In a semiconductor wafer fabrication facility (wafer fab) it is important to accurately predict w... more In a semiconductor wafer fabrication facility (wafer fab) it is important to accurately predict wafer outs, i.e. the remaining cycle time of the wafers in process. A wafer fab consists of multiple work areas, each containing a specific process technology, for example, photolithography, metal deposition or etching. Therefore, to accurately predict the wafer outs, an accurate prediction of the cycle time distribution at each work area is essential. This paper proposes an aggregate model to simulate each of these work areas. The aggregate model is a single server with an aggregate process time distribution and an overtaking distribution. Both distributions are WIP-dependent, but an additional layer-type dependency is introduced for the overtaking distribution. Application on a real-world wafer fabrication facility of a semiconductor manufacturer is presented for the work areas of photolithography, oxidation and dry etch. These experiments show that the aggregate model can, under certain circumstances, accurately predict the cycle time distributions in work areas by layer-type.
The use of general descriptive names, registered names, trademarks, service marks, etc. in this p... more The use of general descriptive names, registered names, trademarks, service marks, etc. in this publication does not imply, even in the absence of a specific statement, that such names are exempt from the relevant protective laws and regulations and therefore free for general use. The publisher, the authors and the editors are safe to assume that the advice and information in this book are believed to be true and accurate at the date of publication. Neither the publisher nor the authors or the editors give a warranty, express or implied, with respect to the material contained herein or for any errors or omissions that may have been made. The publisher remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
This thesis provides an overview of stochastic optimization (SP) problems and looks at how the Sa... more This thesis provides an overview of stochastic optimization (SP) problems and looks at how the Sample Average Approximation (SAA) method is used to solve them. We review several applications of this problem-solving technique that have been published in papers over the last few years. The number and variety of the examples should give an indication of the usefulness of this technique. The examples also provide opportunities to discuss important aspects of SPs and the SAA method including model assumptions, optimality gaps, the use of deterministic methods for finite sample sizes, and the accelerated Benders decomposition algorithm. We also give a brief overview of the Sample Approximation (SA) method, and compare it to the SAA method.
Proceedings of the Winter Simulation Conference, 2005., 2005
The Automated Material Handling System (AMHS) must be designed effectively so that it never becom... more The Automated Material Handling System (AMHS) must be designed effectively so that it never becomes a limiting factor for the capacity of 300mm wafer fabs. Ideally, a fully integrated fab simulation model (i.e. a model containing detailed modeling constructs for the production operations, the tools, the AMHS, and tool AMHS interactions) should be used in order to design the AMHS. However, the problem is that it takes too much time to simulate and analyze these models. Experimentation has demonstrated that certain capacity models with less detailed AMHS representations can generate accurate system predictions in comparison to the values produced by fully integrated models. Because these less detailed models run faster, we can thus assess efficiently the effects of an AMHS design configuration on equipment capacity. A case study comparing the computational efficiency and the quality of the performance predictions at different levels of detail will be presented.
A bi-criteria scheduling problem for parallel identical batch processing machines in semiconducto... more A bi-criteria scheduling problem for parallel identical batch processing machines in semiconductor wafer fabrication facilities is studied. Only jobs belonging to the same family can be batched together. The performance measures are the total weighted tardiness and the electricity cost where a time-of-use (TOU) tariff is assumed. Unequal ready times of the jobs and non-identical job sizes are considered. A mixed integer linear program (MILP) is formulated. We analyze the special case where all jobs have the same size, all due dates are zero, and the jobs are available at time zero. Properties of Pareto-optimal schedules for this special case are stated. They lead to a more tractable MILP. We design three heuristics based on grouping genetic algorithms that are embedded into a non-dominated sorting genetic algorithm II framework. Three solution representations are studied that allow for choosing start times of the batches to take into account the energy consumption. We discuss a heur...
This is a panel paper which discusses the use of simulation modeling in mitigating the effects of... more This is a panel paper which discusses the use of simulation modeling in mitigating the effects of the Covid-19 pandemic. We have gathered a group of expert modelers from around the world who have worked on healthcare simulation projects associated with the pandemic and the paper provides their answers to an initial set of questions. These serve to provide a description of the modeling work that has taken place already and to make suggestions for future directions both in modeling Covid-19 and preparing the world for future healthcare emergencies.
Proceedings of SW21 The OR Society Simulation Workshop, 2021
Tuesday, March 23 Day 3 Wednesday, March 24 Day 4 Thursday, March 25 Day 5 Friday, March 26 Netwo... more Tuesday, March 23 Day 3 Wednesday, March 24 Day 4 Thursday, March 25 Day 5 Friday, March 26 Networking break: Speed networking Lunch Break Lunch (1st 30 mins + networking (2nd half hour):(Trying to) Find calm in the chaos (Informal discussion in groups) Lunch (1st 30 mins + networking (2nd half hour): Lunchtime breakout rooms to discuss a simulation topic of interest Lunch (1st 30 mins + networking (2nd half hour)-3 Lunchtime breakout discussion rooms: 1) Your career in Simulation+ 2) WORAN + 3)ORS Orientation
Modelling has been used extensively by all national governments and the World Health Organisation... more Modelling has been used extensively by all national governments and the World Health Organisation in deciding on the best strategies to pursue in mitigating the effects of COVID-19. Principally these have been epidemiological models aimed at understanding the spread of the disease and the impacts of different interventions. But a global pandemic generates a large number of problems and questions, not just those related to disease transmission, and each requires a different model to find the best solution. In this article we identify challenges resulting from the COVID-19 pandemic and discuss how simulation modelling could help to support decision-makers in making the most informed decisions. Modellers should see the article as a call to arms and decision-makers as a guide to what support is available from the simulation community.
This paper presents two algorithms to analytically approximate work in process (WIP)-dependent in... more This paper presents two algorithms to analytically approximate work in process (WIP)-dependent inter-departure times for tandem queues composed of a series of M/M/1 systems. The first algorithm is used for homogeneous tandem queues, the second for such with bottlenecks. Both algorithms are based on the possible combinations of distributing the WIP on the queues. For each combination the time to the next departure is estimated. A weighted average of all estimated times of each WIP level is calculated to get the expected mean inter-departure time. The generated inter-departure times are used in a simple model of the tandem queue. The inter-departure times, the average WIP and average cycle time of the tandem queue and the simple model are compared in several tandem queue parameterizations. Results show only a small error between the simple model and the tandem queue, rendering this approach applicable in many applications.
IEEE Transactions on Semiconductor Manufacturing, 2017
We study a parallel batch-scheduling problem that involves the constraints of different job relea... more We study a parallel batch-scheduling problem that involves the constraints of different job release times, nonidentical job sizes, and incompatible job families, is addressed. Mixed integer programming (MIP) and constraint programming (CP) models are proposed and tested on a set of common problem instances from a paper in the literature. Then, we compare the performance of the models with that of a variable neighbourhood search (VNS) heuristic from the same paper. Computational results show that CP outperforms VNS with respect to solution quality and run time by 3.4~6.8% and 47~91%, respectively. When compared to optimal solutions, the results demonstrate CP is capable of generating a near optimal solution in a short amount of time.
Operating rooms are important assets for hospitals as they generate the largest revenue and, at t... more Operating rooms are important assets for hospitals as they generate the largest revenue and, at the same time, produce the largest cost for hospitals. The model presented in this paper helps make capacity planning decisions on the combination of open operating rooms (ORs) and estimated overtime to satisfy the allocated OR time to each specialty. The model combines both decisions on determining the amount of OR time to open and to allocate to different surgical specialties. The decisions made are based on OR costs, overutilization and underutilization costs, and contribution margins from allocating OR time. The results show the importance of having a good estimate of specialty usage of OR time to determine the amount of needed capacity and highlighted the tradeoff that the OR manager faces between opening more ORs versus extending the working time of the ORs already in use.
In November 2020 the Dagstuhl Seminar 20452 explored the needs of the semiconductor industry for ... more In November 2020 the Dagstuhl Seminar 20452 explored the needs of the semiconductor industry for making smart semiconductor manufacturing decisions and the information systems to empower flexible decisions for smart production. The seminar participants also spent time identifying the core elements for a simulation testbed which allows for assessing smart planning and control decisions in the semiconductor industry. This Executive Summary describes the process of the seminar and discusses key findings and areas for future research regarding these topics. Abstracts of presentations given during the seminar and the output of breakout sessions are collected in appendices. Seminar November 1–5, 2020 – http://www.dagstuhl.de/20452 2012 ACM Subject Classification Computing methodologies → Modeling and simulation, Information systems
We investigate the impact of the number of human-computer interactions, different interaction pat... more We investigate the impact of the number of human-computer interactions, different interaction patterns, and human inconsistencies in decision maker responses on the convergence of an interactive, evolutionary multiobjective algorithm recently developed by the authors. In our context "an interaction" means choosing the best and worst * Wallenius wishes to acknowledge the financial support from the Academy of Finland (grant number 13253583) and the KAUTE Foundation, Finland.
This paper describes the concepts underlying a new staffing tool, which evaluates staffing requir... more This paper describes the concepts underlying a new staffing tool, which evaluates staffing requirements, plan future shift sizes, perform sensitivity analysis, and locate and measure improvement opportunities in terms of worker operations. The model is based on M/M/S/K queuing theory to analyze the operator-machine-lot (OML) interference especially in semiconductor manufacturing processes. Unlike classical approaches, in which the machines are treated as customers, here wafer lots are taken as the designated customers to be serviced. A comparison of a simulation, OML, and a traditional M/M/S/K model (where the machines are modeled as customers) shows that OML model yields much closer results to the simulation than the traditional M/M/S/K model. Furthermore, from the computational point of view OML is more efficient than the simulation model which can be very time consuming especially in the complex semiconductor manufacturing environment. Thus, OML model is a superior alternative to the classical interference modeling and facilitates staffing recommendations that improve overall equipment effectiveness (GEE) and therefore productivity
Proceedings of the Winter Simulation Conference 2014, 2014
In this paper, we discuss a two-stage flow shop scheduling problem with batch processing machines... more In this paper, we discuss a two-stage flow shop scheduling problem with batch processing machines. The jobs belong to different incompatible job families. Only jobs of the same job family can be batched together. The performance measure is the total weighted tardiness of the jobs. A decomposition heuristic is proposed that is based on the idea to iteratively determine due dates for the jobs in the first stage and earliest start dates of the jobs in the second stage. The two resulting subproblems are solved using a time window decomposition (TWD) heuristic and a variable neighborhood search (VNS) scheme. Results of computational experiments based on randomly generated problem instances are presented. We show that the VNS-based scheme outperforms the TWD heuristic. In addition, we show that the decomposition scheme can be parallelized in a very natural way. As a result, the amount of computing time is modest, even for the computational expensive VNS scheme.
Proceedings of the 31st conference on Winter simulation Simulation---a bridge to the future - WSC '99, 1999
Photolithography is often the constraining equipment in semiconductor wafer fabrication plants du... more Photolithography is often the constraining equipment in semiconductor wafer fabrication plants due to the number of times the product must process through it. Modern day photolithography is performed on a cluster tool that is a combination of a stepper and track. It is obvious that the combined availability of the cluster tool is critical to throughput, but what is not so obvious is the throughput restriction from a secondary constraint known as a reticle. Every layer of a product needs a unique reticle for processing. Setup issues arising from the requirement of reticles affects productivity of photolithography and the entire wafer fabrication line if photolithography is the bottleneck. Efficient management of reticles (with regard to setup and storage on a stepper) based on current system status provides a strategic and tactical advantage. In this paper, a SLAM discrete event simulation model is to mimic the setup and storage of reticles. This enables the collection of information that can used to identify potential gains in tactical reticle management. The simulation model will be explained in detail along with output results for the tactical issues. The relationship between the simulation and the network flow model for proactive reticle management will also be discussed.
ABSTRACT A fundamental issue in designing a supply chain for a semiconductor product is to identi... more ABSTRACT A fundamental issue in designing a supply chain for a semiconductor product is to identify the strategy under which it will operate. Supply Chain strategies can be generally categorized as push, pull, and push-pull. In this research, designed simulation experiments are conducted to screen important factors that affect supply chain performance under different strategies. Statistical results indicate that the lead time customers require, as well as the perceived importance of on-time delivery, are the primary driving factors for choosing the strategy. Demand pattern and process variability need to be considered under some circumstances. The analysis leads to a conceptual decision support framework that guides the selection of push, push-pull, and pull semiconductor supply chain strategies when one is seeking lower supply chain cost and better on-time delivery service.
Manufacturing simulation and digital engineering tools and procedures have had a positive impact ... more Manufacturing simulation and digital engineering tools and procedures have had a positive impact on the manufacturing industry. However, to design a sustainable manufacturing system, a multitude of system dimensions must be jointly optimized. This paper proposes an integrated simulation tool helping to maximize production efficiency and balance environmental constraints already in the system design phase. Lean manufacturing, identification and elimination of waste and production losses, and environmental considerations are all needed during development of a sustainable manufacturing system. Engineers designing the manufacturing system need decision support, otherwise sub-optimization is more likely to occur. We present methods for calculating energy efficiency, CO2 emissions and other environmental impacts integrated into factory simulation software.
International Journal of Production Research, 2000
This paper is motivated by the problem of assigning semiconductor fabrication wafer lots to custo... more This paper is motivated by the problem of assigning semiconductor fabrication wafer lots to customer orders of various sizes. The goal of this research is to develop a method for deciding, on a given day, which orders to ® ll and the assignment of available lots to orders. This problem can be formulated as an integer program with a non-linear objective and non-linear constraints. Because of the complexity of this formulation, the problem is decomposed into two integer linear programs and solved in sequence by heuristic methods. In this paper, heuristic solutions are selected for the two subproblems and the performances of these heuristics are analysed in an experimental design using a representative data set. Based on this analysis, it is shown that the greedy heuristics selected perform signi® cantly better than current practice. Finally, future research is discussed.
In a semiconductor wafer fabrication facility (wafer fab) it is important to accurately predict w... more In a semiconductor wafer fabrication facility (wafer fab) it is important to accurately predict wafer outs, i.e. the remaining cycle time of the wafers in process. A wafer fab consists of multiple work areas, each containing a specific process technology, for example, photolithography, metal deposition or etching. Therefore, to accurately predict the wafer outs, an accurate prediction of the cycle time distribution at each work area is essential. This paper proposes an aggregate model to simulate each of these work areas. The aggregate model is a single server with an aggregate process time distribution and an overtaking distribution. Both distributions are WIP-dependent, but an additional layer-type dependency is introduced for the overtaking distribution. Application on a real-world wafer fabrication facility of a semiconductor manufacturer is presented for the work areas of photolithography, oxidation and dry etch. These experiments show that the aggregate model can, under certain circumstances, accurately predict the cycle time distributions in work areas by layer-type.
The use of general descriptive names, registered names, trademarks, service marks, etc. in this p... more The use of general descriptive names, registered names, trademarks, service marks, etc. in this publication does not imply, even in the absence of a specific statement, that such names are exempt from the relevant protective laws and regulations and therefore free for general use. The publisher, the authors and the editors are safe to assume that the advice and information in this book are believed to be true and accurate at the date of publication. Neither the publisher nor the authors or the editors give a warranty, express or implied, with respect to the material contained herein or for any errors or omissions that may have been made. The publisher remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
This thesis provides an overview of stochastic optimization (SP) problems and looks at how the Sa... more This thesis provides an overview of stochastic optimization (SP) problems and looks at how the Sample Average Approximation (SAA) method is used to solve them. We review several applications of this problem-solving technique that have been published in papers over the last few years. The number and variety of the examples should give an indication of the usefulness of this technique. The examples also provide opportunities to discuss important aspects of SPs and the SAA method including model assumptions, optimality gaps, the use of deterministic methods for finite sample sizes, and the accelerated Benders decomposition algorithm. We also give a brief overview of the Sample Approximation (SA) method, and compare it to the SAA method.
Proceedings of the Winter Simulation Conference, 2005., 2005
The Automated Material Handling System (AMHS) must be designed effectively so that it never becom... more The Automated Material Handling System (AMHS) must be designed effectively so that it never becomes a limiting factor for the capacity of 300mm wafer fabs. Ideally, a fully integrated fab simulation model (i.e. a model containing detailed modeling constructs for the production operations, the tools, the AMHS, and tool AMHS interactions) should be used in order to design the AMHS. However, the problem is that it takes too much time to simulate and analyze these models. Experimentation has demonstrated that certain capacity models with less detailed AMHS representations can generate accurate system predictions in comparison to the values produced by fully integrated models. Because these less detailed models run faster, we can thus assess efficiently the effects of an AMHS design configuration on equipment capacity. A case study comparing the computational efficiency and the quality of the performance predictions at different levels of detail will be presented.
A bi-criteria scheduling problem for parallel identical batch processing machines in semiconducto... more A bi-criteria scheduling problem for parallel identical batch processing machines in semiconductor wafer fabrication facilities is studied. Only jobs belonging to the same family can be batched together. The performance measures are the total weighted tardiness and the electricity cost where a time-of-use (TOU) tariff is assumed. Unequal ready times of the jobs and non-identical job sizes are considered. A mixed integer linear program (MILP) is formulated. We analyze the special case where all jobs have the same size, all due dates are zero, and the jobs are available at time zero. Properties of Pareto-optimal schedules for this special case are stated. They lead to a more tractable MILP. We design three heuristics based on grouping genetic algorithms that are embedded into a non-dominated sorting genetic algorithm II framework. Three solution representations are studied that allow for choosing start times of the batches to take into account the energy consumption. We discuss a heur...
This is a panel paper which discusses the use of simulation modeling in mitigating the effects of... more This is a panel paper which discusses the use of simulation modeling in mitigating the effects of the Covid-19 pandemic. We have gathered a group of expert modelers from around the world who have worked on healthcare simulation projects associated with the pandemic and the paper provides their answers to an initial set of questions. These serve to provide a description of the modeling work that has taken place already and to make suggestions for future directions both in modeling Covid-19 and preparing the world for future healthcare emergencies.
Proceedings of SW21 The OR Society Simulation Workshop, 2021
Tuesday, March 23 Day 3 Wednesday, March 24 Day 4 Thursday, March 25 Day 5 Friday, March 26 Netwo... more Tuesday, March 23 Day 3 Wednesday, March 24 Day 4 Thursday, March 25 Day 5 Friday, March 26 Networking break: Speed networking Lunch Break Lunch (1st 30 mins + networking (2nd half hour):(Trying to) Find calm in the chaos (Informal discussion in groups) Lunch (1st 30 mins + networking (2nd half hour): Lunchtime breakout rooms to discuss a simulation topic of interest Lunch (1st 30 mins + networking (2nd half hour)-3 Lunchtime breakout discussion rooms: 1) Your career in Simulation+ 2) WORAN + 3)ORS Orientation
Modelling has been used extensively by all national governments and the World Health Organisation... more Modelling has been used extensively by all national governments and the World Health Organisation in deciding on the best strategies to pursue in mitigating the effects of COVID-19. Principally these have been epidemiological models aimed at understanding the spread of the disease and the impacts of different interventions. But a global pandemic generates a large number of problems and questions, not just those related to disease transmission, and each requires a different model to find the best solution. In this article we identify challenges resulting from the COVID-19 pandemic and discuss how simulation modelling could help to support decision-makers in making the most informed decisions. Modellers should see the article as a call to arms and decision-makers as a guide to what support is available from the simulation community.
This paper presents two algorithms to analytically approximate work in process (WIP)-dependent in... more This paper presents two algorithms to analytically approximate work in process (WIP)-dependent inter-departure times for tandem queues composed of a series of M/M/1 systems. The first algorithm is used for homogeneous tandem queues, the second for such with bottlenecks. Both algorithms are based on the possible combinations of distributing the WIP on the queues. For each combination the time to the next departure is estimated. A weighted average of all estimated times of each WIP level is calculated to get the expected mean inter-departure time. The generated inter-departure times are used in a simple model of the tandem queue. The inter-departure times, the average WIP and average cycle time of the tandem queue and the simple model are compared in several tandem queue parameterizations. Results show only a small error between the simple model and the tandem queue, rendering this approach applicable in many applications.
IEEE Transactions on Semiconductor Manufacturing, 2017
We study a parallel batch-scheduling problem that involves the constraints of different job relea... more We study a parallel batch-scheduling problem that involves the constraints of different job release times, nonidentical job sizes, and incompatible job families, is addressed. Mixed integer programming (MIP) and constraint programming (CP) models are proposed and tested on a set of common problem instances from a paper in the literature. Then, we compare the performance of the models with that of a variable neighbourhood search (VNS) heuristic from the same paper. Computational results show that CP outperforms VNS with respect to solution quality and run time by 3.4~6.8% and 47~91%, respectively. When compared to optimal solutions, the results demonstrate CP is capable of generating a near optimal solution in a short amount of time.
Operating rooms are important assets for hospitals as they generate the largest revenue and, at t... more Operating rooms are important assets for hospitals as they generate the largest revenue and, at the same time, produce the largest cost for hospitals. The model presented in this paper helps make capacity planning decisions on the combination of open operating rooms (ORs) and estimated overtime to satisfy the allocated OR time to each specialty. The model combines both decisions on determining the amount of OR time to open and to allocate to different surgical specialties. The decisions made are based on OR costs, overutilization and underutilization costs, and contribution margins from allocating OR time. The results show the importance of having a good estimate of specialty usage of OR time to determine the amount of needed capacity and highlighted the tradeoff that the OR manager faces between opening more ORs versus extending the working time of the ORs already in use.
In November 2020 the Dagstuhl Seminar 20452 explored the needs of the semiconductor industry for ... more In November 2020 the Dagstuhl Seminar 20452 explored the needs of the semiconductor industry for making smart semiconductor manufacturing decisions and the information systems to empower flexible decisions for smart production. The seminar participants also spent time identifying the core elements for a simulation testbed which allows for assessing smart planning and control decisions in the semiconductor industry. This Executive Summary describes the process of the seminar and discusses key findings and areas for future research regarding these topics. Abstracts of presentations given during the seminar and the output of breakout sessions are collected in appendices. Seminar November 1–5, 2020 – http://www.dagstuhl.de/20452 2012 ACM Subject Classification Computing methodologies → Modeling and simulation, Information systems
We investigate the impact of the number of human-computer interactions, different interaction pat... more We investigate the impact of the number of human-computer interactions, different interaction patterns, and human inconsistencies in decision maker responses on the convergence of an interactive, evolutionary multiobjective algorithm recently developed by the authors. In our context "an interaction" means choosing the best and worst * Wallenius wishes to acknowledge the financial support from the Academy of Finland (grant number 13253583) and the KAUTE Foundation, Finland.
This paper describes the concepts underlying a new staffing tool, which evaluates staffing requir... more This paper describes the concepts underlying a new staffing tool, which evaluates staffing requirements, plan future shift sizes, perform sensitivity analysis, and locate and measure improvement opportunities in terms of worker operations. The model is based on M/M/S/K queuing theory to analyze the operator-machine-lot (OML) interference especially in semiconductor manufacturing processes. Unlike classical approaches, in which the machines are treated as customers, here wafer lots are taken as the designated customers to be serviced. A comparison of a simulation, OML, and a traditional M/M/S/K model (where the machines are modeled as customers) shows that OML model yields much closer results to the simulation than the traditional M/M/S/K model. Furthermore, from the computational point of view OML is more efficient than the simulation model which can be very time consuming especially in the complex semiconductor manufacturing environment. Thus, OML model is a superior alternative to the classical interference modeling and facilitates staffing recommendations that improve overall equipment effectiveness (GEE) and therefore productivity
Proceedings of the Winter Simulation Conference 2014, 2014
In this paper, we discuss a two-stage flow shop scheduling problem with batch processing machines... more In this paper, we discuss a two-stage flow shop scheduling problem with batch processing machines. The jobs belong to different incompatible job families. Only jobs of the same job family can be batched together. The performance measure is the total weighted tardiness of the jobs. A decomposition heuristic is proposed that is based on the idea to iteratively determine due dates for the jobs in the first stage and earliest start dates of the jobs in the second stage. The two resulting subproblems are solved using a time window decomposition (TWD) heuristic and a variable neighborhood search (VNS) scheme. Results of computational experiments based on randomly generated problem instances are presented. We show that the VNS-based scheme outperforms the TWD heuristic. In addition, we show that the decomposition scheme can be parallelized in a very natural way. As a result, the amount of computing time is modest, even for the computational expensive VNS scheme.
Proceedings of the 31st conference on Winter simulation Simulation---a bridge to the future - WSC '99, 1999
Photolithography is often the constraining equipment in semiconductor wafer fabrication plants du... more Photolithography is often the constraining equipment in semiconductor wafer fabrication plants due to the number of times the product must process through it. Modern day photolithography is performed on a cluster tool that is a combination of a stepper and track. It is obvious that the combined availability of the cluster tool is critical to throughput, but what is not so obvious is the throughput restriction from a secondary constraint known as a reticle. Every layer of a product needs a unique reticle for processing. Setup issues arising from the requirement of reticles affects productivity of photolithography and the entire wafer fabrication line if photolithography is the bottleneck. Efficient management of reticles (with regard to setup and storage on a stepper) based on current system status provides a strategic and tactical advantage. In this paper, a SLAM discrete event simulation model is to mimic the setup and storage of reticles. This enables the collection of information that can used to identify potential gains in tactical reticle management. The simulation model will be explained in detail along with output results for the tactical issues. The relationship between the simulation and the network flow model for proactive reticle management will also be discussed.
ABSTRACT A fundamental issue in designing a supply chain for a semiconductor product is to identi... more ABSTRACT A fundamental issue in designing a supply chain for a semiconductor product is to identify the strategy under which it will operate. Supply Chain strategies can be generally categorized as push, pull, and push-pull. In this research, designed simulation experiments are conducted to screen important factors that affect supply chain performance under different strategies. Statistical results indicate that the lead time customers require, as well as the perceived importance of on-time delivery, are the primary driving factors for choosing the strategy. Demand pattern and process variability need to be considered under some circumstances. The analysis leads to a conceptual decision support framework that guides the selection of push, push-pull, and pull semiconductor supply chain strategies when one is seeking lower supply chain cost and better on-time delivery service.
Manufacturing simulation and digital engineering tools and procedures have had a positive impact ... more Manufacturing simulation and digital engineering tools and procedures have had a positive impact on the manufacturing industry. However, to design a sustainable manufacturing system, a multitude of system dimensions must be jointly optimized. This paper proposes an integrated simulation tool helping to maximize production efficiency and balance environmental constraints already in the system design phase. Lean manufacturing, identification and elimination of waste and production losses, and environmental considerations are all needed during development of a sustainable manufacturing system. Engineers designing the manufacturing system need decision support, otherwise sub-optimization is more likely to occur. We present methods for calculating energy efficiency, CO2 emissions and other environmental impacts integrated into factory simulation software.
International Journal of Production Research, 2000
This paper is motivated by the problem of assigning semiconductor fabrication wafer lots to custo... more This paper is motivated by the problem of assigning semiconductor fabrication wafer lots to customer orders of various sizes. The goal of this research is to develop a method for deciding, on a given day, which orders to ® ll and the assignment of available lots to orders. This problem can be formulated as an integer program with a non-linear objective and non-linear constraints. Because of the complexity of this formulation, the problem is decomposed into two integer linear programs and solved in sequence by heuristic methods. In this paper, heuristic solutions are selected for the two subproblems and the performances of these heuristics are analysed in an experimental design using a representative data set. Based on this analysis, it is shown that the greedy heuristics selected perform signi® cantly better than current practice. Finally, future research is discussed.
Uploads
Papers by John Fowler