See discussions, stats, and author profiles for this publication at: https://www.researchgate.

net/publication/228549165

Implementing Advanced Planning and Scheduling (APS) System for

Semiconductor Manufacturing: A Case at Korean Semiconductor
Manufacturing Company

Article  in  IE interfaces · January 2007

CITATIONS READS

0 1,695

2 authors, including:

Seung-Kil Lim
Sungkyul University
35 PUBLICATIONS   276 CITATIONS   

SEE PROFILE

Some of the authors of this publication are also working on these related projects:

Identical parallel machine View project

All content following this page was uploaded by Seung-Kil Lim on 10 July 2014.

The user has requested enhancement of the downloaded file.

Implementing Advanced Planning and Scheduling (APS) System for

Semiconductor Manufacturing: A Case at Korean Semiconductor

Manufacturing Company

Seung-Kil Lim

Division of e-businessIT, Sungkyul University, KOREA

[email protected]

Yong-Ho Shin*

ISEL (School of Logistics), University of Le Havre, FRACE

[email protected]

*Corresponding Author: Yong-Ho Shin

ISEL (School of Logistics)

Quai Frissard

B.P. 1137

76063 Le Havre CEDEX-FRANCE

E-mail : [email protected]

Telephone: +33 (0)2 32 74 49 18

 Abstract

Semiconductor manufacturing is one of the most complex and capital-intensive processes composed of

several hundreds of operations. In today's competitive business environments, it is more important than ever

before to manage manufacturing process effectively to achieve better performances in terms of customer

satisfaction and productivity than those of competitors. To obtain such fruitful outcome, an effective

management of IC chip production process including production planning and scheduling decisions should be

made intensively. These decisions include demand planning, capacity planning, production planning and

production scheduling decisions. Therefore, many semiconductor manufacturing companies implemented

advanced planning and scheduling (APS) system as a management tool for the complex semiconductor

manufacturing process Although there exist a few articles and survey results that introduce industrial cases of

implementing production planning and scheduling systems for semiconductor manufacturing, research articles

are very rare that explicitly reveals key issues to be resolved for successful implementation of production

planning and scheduling systems. Sharing industrial experiences in terms of key issues and lessons learned

are very important and useful to both academies and practitioners because challengeable research topics and

critical factors for successful system implementation can be identified with the experiences. From these

reasons, in this paper, we briefly describe several key issues resolved during APS system implementation

project at a Korean semiconductor manufacturing company, which may be useful in building similar APS

systems at other semiconductor manufacturing companies. We firstly explain roles of production planning

and scheduling in semiconductor manufacturing and principal factors that make the production planning and

scheduling more difficult. Then we describe the APS system implementation project at Korean semiconductor

manufacturing company in terms of key issues. The analysis of several alternative solutions for those issues is

also revealed. Finally, we show the results of productivity improvement due to the APS system

implementation with actual performance statistics.

Keywords: productivity improvement, semiconductor manufacturing, advanced planning and scheduling,

system implementation
0. Introduction

Leading-edge technology and high capital investment are the most distinct features of semiconductor business.

Recently, effective management of manufacturing process is more important than ever before in this business

area since lower cost and higher customer satisfaction, which become essential factors for competitive

advantage in today's marketplace, can be achieved with better management capability, while design and

production technologies for high-density integrated circuits (IC) have been main factors for competitiveness

in semiconductor industry. Although effective management should be made in all areas of semiconductor

manufacturing process including design of integrated circuits, production of IC chips and delivery of

produced chips to customers, IC chip production process is one of the most challengeable area for reducing

cost and improving customer satisfaction since manufacturing cost and customer service quality are

significantly influenced by the chip production process due to its quite long production lead-time (at least 30

days and even longer than 90 days for some product types) and very complex operational nature (several

hundreds of operations performed by thousands of technology-intensive equipments). Here, production lead-

time, called as cycle time, is the duration from the release time at the first production operation to the time it

is packed ready for shipment.

Basically, IC chip production process consists of four sequential stages; wafer fabrication, electrical die

sorting, assembly and final test. First two production stages are generally referred as front-end stages, while

the others are referred as back-end stages. At the wafer fabrication stage, electronic circuits are built up layer

by layer on raw wafers, thin discs made of silicon or gallium arsenide, by performing several hundreds of

operations. A few hundreds of identical chips (called individual chip on the wafer as die) are made on each

wafer through the wafer fabrication process. Individual die on wafer is electrically tested at the electrical die

sorting stage from which good and bad dies are sorted by marking ink dots on bad dies physically or making

electronic map (that contains information about electrical test results for dies) for each wafer using some test

equipments. At the back-end stages, wafers are cut into individual dies (that is IC chips) and each chip is
placed with lead-frame in plastic or ceramic packages at assembly stage, while burn-in and functional tests for

packaged chips are performed at final test stage.

For effective management of IC chip production process, some of planning and scheduling decisions should

be made carefully. These decisions include demand planning, capacity planning, production planning and

production scheduling decisions. Demand planning is associated with formulating sales plan (based on

forecasts and sales targets given by top management) that provides types and quantities of IC chips to be sold

monthly, weekly and/or daily for given planning horizon. On the other hand, production planning is

associated with determining supply plan (that is production plan to meet sales plan) that provides types and

quantities of IC chips to be produced monthly, weekly and/or daily for given planning horizon while capacity

planning is associated with deciding how to maintain or acquire production resources such as workers and

machine equipments needed to satisfy the production plan. Finally, detail schedules, which are used to

determine sequences for jobs and operations at machine equipments in production lines, are generated through

production scheduling. In other words, when machine equipment of production lines is available after

completing a job assigned to the machine, next job or operation to be performed on the machine is determined

by the scheduling decision.

There are a lot of theoretical studies on production planning and scheduling for semiconductor manufacturing.

Especially, production planning and scheduling for wafer fabrication have been main research topic of many

researchers since wafer fabrication is the most important and complex process involving hundreds of

operations with very long production lead-time. Previous studies on production planning and scheduling for

wafer fabrication can be classified into two types according to main objectives of planning and scheduling.

The first type focuses on maximizing throughput with minimum cost (Rupp and Ristic, 2000; Vargas-Villami

and Rivera, 2000; Elif Akcah and Uzsoy, 2001; Hsieh et al., 2001; Lee et al., 2002; Lee and Kim, 2002; Kim

et al., 2002), while the second type has the objective of meeting due-dates of customers' orders or production

orders (Dabbas et al., 2001; HoriGuchi et al., 2001; Kim et al., 1998, 2001, 2003; Hwang and Chang, 2003;

Vargas-Villamil et al., 2003) without sacrificing system utilization. Besides the research results on production
planning and scheduling at the wafer fabrication stage, some researchers study production planning and

scheduling at other stages of semiconductor production process such as electrical die sorting (Lee et al., 2000;

Ellis et al., 2004), assembly (Potoradi et al., 2002) and final test process (Freed and Leachman, 1999;

Sivakumar, 1999). Hsieh et al. (2001), Hwang and Chang (2003), Ellis et al. (2004), and Kim et al. (2003),

well summarize previous research results on production planning and scheduling for semiconductor

manufacturing.

Table 1. Planning and scheduling systems used in major semiconductor companies

Company
1 2 3 4 5 6 7 8
Oracle,
SAP, SAP, SAP, SAP, Homegrown, SAP,
ERP Not Baan
5 Years 1 year 5 Years 13 Years 5 years
implemented

APO And i2
i2 Mimi and
Demand APO 1 Year i2 i2 i2 Homegrown
Not Homegrown Adexa
Planning i2 Not 2 years 1.5 Years 0.5 Years 2 Years
SCM Applications used

implemented Mimi 4 years,

implemented

Supply i2 i2
APO, Homegrown i2 Homegrown Adexa
Chain Not Not Homegrown
Pilot 1 Year 3 Years 10 Years 1 Year
Planning implemented implemented
i2
Factory i2 Homegrown i2 i2 Homegrown Adexa Adexa
Not
Planning Pilot 2 years 2 Years 0.5 Years 10 Years Pilot 1 Year
implemented
Homegro
Detailed Homegrown MES MES Homegrown Adexa
wn Homegrown
Scheduling 5 Years 5 Years 3 Years 2 Years 1 Year
3 Years
Oracle
Material SAP Homegrown SAP SAP Homegrown SAP
Not Homegrown
Planning 3 Years 5 Years 3 Years 3 Years 1 Year 3 Years
implemented

Source: Semiconductor Supply Chain Benchmark, 2001, PriceWaterhouseCoopers

Despite lots of theoretical studies on production planning and scheduling for semiconductor manufacturing,

only a few research results are practically used for effective management of real semiconductor production

processes (Murty and Bienvenu, 1995; Malmstrom, 1997; Pickett and Zuniga, 1997). On the other hand,

many semiconductor manufacturing companies implement commercial software packages or develop their

own (homegrown) planning and scheduling systems for effective management of their production processes.

Table 1 shows survey results conducted by PriceWaterhouseCoopers, one of the largest global business
consulting firm, about planning and scheduling systems used in 8 major semiconductor manufacturing

companies. Note that major semiconductor manufacturing companies like Intel, Samsung and AMD

participated in the survey although real names of the semiconductor companies are not given in Table 1 due to

confidential issues. As shown in Table 1, commercial software packages from professional solution vendors

such as i2 Technologies, Adexa and SAP are used for demand planning, supply chain planning, factory

planning and detail scheduling. Detail roles of supply chain planning and factory planning for semiconductor

production will be described later. Professional solution vendors mentioned above called their software

packages for planning and scheduling as advanced planning and scheduling (APS) suites since optimization

techniques and information technologies are applied into their software packages such as mixed integer linear

programming, TOC (Theory of Constraints), Web-based multi-tier client-server architecture with large-scale

integrated database.

Although there exist a few articles and survey results that introduce industrial cases of production planning

and scheduling systems implementation for semiconductor manufacturing, articles are very rare that explicitly

reveal key issues to be resolved for successful implementation of production planning and scheduling systems.

Sharing industrial experiences in terms of key issues and lessons learned are very important and useful to both

academies and practitioners because challengeable research topics and critical factors for successful system

implementation can be identified with the experiences. From these reasons, in this paper, we briefly describe

several key issues resolved during APS system implementation project at a Korean semiconductor

manufacturing company, which may be useful in building similar APS systems at other semiconductor

manufacturing companies.

In the following sections, we firstly explain roles of production planning and scheduling in semiconductor

manufacturing and principal factors that make the production planning and scheduling more difficult. Then

we describe the APS system implementation project at Korean semiconductor manufacturing company in

terms of key issues.

1. Production Planning and Scheduling for Semiconductor Manufacturing

As mentioned earlier, IC chip production process consists of four sequential production stages, wafer

fabrication, electrical die sorting, assembly and final test. Typically, there exist several production facilities

(in other words, production lines) at each production stage, that is, several wafer fabrication lines (called as

wafer FAB shortly), electrical die sorting lines, assembly lines and final test lines exist at each production

stages. Wafers fabricated at wafer fabrication lines are transferred to electrical die sorting lines, and then

sorted wafers are moved to assembly lines. Finally, packaged IC chips at assembly lines are sent to final test

lines. Production planning in such production structure is associated with determining

(daily/weekly/monthly) production targets at individual production facilities of each production stages to meet

sales plan determined by demand planning for given product (that is IC chip) types over planning horizon.

Sometimes the production planning for semiconductor manufacturing is called as supply chain planning at the

manufacturing-level since production facilities at production stages are connected like typical supply chains.

There exist a large number of machine equipments at each production facilities. For example, a wafer

fabrication facility has several thousands of machine equipments that are used to perform operations needed

to build integrated circuits on wafers. Similarly, other facilities for electrical die sorting, assembly and final

test have large numbers of machine equipments for their operations. Typically, several hundreds of operations

are performed on wafers at wafer fabrication facilities and tens of operations are performed on wafers or IC

chips at other production facilities for electrical die sorting, assembly and final test. Since wafers and IC chips

visit the machine equipments as a form of lot (composed of multiple wafers or several thousands of IC chips),

job sequences (that is, which lot should be processed next when machine equipment is available) at machine

equipments need to be determined. This is the main role of production scheduling for semiconductor

manufacturing. Here, a lot is composed of a number of wafers, typically ranged from 2 to 25 wafers, and the

wafers of a lot are cut into individual IC chips at assembly stage. After the wafers are cut into the IC chips,

several thousands of IC chips are included into a lot again that visits machine equipments of assembly and

final test stages for completing remained operations.

Now, we examine some principal factors that make production planning and scheduling for semiconductor

manufacturing more difficult. These factors can be classified into four categories: uncertainty, operational

complexity, and multiple objectives and data management.

1.1 Uncertainty

Availability and reliability of machine equipments are well known factors that cause difficulties in effective

management of semiconductor production processes. Note that machine availability and reliability are related

with machine breakdowns and inconsistent operational performances of machine equipments, respectively. It

is said that machine equipment is not available when the machine breaks down or machine is under the

preventive maintenance while machine equipment is not reliable when outputs from the machine are not

consistent in terms of quantity and/or quality level. For example, machine equipments for photolithography

operation in wafer fabrication facility are not reliable since outputs (in terms of quality and quantity) from the

machines are not consistent according to product types and times when the operations are performed,

although same operator utilizes the same machine equipment to perform similar operations. Typical results

from limited machine availability and unreliable operational performances in semiconductor production are

lower utilizations of machine equipments and yield losses of wafers and/or IC chips. Here, yield is usually

defined as the ratio of the number of usable items (wafers or IC chips) after the completion of production

processes to the number of potentially usable items at the beginning of production (Cunningham et al., 1995).

Considering the machine utilization and yield loss in determining optimal production targets at each

production facilities and generating detail schedules of lots of wafers or IC chips at machine equipments are

not easy since they are significantly influenced by uncertain factors such as machine breakdowns and

inconsistent operational performances as mentioned earlier.

1.2 Operational Complexity

Operational complexity is another factor that makes production planning and scheduling for semiconductor

manufacturing more difficult. As mentioned in many research papers (Hsieh et al., 2001; Hwang and Chang,

2003; Vargas-Villamil et al., 2003), production processes of semiconductor manufacturing have several

distinct features such as reentrant workflows, batching, time-window constraint, sequence-dependent setup
time and auxiliary resources. Reentrant workflows, which occur at the wafer fabrication stage, can be

explained as multiple visits of lots of wafers to the same machine equipments for several similar operations.

Remember that wafers visit machine equipments as a form of lot for their operations as mentioned earlier.

Also, note that several layers are built up on wafers layer by layer at wafer fabrication stage and a layer is

completed when a series of operations performed by a set of machine equipments are finished. Since the

series of operations required for a layer is similar to those for other layers, a lot of wafers can visit the same

machine equipment several times for similar operations required by each integrated circuit layer. We call this

as reentrant workflow since wafers revisit same machine equipments several times. Batching is another

operational characteristic of semiconductor production that can be explained as simultaneous processing of

multiple lots of wafers or a large number of IC chips at machine equipment. Batching occurs at some

operations of wafer fabrication and final test stages such as diffusion and burn-in test operations. Also, time-

window constraints impose additional complexity on production scheduling since those require that an

operation be started within specified time-window after previous operation for the operation is finished.

Typical example operations that should satisfy the time-window constraints are pre-processing operations

performed to prepare main operations such as cleaning operations. On the other hand, sequence-dependent

setup times mean that different setup times are needed to perform an operation according to types of

operations that were completed before the operation. Sequence-dependent setup time requirements appear in

some etching and test operations. Finally, auxiliary resources should be available in order to start some

operations for semiconductor production. Typical examples for auxiliary resources are reticles (called as

masks) that are used in doing photolithography operations at wafer fabrication stage. Ultraviolet rays are

exposed on wafers through the reticles, from which circuits are printed on the wafers since reticles transmits

ultraviolet rays according to the circuit image embedded on the reticles. Operational characteristics of

semiconductor production mentioned so far make production planning and scheduling more difficult since

these characteristics force additional considerations (such as batching, sequence-dependent setup time and

auxiliary resources availability) to be included to make production plan and detail schedule.

1.3 Multiple Objectives

Although semiconductor manufacturing companies' objectives for production planning and scheduling may

be different among each other according to their market environments and marketing and/or production

strategies, there exist common objectives that are considered as more important ones by many semiconductor

manufacturing companies. For production planning, maximizing both profit and customer satisfaction are the

most important objectives when determining production targets that meet sales plan and/or customer orders.

On the other hand, maximizing throughput and utilization, as well as meeting production targets given by

production planning are important objectives for production scheduling. Typically, machine equipments in

semiconductor production facilities are very expensive. As a result, many semiconductor manufacturing

companies want to fully utilize their production resources. On the other hand, they also want to reduce (or

optimize) cycle time and its variation in order to improve responsiveness to external and/or internal

requirements such as urgent customer orders and order change requests (that is, maximizing customer

satisfaction). For reducing cycle time, level of work-in-process should be reduced since waiting times of

wafer lots in semiconductor production facilities can be cut down with lower work-in-process inventory.

However, utilization of machine equipments may be harmful as the WIP level decreases for reducing cycle

time since the possibility of being machine idle increases due to less work-in-process inventory. Optimizing

several conflicting objectives simultaneously, which are required in production planning and scheduling for

semiconductor manufacturing, are very difficult tasks.

1.4 Data Management

Huge amount of data to be managed is another factor that causes high complexity of production planning and

scheduling for semiconductor manufacturing. In general, it is necessary to manage data related to demands,

products, operations and resources in order to perform production planning and scheduling. Demand-related

data includes product type (product code), due-date, production quantity and customer name or code, while

product-related data is product code (product number or product name), bill-of-materials and sales price. On

the other hand, operation-related data specifies which operations are needed to make a product type and

includes operation name and sequence, machine equipments capable of performing the operation and needed

time for the operation. Resource-related data includes names and types of machine equipments and auxiliary

resources such as reticles, and set up time if required when performing operations at the machine equipments.
In addition to data mentioned above, there exist huge amount of data to be managed. Since several thousands

of operations are needed to produce IC chips and tens of thousands machine equipments are utilized for the

operations, very large-sized data about operations and resources should be managed. Furthermore, it is

difficult and time-consuming tasks to gather these data needed to generate production plan and schedule since

many companies often manage these data at different and/or separate information systems.

Factors discussed above are strongly related with key issues to be resolved for successful implementation of

production planning and scheduling system for semiconductor manufacturing. In the next section, we will

give details for the key issues mainly caused by the factors when implementing advanced planning and

scheduling system based on experiences at a Korean semiconductor manufacturing company.

2. APS System Implementation at a Korean Semiconductor Company

At the beginning of 21st century, a global semiconductor manufacturing company in Korea has devoted to the

construction of an optimized supply chain to strengthen its competitiveness in marketplace. The company

examined all functional areas included in its supply chain from procurement to logistics as well as production

in order to identify possible opportunities for improvement or innovation. Several projects were initiated at

that time and some continued until early of 2004. The semiconductor company not only produces typical

memory chips such as DRAM and SRAM but also some SOCs (system-on-chips) such as ASIC (Application

Specific Integrated Circuit) used by most of electronic products like digital camera and cellular phones. The

company operates multiple facilities for each production stages of semiconductor production in Korea and

some facilities in other countries. For optimizing supply chain, the company focused on the development of

APS system since the company thought APS system is very powerful tool in managing their complex

semiconductor production process effectively. So, APS system implementation project started at early of 2000

in which most of planning and scheduling activities are included such as demand planning, capacity planning,

and production planning and scheduling. Software packages from i2 Technologies were selected and

implemented for the core planning and scheduling activities except for production scheduling. The company
developed their own production scheduling system after evaluation of several commercial software packages'

functionality for semiconductor production scheduling.

In this section, we give brief descriptions of issues occurred during the APS system implementation project

but we focus on issues related to production planning and scheduling at wafer fabrication stage, which are the

most complex and difficult tasks in the APS system implementation for semiconductor manufacturing. Also,

we limit our discussion to production planning and scheduling for SOCs manufacturing since the complexity

of the APS system implementation for those products is greater than that for typical memory chips. First, we

briefly introduce main tasks conducted during the implementation project and then the issues identified are

described. The following are six main tasks performed during the implementation project.

Process Analysis

Main objectives of this task are to understand current production and business processes, called as AS-IS

processes, and to analyze these processes in order to find opportunities for improvement with newly

implemented APS system. Large number of interviews and workshops are done for this task. Note that target

activities of this analysis are business processes related to production planning and scheduling that are

performed by production planners and managers, and operators' works at shop floor for semiconductor

production.

Issues and Requirements Finding

With the results of Process Analysis task, issues to be resolved for successful implementation and

functionalities requested by users of the APS system are identified. Note that issues and requirements are

different in that issues are problems to be solved for successful implementation of the APS system, while

requirements are functionalities requested by system users who will use the implemented system for their

works.

Solution Mapping
This task is performed to generate solutions for the issues identified by Issues and Requirements Finding task.

For mapping final solutions to the identified issues, several alternative solutions for each of issues are

compared in terms of time and cost needed as well as expected performances on goals of APS system

implementation. Also, from this task, we determine which and what functionalities should be provided for

users of the APS system.

Process Design

Based on functionalities of newly implemented APS system, we design improved business and production

processes, called as TO-BE processes. Improved processes are devised through several workshops and/or

brainstorming of which participants are all of persons related with the APS system such as system users

including production managers (of course, top managers of production division) and operators of production

lines, and system developers including external consultants and system managers.

System Design and Implementation

Included in System Design task are modeling database structure and configuring system architecture while

System Implementation task focuses on installation and customization of selected software packages and

developing some computer programs needed for data gathering from several different information systems

and providing graphical user interfaces. Here, configuring system architecture is to determine overall system

environments such as client-server and/or Web-based three-tier architecture, OS (Operating System) and

hardware platforms, and communication structure like intranet and/or extranet.

Test and Go-Live

This is the final task from which we test functionalities of the implemented APS system and assess quality of

production plans and schedules generated by the system. If the test is passed, users now can use the

implemented system (that is, the system is in go-live status), additional works should be done until the test is

passed otherwise.
3. Critical Issues in the APS Implementation Project

Now, we describe most critical issues identified in the APS system implementation project.

Issue 1. Hierarchy of Production Planning and Scheduling

First of all, we should determine the best hierarchy of production planning and scheduling since production

planning and scheduling for semiconductor manufacturing are very difficult tasks due to several factors as

mentioned earlier. Like we often divide a very difficult problem into several tractable subproblems in order to

obtain feasible solutions in a reasonable amount of time, the hierarchy of production planning and scheduling

is needed to divide whole production planning and scheduling decision-making problem into several

subproblems and define their relationships.

Issue 2. Functionality of Software Package

The second issue is associated with limited functionality of software package. Since most of software

packages from professional solution vendors are general-purpose ones, some of functionality needed for

semiconductor manufacturing are not supported or do not sufficiently meet requirements of system users. As

a result, it is necessary to determine which functionality of software package should be used at what level of

satisfaction compared to the requirements of system users.

Issue 3. Data Preparation

Since huge amount of data is needed for APS system implementation, companies that want to build APS

system should prepare the data. However, preparing this data is time-consuming tasks since much of data is

dispersed among several information systems and even some are not managed. As a result, it is necessary to

make data management plan carefully considering future system developments as well as the APS system

implementation.

Issue 4. System Modeling

As mentioned earlier, there exist several factors that make production planning and scheduling for

semiconductor manufacturing more difficult. Also, facilities at each production stage have large number of

machine equipments and there are large number of product types and operation types. Since most of software

packages need a model in order to represent actual system configuration and the performance in terms of

quality of plan and schedule, and time required to obtain the plan and schedule is affected by the model, it is

necessary to determine how to model the actual system and how to consider the factors such uncertainty and

operational characteristics.

Issue 5. Test and Quality Check

One of the most difficult problems in implementing planning and scheduling system is the assessment of

quality of plans and schedules that are generated by the implemented system. For the assessment, it is

necessary to devise suitable performance measures and systematic test/assessment procedures.

Issue 6. System Architecture

It is necessary to determine system architecture that determines the configuration of hardware platforms, O/S

platforms and their roles in the entire system configuration. Also, we need to determine appropriate

specifications of hardware systems such as memory size, CPU speed and so on.

Issue 7. Training and System Management

When software package from a professional solution vendor is used, it is required that some people should be

trained for doing some system management tasks such as modifying and/or upgrading system modeling and

system configuration. As a result, it is necessary to determine suitable persons for undertaking the system

management tasks and to make a plan for the training.

4. Concluding Remarks

In this study, we introduce a real APS system implementation project conducted at one of the largest

semiconductor manufacturing company in Korea. Based on actual experiences, we briefly describe several
key issues resolved during the APS system implementation project that may be useful in building similar APS

systems at other semiconductor manufacturing companies. Since articles and research works are very rare that

explicitly reveal key issues to be resolved for successful implementation of production planning and

scheduling systems, our study that focuses on sharing industrial experiences in terms of key issues are very

useful to both academies and practitioners in that both people may find challengeable research topics and

critical factors for successful system implementation with the results of this study.

References

Cunningham, S. P., Spanos, C. J., and Voros, K. (1995) Semiconductor yield improvement: Results, and best

practices. IEEE Transactions on Semiconductor Manufacturing 8: 103-109

Dabbas, R. M., Chen, H. N., Fowler, J. W., and Shunk D. (2001) A combined dispatching criteria approach to

scheduling semiconductor manufacturing systems. Computers and Industrial Engineering 39: 307-324

Elif Akcah, K. N. and Uzsoy, R. (2001) Cycle-time improvements for photolithography process in

semiconductor manufacturing. IEEE Transactions on Semiconductor Manufacturing 14 (1): 48-56

Ellis, K. P., Lu, Y., and Bish, E. K. (2004) Scheduling of wafer test processes in semiconductor

manufacturing. International Journal of Production Research 42 (2): 215-242

Freed, T. and Leachman, R. C. (1999) Scheduling semiconductor device test operations on multihead testers.

IEEE Transactions on Semiconductor Manufacturing 12 (4): 523-530

Horiguchi, K., Raghavan, R., Uzsoy, R., and Venkateswaran, S. (2001) Finite-capacity production planning

algorithms for a semiconductor wafer fabrication facility. International Journal of Production Research

39 (5): 825-842

Hsieh, B. W., Chen, C. H., and Chang, S. C. (2001) Scheduling semiconductor wafer fabrication by using

ordinal optimization-based simulation. IEEE Transactions on Robotics and Automation 17 (5): 599-608

Hwang, T. K. and Chang S. H. (2003) Design of a Lagrangian relaxation-based hierarchical production

scheduling environment for semiconductor wafer fabrication. IEEE Transactions on Robotics and

Automation 19 (4): 566-578

Kim, S., Yea, S. H., and Kim, B. K. (2002) Shift scheduling for steppers in the semiconductor wafer

fabrication process. IIE Transactions 34: 167-177

Kim, Y. D., Kim, J. U., Lim, S. K., and Jun, H. B. (1998) Due-date based scheduling and control policies in a

multiproduct semiconductor wafer fabrication facility. IEEE Transactions on Semiconductor

Manufacturing 11 (1): 155-164

Kim, Y. D., Kim, J. G., Choi, B., and Kim, H. U. (2001) Production scheduling in a semiconductor wafer

fabrication facility producing multiple product types with distinct due dates. IEEE Transactions on

Robotics and Automation 17 (5): 589-598

Kim, Y. D., Shim, S. O., Choi, B., and Hwang, H. (2003) Simplification methods for accelerating simulation-

based real-time scheduling in a semiconductor wafer fabrication facility. IEEE Transactions on

Semiconductor Manufacturing 16 (2): 290-298

Lee, Y. H. and Lee, B. K., and Jeong, B. (2000) Multi-objective production scheduling of probe process in

semiconductor manufacturing. Production Planning and Control 11 (7): 660-669

Lee, Y. H. and Kim, T. (2002) Manufacturing cycle time reduction using balance control in the

semiconductor fabrication line. Production Planning and Control 13 (6): 529-540

Lee, Y. H., Park, J., and Kim, S. (2002) Experimental study on input and bottleneck scheduling for a

semiconductor fabrication line. IIE Transactions 34: 179-190

Malmstrom, C. (1997) An integrated approach to planning and scheduling at Phillips Semiconductors.

Proceedings of IEEE/SEMI Advanced Semiconductor Manufacturing Conference D27-D29

Murty, S. and Bienvenu, J. W. (1995) Global Planning at Harris Semiconductor. Proceedings of International

Symposium on Semiconductor Manufacturing 18-23

Pickett, B. and Zuniga, M. (1997) Modeling, scheduling, and dispatching in the dynamic environment of

semiconductor manufacturing at FASL, Japan. Proceedings of IEEE/SEMI Advanced Semiconductor

Manufacturing Conference 448-450

Potoradi, J., Boon, O. S., and Mason S. J. (2002) Using simulation-based scheduling to maximize demand

fulfillment in a semiconductor assembly facility. Proceedings of the 2002 Winter Simulation Conference

1857-1861
 Rupp, T. M. and Ristic, M. (2000) Fine planning for supply chains in semiconductor manufacture. Journal of

Material Processing Technology 107: 390-397

Sivakuma, A. I. (1999) Optimization of cycle time and utilization in semiconductor test manufacturing using

simulation based, on-line, near-real-time scheduling system. Proceedings of the 1999 Winter Simulation

Conference 727-735

Vargas-Villamil, F. D. and Rivera D. E. (2000) Multilayer optimization and scheduling using model

predictive control: application to reentrant semiconductor manufacturing lines. Computers and Chemical

Engineering 24: 2009-2021

Vargas-Villamil, F. D., Rivera D. E., and Kempf, K. G. (2003) A hierarchical approach to production control

of reentrant semiconductor manufacturing lines. IEEE Transactions on Control systems Technology 11

(4): 578-587

View publication stats