheng, Yunzeng Wang --- Univ. of PA, Wharton Sch., OPIM Dept., 1300 SH-DH, Philadelphia, PA 19104 , (cohen@opim.wharton.upenn.edu)

We model the operations of a logistics network which provides and¨ repairs service parts for world-wide customers for high technology¨ products. The system is complicated by the interaction of 2¨ customer classes, one with scheduled repairs and 'negative'¨ leadtimes. The model can be used to optimally allocate inventory¨ within the network.

SC18.3 Effective Strategy for Inspection Sampling in the Presence of Inter- & Intra-Lot Variations Kyle D. Chen, Ram Akella --- Stanford Univ., IEEM Dept., Stanford, CA 94305 , (kyle.chen@stanford.edu)

We formulate a model to design inspection sampling strategies based¨ on semiconductor fab-lines. Unlike many traditional models, our¨ model takes into account the lot-to-lot variations as well as the¨ wafer-to-wafer variations and minimizes the lots at risk.

SC19.1 Global Operations, Business Strategy & Firm Performance M. Therese Flaherty --- Univ. of PA, Dept. of OPIM, The Wharton Sch., Philadelphia, PA 19104 , (flaherty@opim.wharton.upenn.edu)

This presentation empirically examines the determinants of firm and¨ business performance for large firms in several broadly-defined¨ international industries between 1977 and 1994. The relative effects¨ of business strategy and global operations strategy, firm-level and¨ business level management and industry conditions are estimated.

SC19.2 Neo-Operations Strategy: An Empirical Study of Global Manufacturing Strategies Aleda V. Roth --- Univ. of NC, Kenan-Flagler Bus. Sch., Dept. of Op. Mgmt., Chapel Hill, NC 27599-3490, (rotha.bsacdl.@mhs.unc.edu)

Neo-operations strategy links combinative competitive capabilities¨ and advanced manufacturing technology under the rubric of economies¨ of knowledge. Competitive progression theory is proposed to explain¨ how sustainable competitive capabilities are accumulated. A sample¨ of 1200 manufacturers in North America, Japan and Europe is used to¨ test hypotheses linking progression theory, technology and people.

SC19.3 Evaluation of Supply Chain Structures Through Modularization & Postponement Ricardo Ernst --- Georgetown Univ., Bus. Sch., Washington, DC 20057 , (ernstr@gunet.georgetown.edu)

This presentation introduces a conceptual framework for evaluating¨ different supply chain structures in the context of modularization¨ (linked to inbound logistics through the manufacturing process) and¨ postponement (linked to outbound logistics). The analysis is used to¨ quantify the total cost for utilizing a particular supply chain¨ structure and allows an objective comparison.

SC19.4 Operations & Logistics Strategies of Small- & Medium-Sized US Firms Markus Biehl, Edmund Prater, Michael Smith, Soumen Ghosh --- GA Inst. of Tech., Sch. of Mgmt., Atlanta, GA 30332-0520, (markus.biehl@mgt.gatech.edu)

We examine ways that small- and medium-sized US firms are expanding¨ into Europe. We investigate the evolution of these firms' operations¨ and logistics strategies as they expand and mature in a global¨ market. These strategies are compared with those of larger firms and¨ multinationals to assess their differences.

SD18.1 Effective Strategy for Multi-Stage Inspection Planning in Semi- Conductor Fab-Lines Kyle D. Chen, Ram Akella --- Stanford Univ., IEEM Dept., Stanford, CA 94305 , (kyle.chen@stanford.edu)

We formulate a multi-stage inspection planning model based on setups¨ in real fab-lines, taking into account capacity constraints and¨ queueing effects. Employing a mixed FCFS-LCFS policy, we approximate¨ the queueing delays in the 'time-to-signal' for process shifts. We¨ further develop algorithms to obtain optimal/near-optimal policies¨ for scheduling inspections.

SD18.2 Adaptive Sampling for Yield Management in Semiconductor Manufacturing William Kuo, George Shanthikumar, Ram Akella --- Univ. of CA, Dept. of IE/OR, Berkeley, CA 94720 , (wkuo@macsmtp.kla.com wykuo@cimsim.berkeley.edu)

We describe new models for describing adaptive sampling for¨ excursion monitoring in semiconductor manufacturing, characterized¨ by units described in terms of lots and wafers. We describe the¨ initial results including the sufficient statistics, policy¨ structures and results for a typical fab.

SD18.3 Optimal Wafer Inspection Strategy with Learning Effects Eric H. Wang, Ram Akella --- Stanford Univ., Dept. of EES, Terman Engineering Ctr. 311, Stanford, CA 94305 , (ericwang@leland.stanford.edu)

Traditional static SPC models cannot capture the continuous learning¨ phenomena that characterize the wafer fabrication process,¨ particularly during the ramp-up phase. As part of the CSM focus¨ study, we have developed a dynamic cost-based model that extends the¨ classic economic design of SPC strategy to incorporate learning¨ effects...

SD18.4 Financial Justification for Equipment Purchases in Semiconductor Fabs Marius Holton, Ram Akella --- Stanford Univ., Dept. of EES, Stanford, CA 94305 , (marius@leland.stanford.edu)

We describe the way the semiconductor manufacturing industry is¨ currently evaluating purchases of new production and inspection¨ equipment. A preliminary evaluation model, taking into account how¨ flexibility in decisions should be evaluated, is described. The¨ limitations of CAPM and the value of dynamic financial theory in the¨ financial justification and investment evaluationn analysis are also¨ discussed.

SD19.1 Continuous Review Inventory Models Where Random Lead Times Depend on Lot Size & Reserved Capacity Yigal Gerchak, James H. Bookbinder, Metin Cakanyildirim --- Univ. of Waterloo, Dept. of MS, Waterloo, Ontario, , Canada N2L 3G1

The processing time of a lot is stochastically proportional to its¨ size, or to some function thereof. In the latter case, we obtain¨ closed-form solutions (no iterations required). We also consider the¨ case where a customer can secure shorter processing times by¨ reserving capacity at the manufacturer's facility.

SD19.2 The Economic Bottleneck in Complex Production Settings Arnold Buss, Stephen R. Lawrence --- Naval Postgrad. Sch., OR Dept., Glasgow Hall, Monterey, CA 93943-5000,

In previous work, we have explored the concept of the economic¨ bottleneck in serial production systems. Here, we extend the concept¨ to assembly systems, distribution systems and network systems, and¨ demonstrate how production control policies affect the management of¨ economic bottlenecks.

SD19.3 Modeling the Benefits of Supply Chain Improvement in Practice Charles Corbett --- Vanderbilt Univ., Owen Grad. Sch. of Mgmt., Nashville, TN 37203 , (corbetcj@ctrvax.vanderbilt.edu)

A multinational chemical company has undertaken various efforts,¨ e.g., SKU rationalization and closer coordination with customers, to¨ improve the structure of its supply chain. True costs and benefits¨ of these improvements, and where in the supply chain they accrue,¨ are unclear. We develop a supply chain model for this company to¨ better understand the impact of these changes and the nature of the¨ trade-offs involved.

SD19.4 Price & Reliability in Supplier Relations: A Newsboy Analysis Nicholas C. Petruzzi, Maqbool Dada, Leroy B. Schwarz --- Univ. of IL, 350 Commerce W Bldg., 1206 S 6th St., Champaign, IL 61820 , (petruzzn@mgmt.purdue.edu)

We develop the optimal policy for a firm that purchases its product¨ from some subset of unreliable suppliers. We then analyze how a¨ given supplier's per-unit price and delivery reliability affects its¨ competitiveness. We find that the price 'qualifies' the supplier,¨ while reliability 'wins' a qualifying supplier its market share.

SE19.1 New Paradigms for Responding to Requests for Quotations in Electronic Commerce Raj Veeramani --- Univ. of WI, 1513 University Ave., Dept. of IE, Madison, WI 53706-1572, (raj@ie.engr.wisc.edu)

The ability to respond quickly to RFQs can provide manufacturing¨ companies with a significant competitive advantage. We present¨ industry applications and underlying concepts of fundamentally new¨ approaches that can reduce the RFQ processing lead time from days to¨ seconds.

SE19.2 An Analytical Framework for the Flexibility of Mixed-Model Assembly Lines Begue Christophe --- Phillips Corp., , ,

This presentation concentrates on the relationship between the¨ general architecture of an assembly line, the complexity of the¨ environment in which it is operated and its performance. In¨ particular, we focus on the comparison of the relative ability of¨ different assembly line architectures to operate in the context of¨ various levels of environment complexity.

SE19.3 Managing an Automated Assembly System with Deteriorating Machines Stephen M. Gilbert --- Case Western Reserve Univ., Weatherhead Sch. of Mgmt., 10900 Euclid Ave., Cleveland, OH 44106-7235, (gilberts@pyrite.som.cwru.edu)

We address a problem that arises when products are processed by a¨ series of machines, each of which deteriorates over time. We¨ describe and model a situation where interchangeable assembly¨ machines are selected and set up for a job, and no adjustments can¨ be made during processing. We then incorporate opportunities to¨ continuously monitor and adjust machines during processing.

SE19.4 Productivity & Quality in Low-Inventory, Lean Production Systems John W. Boudreau, David Juran, John McClain, Kenneth Schultz, Joseph Thomas --- Cornell Univ., Malott Hall, Ithaca, NY 14853-4201,

Everyone 'knows' that JIT systems improve productivity and quality,¨ but why AND how? This research investigates low-inventory production¨ systems experimentally to discover what effect different work¨ organizations and inventory control rules have on both productivity¨ and quality. Personality variables, team-building methods and¨ buffer-control rules are studied...

MA18.1 Order Fulfillment Windows in an Assemble-to-Order System with Stochastic Leadtimes Jing-Sheng Song, Susan H. Xu, Bin Liu --- Univ. of CA, Grad. Sch. of Mgmt., Irvine, CA 92717 , (jssong@uci.edu)

We study a multi-item production and inventory system in which¨ individual items are made to stock but final products are customized¨ and assembled quickly to customer orders. Each item is controlled by¨ an independent base-stock policy and the replemishment leadtimes are¨ stochastic. For any given policy, we derive the key performance¨ measures...

MA18.2 Bounds & Heuristics for Distribution Systems Guillermo Gallego, Paul Zipkin --- Columbia Univ., Dept. of IE/OR, 312 Mudd Bldg., 500 W 120th St., New York, NY 10027 , (gmg15@cunixa.columbia.edu)

We develop simple bounds and heuristics for the 1-warehouse¨ multi-retailer system with random demands. These results provide¨ insight into where to position stock and the drivers of system¨ performance.

MA18.3 Time Delay Measure in a Multi-Item Production/Inventory System Rachel Q. Zhang --- Univ. of MI, Dept. of IOE, Ann Arbor, MI 48109-2117, (rzhang@engin.umich.edu)

We consider a multi-item production/inventory system where a¨ customer order may require several different spare parts and¨ customer satisfaction is measured by the down time seen by customers¨ who have service contracts with the system. We derive the expression¨ for the down time and compare it with when items are treated¨ independently. A series of bounds are developed and some¨ optimization issues discussed.

MA19.1 Approaches for Identifying Customer Preferences in Quality Function Deployment Rohit Verma, Madeleine Pullman --- DePaul Univ., Dept. of Mgmt., Chicago, IL 60604 , (rverma@wppost.depaul.edu)

QFD is a structured approach for integrating customer preferences¨ into the product design and development process. We present a¨ comparative analysis of different approaches for identifying¨ customer preferences in QFD. Empirical data collected for a¨ recreational product demonstrate the strengths and weaknesses of the¨ approaches.

MA19.2 Intra-Firm Coordination Through Irrational Markets Martin A. Lariviere, Panagiotis Kouvelis --- Duke Univ., Fuqua Sch. of Bus., Box 90120, Durham, NC 27708 , (larivier@mail.duke.edu)

Many activities within a firm are carried out by self-interested¨ agents who may take actions that are not optimal for the firm as a¨ whole. We examine coordinating actions through transfer prices¨ implemented via an internal market that operates at a loss.¨ Applications to the marketing-manufacturing interface are¨ considered.

MA19.3 Inside the Fuzzy Front End: An Empirical Investigation of Project Objectives in Product Development Projects Mohan V. Tatikonda --- Univ. of NC, Kenan-Flagler Bus. Sch., CB 3490, Carroll Hall, Chapel Hill, NC 27599-3490, (mohan_tatikonda@unc.edu)

Setting appropriate project objectives is an important but little¨ understood aspect of the 'front end' of successful product¨ development projects. Findings on the objectives-setting process and¨ characteristics of appropriate objectives are presented, based on¨ in-depth case studies and a large-scale survey of product¨ development projects.

MA19.4 An Empirical Test: The Impact of Market Requirements Focus on Manufacturing Focus & Performance Cecil Bozarth, Steve Edwards --- NC State Univ., Sch. of Mgmt., Raleigh, NC 27695 , (cecil_bozarth@ncsu.edu)

This study evaluates the impact of an inconsistent market on¨ manufacturing performance and infrastructure. In an empirical test¨ of a model from Bozarth (1993), we show that the negative or¨ positive impact of market requirements focus on manufacturing¨ performance is mediated by the relationship between market¨ requirements focus and manufacturing characteristics focus.

MC18.1 Dynamic Delivery Routing & Inventory Allocation in a One-Warehouse, N-Retailer Distribution System Sangwook Park, Leroy B. Schwarz, James E. Ward --- Purdue Univ., Krannert Grad. Sch. of Mgmt., W Lafayette, IN 47907 ,

We model a dynamic routing and allocation problem in a one-warehouse¨ N-retailer distribution system operating in a periodic-review model¨ to study the cost-reduction effect of dynamic routing. The optimal¨ routing policy is identified and efficient heuristics are developed¨ and evaluated for the inventory allocation and system replenishment¨ problem.

MC18.2 Leadtimes, Batch Sizes, Demand Information & Supply Chain Performance Fangruo Chen --- Columbia Univ., Sch. of Bus., New York, NY 10027 , (fchen@research.gsb.columbia.edu)

We consider a multi-stage serial system with random demand. Each¨ stage uses a reorder-point, order-quantity policy. For fixed order¨ quantities, we show how to determine the optimal reorder points. We¨ then use this result to show how changes in lead times, batch sizes¨ and demand information (global or local) affect supply chain¨ performance.

MC18.3 Centralization of Stocks: Retailers vs. Manufacturer Ravi Anupindi, Yehuda Bassok --- Northwestern Univ., Kellogg Grad. Sch. of Bus Adm., Evanston, IL 60208 ,

In a system consisting of multiple retailers, centralization of¨ stocks decreases total costs and increases total profits is well¨ known from standard inventory literature. However, does¨ centralization benefit the manufacturer who wishes to increase¨ market share (or expected revenues) when unsatisfied customers are¨ considered lost sale? We show that centralization does not always¨ benefit the manufacturer.

MC19.1 Empirical Research on Manufacturing Performance Outcomes Roger Schroeder, E. James Flynn, Barbara Flynn, David Hollingworth --- Univ. of MN, 332 Mgmt. & Econ. Bldg., 271 19th Ave. S, Minneapolis, MN 55455 , (rschroeder@scsom.umn.edu)

We describe the result of Phase I of an NSF-funded research study of¨ manufacturing performance tradeoffs. Results include the development¨ and testing of a new theory which considers conformance quality,¨ performance quality and throughput time as drivers of other¨ performance outcomes including cost, flexibility and delivery.

MC19.2 World Class Manufacturing: An Overview of 10 Years of Empirical Research Aleda V. Roth --- Univ. of NC, Kenan-Flagler Bus. Sch., Dept. of Op. Mgmt., Chapel Hill, NC 27599-3490, (rotha.bsacdl.@mhs.unc.edu)

This presentation will report on a series of studies, captured¨ collectively under the rubric of world class manufacturing. Using¨ methods traigulation, I will pull together a 10-year history of my¨ cases, survey and action research. Consistent and cumulative results¨ that contribute to the understanding of world class manufacturing¨ will be presented.

MC19.3 Work Team Implementation Phases & Their Quality Impact: A Longitudinal Field Study Rajiv Banker, Joy Field, Roger Schroeder, Kingshuk Sinha --- Univ. of MN, Carlson Sch. of Mgmt., Minneapolis, MN 55455 , (rbanker@csom.umn.edu)

We report the findings of a study aimed at identifying different¨ phases in the process of work team implementation and examining¨ their impact on quality. The research site is a unionized electro-¨ mechanical plant operated by a division of a Fortune 500 company.

MC19.4 Firm Characteristics, Total Quality Management & Financial Performance: An Empirical Investigation Kevin B. Hendricks, Vinod R. Singhal --- Coll. of William & Mary, Sch. of Bus., PO Box 8795, Williamsburg, VA 23187-8795, (kbhend@dogwood.tyler.wm.edu kbhend@mailwm.edu)

We present results from a study that relates the financial¨ performance from effective TQM programs to various firm¨ characteristics. The characteristics examined are firm size, degree¨ of capital intensity, degree of diversification, the timing of TQM¨ implementation and the maturity of the TQM program.

MD19.1 Flexibility Planning Under Uncertainty in Manufacturing Systems Elena Katok, Terry P. Harrison --- PA State Univ., Dept. of MSIS, Smeal Coll. of Bus., University Park, PA 16802 , (exk106@silmaril.smeal.psu.edu)

The primary role of flexibility in manufacturing systems is to¨ buffer against uncertainty in the production environment. Since a¨ firm is often able to influence the level of uncertainty in its¨ environment, we propose a general framework to simultaneously¨ determine the appropriate level of manufacturing flexibility and¨ environmental uncertainty...

MD19.2 Market Share Effects of Process Improvement in Dynamic Markets Robert Burgess, Cheryl Gaimon --- GA Inst. of Tech., Sch. of Mgmt., Atlanta, GA 30332-0520,

A model is introduced evaluating investment in process improvement¨ as a means of responding to dynamic market forces. Optimal decisions¨ include the rate and timing for investment in process improvements¨ and the dynamic rate of product variety offered. The firm's optimal¨ level of market share is obtained reflecting the dynamic revenue¨ earned and costs incurred.

MD19.3 Effects of Process Variance Increase on Rejection & Taguchi Losses: An Optimum Variance Reduction Model Samar Mukhopadhyay, Debopam Chakraborty --- Univ. of WI, Sch. of Bus. Admin., Milwaukee, WI 53201 ,

Output variance of a manufacturing process increases over time as¨ the process deteriorates. We study the effects of this increase on¨ the rejection and Taguchi losses. Action can be taken to reduce the¨ variance at a cost. We present a model to obtain the optimal¨ variance reduction policy.

MD19.4 A Framework for Strategic Analysis of Process Improvement Janice Carrillo, Cheryl Gaimon --- GA Inst. of Tech., Sch. of Mgmt., Atlanta, GA 30332-0520,

While the importance of process improvements is well recognized,¨ research has been somewhat fragmented. A comprehensive framework is¨ introduced characterizing various process improvement choices.¨ Taxonomies of process change are discussed to aid managers with¨ successful selection and implementation strategies. Current¨ literature is discussed in relation to the taxonomy introduced and¨ opportunities for future research identified.

ME19.1 Manufacturing Process Monitoring Using On-Line Sensors Linguo Gong, Kwei Tang, Jen Tang --- LA State Univ., Dept. of Info. Sys. & Dec. Sci, Coll. of Bus. Admin., Baton Rouge, LA 70803 ,

We develop an economic model for using a single on-line sensor to¨ monitor a manufacturing process and compare the performance of the¨ single sensor with that of a process control chart.

ME19.2 Quality, Price & Promotion Dynamics for Profit Maximization Soumen Ghosh --- GA Inst. of Tech., Sch. of Mgmt., Atlanta, GA 30332-0205, (soumen.ghosh@mgt.gatech.edu)

We present a model that integrates a firm's marketing and¨ manufacturing-related activities concerning its product quality,¨ pricing and promotion decisions in a competitive environment. It is¨ shown that a firm's ability to maximize its profit is strongly¨ influenced by the dynamic interaction among these 3 decisions.

ME19.3 Control Charts for Autocorrelated & Skewed Data Amitava Mitra, Stanley C. Gardiner --- Auburn Univ., Coll. of Bus., Dept. of Mgmt., Auburb, AL 36849-5241, (mitraam@mail.auburn.edu)

The usual assumption used to determine control chart limits is that¨ observations are independent and identically distributed. Some¨ quality characteristics in service systems, such as customer waiting¨ time, yield observations that are correlated. We explore the¨ estimation of the process standard deviation under conditions of¨ autocorrelation. Control limits are proposed for data from skewed¨ distributions.

TA19.1 Product & Process Selection: Integration of Marketing & Manufacturing Leslie Olin Morgan --- Duke Univ., Fuqua Sch. of Bus., Chapel Hill, NC 27514 , (lso@mail.duke.edu)

As a firm updates it product lines, it must evaluate both the¨ marketplace and its own manufacturing environment in order to¨ provide an attractive yet cost competitive product mix. The focus of¨ this research is to examine the relationship between product line¨ decisions and process selection.

TA19.2 The Impact of Information Technology on Prices, Inventory Policy & Response Times Maqbool Dada, Raj Mehta --- Purdue Univ., Krannert Grad. Sch. of Mgmt., W Lafayette, IN 47907-1310, (dada@mgmt.purdue.edu)

Information technology has resulted in fundamental changes in¨ organizations and markets. We examine the supposition that¨ improvements in IT lead to increased competition, lower prices and¨ quicker response. We focus the analysis on fashion goods, i.e., good¨ for which styles change with some regularity. Demand for these goods¨ must be estimated...

TA19.3 Product Option Structure Complexity & Its Implications for Production Planning Nagesh Murthy, W. C. Benton --- GA Inst. of Tech., Sch. of Mgmt., Atlanta, GA 30332-0205, (nmurth@magnus.acs.ohio-state.edu)

The strategic decision of creating product variants has been¨ operationalized by developing measures for product option structure¨ complexity. An assemble-to-order environment is simulated to examine¨ the impact of demand management strategies in the context of product¨ variety, delivery speed, delivery reliability and manufacturing¨ stability.

TA19.4 Technology Investment for Strategic Marketing/Operations Success Alysse Rosewater, Cheryl Gaimon --- GA Inst. of Tech., Sch. of Mgmt., Atlanta, GA 30332-0520,

As product life cycles shorten, the strategic interface between¨ operations and marketing becomes crucial. A model is develped which¨ evaluates technology choice, capacity and pricing decisions for a¨ firm which produces multiple generations of high-volume products.¨ R&D spending, demand patterns, experience effects and technology¨ investment costs are shown to impact strategic decisions.

TB19.1 A Framework for Analyzing Access Costs in Services Pamela K. Armstrong --- Georgetown Univ., Sch. of Bus., G-4 Old North, Washington, DC 20057 ,

Consumers often incur a variety of nonmonetary costs when purchasing¨ a service, such as time, transportation, physical effort and risk.¨ This study characterizes and classifies these costs, examines the¨ impact of these costs on perceived quality and value and develops a¨ framework for reducing these costs while improving service¨ effectiveness.

TB19.2 Service in the USA: An Empirical Study of International Service Elizabeth L. Rose, Richard Chase, Aleda V. Roth --- Univ. of Southern CA, Sch. of Bus. Admin., Los Angeles, CA 90089-1421, (erose@kecknet.usc.edu)

The International Service Study is a large-scale, multi-national¨ study of service management in the US and Europe. Using¨ questionnaire-based interviews with corporate executives in leading¨ service firms, the study's goal is to understand the drivers of¨ service quality and performance. Theoretical and methodological¨ issues associated with the ISS are described.

TB19.3 Service in the US: An International Study Aleda V. Roth, Richard Chase, Elizabeth L. Rose, Chris Voss --- Univ. of NC, Kenan-Flagler Bus. Sch., Dept. of Op. Mgmt., Chapel Hill, NC 27599-3490, (rotha.bsacdl.@mhs.unc.edu)

Excellence in the management of service has become vital in winning¨ competitive advantage and improving business performance. This¨ presentation benchmarks the findings of the ISS, a collaborative¨ empirical research project. We present a model of service in the US¨ and highlight significant international and sector differences.

TD19.1 Remnant Inventory Systems with Process Scrap Dan Adelman, George L. Nemhauser --- GA Inst. of Tech., Sch. of ISyE, Atlanta, GA 30332-0205, (bigd@isye.gatech.edu)

Motivated by cable manufacturing, we consider IP-based, dual price¨ directed control of remnant inventory systems in the presence of¨ both allocation and process scrap. In such systems, orders are¨ satisfied with, and in turn may generate, partially consumed raw¨ maaterials, or remnants that can be used on future orders.

TD19.2 Scheduling Semiconductor Burn-In Operations in the Presence of Resource Constraints Reha Uzsoy, Cheng-Shuo Wang --- Purdue Univ., Sch. of IE, 1287 Grissom Hall, W Lafayette, IN 47907 , (uzsoy@ecn.purdue.edu)

We consider the problem of scheduling a single semiconductor burn-in¨ oven, modeled as a batch processing machine. In addition to limited¨ machine capacity, the jobs require a second resource (burn-in¨ boards) in order to be processed. We develop mathematical¨ programming formulations and heuristics to minimize makespan and¨ total completion times and presnt computational results.

TD19.3 A Survey of New Results for Flow Shop & Parallel Machine Scheduling Problems Lap Mui Ann Chan, Ana Muriel, Philip Kaminsky, David Simchi-Levi --- Philips Labs., 345 Scarborough Rd., Briarcliff Manor, NY 10510 , (lmc@philabs.philips.com)

We review recent analytical results obtained for a variety of¨ machine scheduling problems. This includes the flow shop mean¨ completion time problem as well as a class of parallel machine¨ scheduling models.

TD19.4 Coordination of Production/Distribution Networks with Unbalanced Leadtimes Julien Bramel, Shobhna Goyal, Paul Zipkin --- Columbia Univ., Grad. Sch. of Bus., 406 Uris Hall, New York, NY 10027 , (jbramel@groucho.gsb.columbia.edu)

We study multi-product, multi-location production/distribution¨ networks with deterministic, stationary demands and positive¨ leadtimes. Following the approach of Roundy, for a large class of¨ networks, we construct a policy that is within 44.3% of a lower¨ bound on all feasible policies. We extend this to 34.7% using a¨ slightly different heuristic.

TE19.1 Generic & Problem-Specific Methods for Identifying Conflicting Constraints & Decisions in Constraint-Based Scheduling Claude Le Pape --- Bouygues - Direction Scientifique, 1 ave. Eugene Freyssinet, St. Quentin en Yveli, 78061 , France (lepape@dmi.ens.fr)

In the scheduling domain, constraints and decisions often conflict,¨ either because parameters such as job due-dates were set without¨ evaluating the scheduling situation in detail, or because unexpected¨ events force to invalidate previous scheduling decisions. Both¨ generic and problem-specific methods can be used to identify¨ conflicting contraints and decisions...

TE19.2 Analyzing Infeasibilities in Master Production Schedules Mike Hegedus, Tom Ervolina --- IBM Charlotte, Microelectronics Div., 6800 IBM Dr., Charlotte, NC 28262-8563, (mhegedus@vnet.ibm.com)

We discuss the decision support tool developed at IBM to support a¨ PCB assembly facility in generating their master production¨ schedules. Schedules are constructed using optimization and/or¨ heuristic algorithms. Some of the output includes prioritized¨ materials to expedite, recommended schedule changes and projected¨ resource (materials and equipment) requirements.

TE19.3 Infeasibility Diagnostics in Finite-Capacity Production Environments Valerie Tardif, Mark L. Spearman --- Univ. of TX, Dept. of Mech. Eng., Austin, TX 78712-1063, (vtardif@mail.utexas.edu)

In finite-capacity production environments, it is often impossible¨ to meet all due dates without overtime or some due date¨ renegotiation. We propose a procedure which evaluates the¨ feasibility of proposed plans by attempting to develop a schedule¨ and provides diagnostic information on how to correct the¨ infeasibilities if no feasible schedule is found.

TE19.4 Using Detailed Analysis of Constraints to Detect & Correct Schedule Infeasibilities Howard Beck --- Univ. of Edinburgh, AI Applications Inst., 80 South Bridge, Edinburgh EH1 1HN, , UK (hab@aiai.ed.ac.uk)

While MRP-II systems have introduced a number of benefits in terms¨ of overall factory communication, significant gaps still exist at¨ the detailed production scheduling and control level. Kanban¨ approaches for scheduling and control have proven successful in¨ streamlining production activities within plants with high volume¨ and a limited range of products...

WA19.1 The Facility Layout: Who Cares? The Production System, That's Who! Russell D. Meller --- Auburn Univ., Dept. of IE, 308 Dunstan Hall, Auburn, AL 36849-5346, (rmeller@eng.auburn.edu)

The significance of a facility layout is derived from the fact that¨ it is the substrate upon which a production system functions. Yet,¨ by the time we solve the facility layout problem, little detail¨ remains with respect to the production system besides aggregate flow¨ data. Does this matter? Let's investigate.

WA19.2 On Facilities Layout Using Virtual Reality P. Banerjee --- Univ. of IL, Dept. of Mech. Eng., Chicago, IL 60607-7022, (banerjee@uic.edu)

User interfaces, data modeling and model performance on a virtual¨ reality implementaiton of facilities layout will be discussed and¨ issues related to telecollaboration using such a model presented.

WA19.3 An Integrated Approach to Solving the Facility Layout & AGV Flowpath Design Sunderesh S. Heragu, Srinivasan Rajagopalan --- RPI, Dept. of DSES, Troy, NY 12180 , (herags@rpi.edu)

Facility layout and material handling system design, the 2 main¨ issues in any manufacturing plant, have always been treated as¨ separate problems and solved. We develop an integrated approach to¨ these problems resulting in a more effective manufacturing facility.

WA19.4 An Integrated Framework for Strategic Design & Configuration of Flexible Manufacturing Facilities Maheswaran Rajasekharan, Brett Peters --- TX A&M Univ., Dept. of IE, College Station, TX 77843-3131,

Flexible manufacturing facility configuration is a key strategy for¨ improving market responsiveness when product demand and mix are¨ uncertain. An integrated framework to analyze tradeoffs between¨ benefits of flexible technology facilities and higher investment¨ costs. We introduce an integrated model for determining the facility¨ configuration, technology and flexible automation level considering¨ stochastic product demand.

WB19.1 Control of a Make-to-Stock/Make-to-Order System Scott Carr, Izak Duenyas --- Univ. of MI, IOE Dept., Ann Arbor, MI 48109 ,

We consider the stochastic control of a hybrid make-to-stock,¨ make-to-order production system. We characterize the optimal¨ production as well as order acceptance policies for this system.

WB19.2 Stochastic Models & State-Dependent Behavior John W. Boudreau, David Juran, John McClain, Kenneth Schultz, Joseph Thomas --- Cornell Univ., Malott Hall, Ithaca, NY 14853-4201,

Stochastic models of manufacturing systems make assumptions known to¨ be unrealistic to gain insights into the behavior of such systems.¨ We report on a simultaneous experimental and model-building effort¨ to investigate state-dependent worker behavior. The hypotheses are¨ that workers do have different production rates depending on factors¨ such as inventory levels, whether the current task is their 'normal'¨ task and complexity of the rules for determining what to do next...

WB19.3 Optimal Order Release in a Stochastic Production Line Mark L. Spearman, C. Nirmal Hasan --- GA Inst. of Tech., Sch. of ISyE, Atlanta, GA 30332-0205, (mark.spearman@isye.gatech.edu)

We consider scheduling job releases to a serial production line with¨ stochastic process times and pre-existing WIP. From the basic sample¨ path equation, we use stochastic approximation with infinitesimal¨ perturbation analysis to compute optimal release times. We provide¨ extensions for use within an existing MRP II system.

WB19.4 Scheduling a Machine Shared by Multiple CONWIP Lines Mark P. Van Oyen, Eungab Kim, Esma G. Senturk --- Northwestern Univ., Dept. of IE/MS, Evanston, IL 60208-3119, (vanoyen@iems.nwu.edu)

When 2 or more fabrication lines operate under CONWIP release¨ policies and share a common machine, dynamic job scheduling is¨ difficult. We provide policies and performance optimization methods¨ for scheduling of the shared machine to achieve high throughput and¨ to avoid starvation at downstream workstations.

WC19.1 After-Sales Service Logistics Support: Saturn's Market Area Strategy Morris A. Cohen, Vipul Agrawal --- Univ. of PA, Wharton Sch., OPIM Dept., 1300 SH-DH, Philadelphia, PA 19104 , (cohen@opim.wharton.upenn.edu)

The Saturn service parts logistics network has been consistently¨ rated among the best in the country in terms of parts availability,¨ customer satisfaction and operating costs. This presentation reports¨ on the system design, which was motivated by a multi-echelon¨ inventory model-based analysis, and a current study of coordination¨ within multi-site retail market areas.

WC19.2 Vendor Managed Inventory in Multi-Echelon Distribution Jarrod D. Goentzel, Paul Griffin, H. Donald Ratliff --- GA Inst. of Tech., Logistics Eng. Ctr., Sch. of ISyE, Atlanta, GA 30332-0205, (goentzel@isye.gatech.edu)

Traditionally VMI initiatives give inventory control to outside¨ vendors. In a multi-echelon network, VMI could be applied by having¨ the depot manage inventory levels at the regional distribution¨ centers. Increased flexibility provides the opportunity to make the¨ picking, transportation and restocking operations more efficient.¨ Our research investigates the potential of these opportunities.

WC19.3 Coordinated Replenishment by Capacitated Suppliers Under Periodically Varying Demands Yossi Aviv, Awi Federgruen --- Columbia Univ., Grad. Sch. of Bus., New York, NY 10027 ,

We consider a supplier responsible for the procurement for a chain¨ of retailers, facing aggregated capacity constraints in settings¨ where demand distributions and cost parameters vary according to a¨ periodic pattern. We develop close to optimal and easily¨ implementable replenishment strategies and analytically calculable¨ lower bound approximations for system-wide costs.

WC19.4 Multiproduct Contracts with Stochastic Demands & Business Volume Discounts: Single Supplier Case Ravi Anupindi, Yehuda Bassok --- Northwestern Univ., Kellogg Grad. Sch. of Bus Adm., Evanston, IL 60208 ,

We analyze contracts for multiple products when the supplier offers¨ business volume discounts. The contract takes the form of price¨ discounts for total minimum dollar commitments over the horizon with¨ flexibility to adjust the total purchases upwards by a fixed¨ percentage about this minimum commitment.

WD19.1 Teaching Intuition Using Factory Physics Mark L. Spearman, Wallace J. Hopp --- GA Inst. of Tech., Sch. of ISyE, Atlanta, GA 30332-0205, (mark.spearman@isye.gatech.edu)

Factory physics is a systematic description of the underlying¨ behavior of manufacturing systems. This description is based on a¨ number of basic postulates and several simple models. We discuss the¨ importance of using simple models to enhance a student's intuition -¨ an essential component to effective decision making.

WD19.2 Practical Issues in Implementing Quick Response Manufacturing: Insights from Fourteen Projects with Industry Raj Veeramani, Urban Wemmerlov, Frank Rath, Rajit Gadh --- Univ. of WI, 1513 University Ave., Dept. of IE, Madison, WI 53706-1572, (raj@ie.engr.wisc.edu)

Using case-based research, students and faculty worked with 14¨ companies on analyzing long lead times and recommending ways to¨ reduce them. Teaching the principles of how manufacturing systems¨ behave was instrumental in convincing companies of the validity of¨ the recommendations. Implementation issues and obstacles to¨ implementation are also discussed.