Interim equipment shutdown planning for a wafer fab during economic downturns

Because of the low equipment utilization during periods of economic recession, managers of wafer fabs are forced to plan equipment shutdowns in order to reduce variable cost and reallocate resources. Unfortunately, few studies have proposed effective solutions for equipment shutdown planning in response to economic downturns. Taking into consideration the product mix, corresponding output target, excessive capacity, production performance impact and the variable cost savings, this paper presents a new mechanism for equipment shutdown planning using a developed integer programming model. The proposed mechanism effectively provides valuable recommendations for the managements of wafer fabs regarding the type and quantity of equipment to shut down.     

Two-step simulation method for automatic material handling system of semiconductor fab

To design an effective automated material handling system (AMHS) in semiconductor fab, Samsung Electronics has used the commercial AMHS simulation software. Although this tool is very efficient at simulating material handling system, an additional simulation step is required to apply the production logic to the AMHS simulation model. Therefore, this paper introduces one more simulation step, namely the production simulation step. In the production simulation step, the production logic is implemented and the throughput of the semiconductor line can be predicted. The AMHS simulation step follows the production simulation step. In the AMHS simulation step, the capability of AMHS is predicted in advance, based on which the scope of investment can be estimated.     

Cycle time analysis of dual-arm cluster tools for wafer fabrication processes with multiple wafer revisiting times

For some wafer fabrication processes, the wafers need to visit some processing modules for a number of times, which is referred to as the revisiting process. With wafer revisiting, it is very complicated to analyze the cycle time of a dual-arm cluster tool. Due to the fact that atomic layer deposition (ALD) process is a typical revisiting process in the semiconductor industry, study is conducted on cycle time analysis of dual-arm cluster tools for the ALD process with multiple revisiting times. The system is modeled by a type of Petri net. With this model, it is revealed that the system may never reach a steady state. Based on this finding, a method is presented to analyze the cycle time and analytical expressions are derived to calculate the cycle time for different cases. Several illustrative examples are given to show the applications of the proposed approach.     

Fuzzy-neural approaches with example post-classification for estimating job cycle time in a wafer fab

Estimating the cycle time of a job in a wafer fabrication plant (wafer fab) is a critical task to the wafer fab. Many recent studies have shown that pre-classifying a job before estimating the cycle time was beneficial to the forecasting accuracy. However, most pre-classification approaches applied in this field could not absolutely classify jobs. Besides, whether the pre-classification approach combined with the subsequent forecasting approach was suitable for the data was questionable. For tackling these problems, two hybrid approaches with example post-classification, the equally-divided method and the proportional-to-error method, are proposed in this study in which a job is post-classified by a back propagation network (BPN) instead after the forecasting error is generated. In this novel way, only jobs whose cycle time forecasts are the same accurate will be clustered into the same category, and the classification algorithm becomes tailored to the forecasting approach. For evaluating the effectiveness of the proposed methodology and to make comparison with some existing approaches, production simulation (PS) is applied in this study to generate test data. According to experimental results, the forecasting accuracy (measured with root mean squared error, RMSE) of the proportional-to-error method was significantly better than those of the other approaches in most cases by achieving a 26–56% (and an average of 41%) reduction in RMSE over the comparison basis – multiple-factor linear combination (MFLC). The effect of post-classification was also statistically significant.     

Machine learning-based novelty detection for faulty wafer detection in semiconductor manufacturing

Since semiconductor manufacturing consists of hundreds of processes, a faulty wafer detection system, which allows for earlier detection of faulty wafers, is required. statistical process control (SPC) and virtual metrology (VM) have been used to detect faulty wafers. However, there are some limitations in that SPC requires linear, unimodal and single variable data and VM underestimates the deviations of predictors. In this paper, seven different machine learning-based novelty detection methods were employed to detect faulty wafers. The models were trained with Fault Detection and Classification (FDC) data to detect wafers having faulty metrology values. The real world semiconductor manufacturing data collected from a semiconductor fab were tested. Since the real world data have more than 150 input variables, we employed three different dimensionality reduction methods. The experimental results showed a high True Positive Rate (TPR). These results are promising enough to warrant further study.     

Defective wafer detection using a denoising autoencoder for semiconductor manufacturing processes

Defective wafer detection is essential to avoid loss of yield due to process abnormalities in semiconductor manufacturing. For most complex processes in semiconductor manufacturing, various sensors are installed on equipment to capture process information and equipment conditions, including pressure, gas flow, temperature, and power. Because defective wafers are rare in current practice, supervised learning methods usually perform poorly as there are not enough defective wafers for fault detection (FD). The existing methods of anomaly detection often rely on linear excursion detection, such as principal component analysis (PCA), k-nearest neighbor (kNN) classifier, or manual inspection of equipment sensor data. However, conventional methods of observing equipment sensor readings directly often cannot identify the critical features or statistics for detection of defective wafers. To bridge the gap between research-based knowledge and semiconductor practice, this paper proposes an anomaly detection method that uses a denoise autoencoder (DAE) to learn a main representation of normal wafers from equipment sensor readings and serve as the one-class classification model. Typically, the maximum reconstruction error (MaxRE) is used as a threshold to differentiate between normal and defective wafers. However, the threshold by MaxRE usually yields a high false positive rate of normal wafers due to the outliers in an imbalanced data set. To resolve this difficulty, the Hampel identifier, a robust method of outlier detection, is adopted to determine a new threshold for detecting defective wafers, called MaxRE without outlier (MaxREwoo). The proposed method is illustrated using an empirical study based on the real data of a wafer fabrication. Based on the experimental results, the proposed DAE shows great promise as a viable solution for on-line FD in semiconductor manufacturing.     

Product mix optimization for a semiconductor fab: Modeling approaches and decomposition techniques

To optimize the product mix of a semiconductor fab, the production capabilities and capacities are matched with the demand in the most profitable way. In this paper we address a linear programming model of the product mix problem considering product dependent demand limits (e.g. obligations and demand forecast) and profits while respecting the capacity bounds of the production facility. Since the capacity consumption is highly dependent on choosing from different production alternatives we are implicitly solving a static capacity planning problem for each product mix. This kind of planning approach is supported by the fluid flow concept and complete resource pooling in high traffic. We propose a general model that considers a wide range of objectives, and we introduce a heuristic that is based on the decomposition of the static capacity planning problem. A computational study reports on the quality of the decomposition approaches, and examples from practice demonstrate the versatility of the model.     

Heuristic approaches for master planning in semiconductor manufacturing

In this paper, we propose heuristic approaches for solving master planning problems that arise in semiconductor manufacturing networks. The considered problem consists of determining appropriate wafer quantities for several products, facilities, and time periods by taking demand fulfillment (i.e., confirmed orders and forecasts) and capacity constraints into account. In addition, fixed costs are used to reduce production partitioning. A mixed-integer programming (MIP) formulation is presented and the problem is shown to be NP-hard. As a consequence, two heuristic procedures are proposed: a product based decomposition scheme and a genetic algorithm. The performance of both heuristics is assessed using randomly generated test instances. It turns out that the decomposition scheme is able to produce high-quality solutions, while the genetic algorithm achieves results with reasonable quality in a short amount of time.     

Capacity planning with capability for multiple semiconductor manufacturing fabs

A capacity planning system (CPS) that considers the capacity and capability of equipment is developed for multiple semiconductor manufacturing fabs. On the basis of pull philosophy and the assumption of infinite equipment capacity, the system determines each lot's release time, start fab, and the capability of the equipment. CPS includes three main modules—the WIP-Pulling Module (WPM), the Workload Accumulation Module (WAM) and the Wafer Release Module (WRM). WPM pulls WIP from the end of the process route to meet the master production schedule (MPS). WAM then calculates the expected equipment loading in different time buckets. If WIP cannot meet the MPS requirement, then for each lot to be released, WRM evaluates the expected loading of many fabs, based on the lot's planned start time, and then determines the lot release time, the start fab and the equipment capability, to optimize the workload balance among all fabs. Simulation results indicate the effectiveness and efficiency of this system. A CPS that combines Adjusted Release Time (ART) and Path Load performs best in terms of three performance measures. This finding shows that CPS based on the combination of ART and Path Load can efficiently balance the equipment workload among the various fabs, on various days, and across various equipment at various levels of demands.     

Using an empirical queueing approach to predict future flow times

The traditional method used to describe a steady-state manufacturing system is a queueing model; whereas the common tool used to predict the future performance of a dynamic manufacturing system is a simulation model. This study proposes an empirical queueing approach to obtain the flow time performance measures of a complex dynamic manufacturing system, such as semiconductor wafer fabrication. This approach is easier to implement than a simulation model and more straightforward than a queueing model. Initially, the empirical queueing curve of each work station, which defines the relationship between the utilization rate and the expected waiting time of a small time period, is obtained from the historical database. Then, an iterative scheme is used to predict the future system behavior. Several latest researches have reported that the prediction of future flow times is important in the operation management of a complex manufacturing system. The approach proposed in this study can be easily implemented for such purposes, and the experimental results show that this approach can accurately predict the future flow times.     

Decomposition-based classified ant colony optimization algorithm for scheduling semiconductor wafer fabrication system

Due to its typical features, such as large-scale, multiple re-entrant flows, and hybrid machine types, the semiconductor wafer fabrication system (SWFS) is extremely difficult to schedule. In order to cope with this difficulty, the decomposition-based classified ant colony optimization (D-CACO) method is proposed and analyzed in this paper. The D-CACO method comprises decomposition procedure and classified ant colony optimization algorithm. In the decomposition procedure, a large and complicate scheduling problem is decomposed into several subproblems and these subproblems are scheduled in sequence. The classified ACO algorithm then groups all of the operations of the subproblems and schedules them according to machine type. To test the effect of the method, a set of simulations are conducted on a virtual fab simulation platform. The test results show that the proposed D-CACO algorithm works efficiently in scheduling SWFS.     

A long range production/resource planning system using linear programming

In January 1985, a project was initiated to optimize and update the entire Production Planning function, including long-range and short-range scheduling, Material Requirements Planning, and a Production/Resource Planning System for the Molding and Assembly area. This paper addresses the latter.

The Molding and Assembly area at Johnson Wax molds and assembles caps and their components for our products. We were looking for a long range strategic plan for the area. The natural solution was optimization in the form of linear programming.

The models were set up on Control Data's Cybernet Computer using PDS/Magen (for matrix generation) and APEX IV (for problem solving). This was accessed from an IBM-XT via a modem. (The models have since been moved to our inhouse IBM 3083 Mainframe and use Haverly Systems' OMNI and MPSX-370.) The project involved gathering information on the Molding and Assembly area, developing and testing a prototype, developing the data base and data tables for all caps concerned, expanding the prototype and developing the program to handle all caps, preparing output reports (using FOCUS Information Management System), and final implementation.     

SAP生产订单功能系统实现方案探讨

SAP生产计划与控制模块以生产订单为核心实现其生产控制功能。为适应不同生产型企业的业务特色和管理需要,可以通过SAP后台配置的方式对生产订单相关参数进行设置。本文介绍了常用的生产订单相关配置项,并对如何结合夹际业务对这些配置对象进行赋值加以探讨和研究,以求更加灵活和有效的使用SAP系统生产订单功能。     

Supply planning under uncertainties in MRP environments: A state of the art

Inventory control in a supply chain is crucial for companies desiring to satisfy their customers demands as well as controlling costs. This paper examines specifically supply planning under uncertainties in MRP environments. Models from literature that deal with random demand or lead time uncertainties are described and commented. Promising research areas emerge from this survey. It appears that lead time uncertainty has been ignored in the past, in spite of their significant importance. In particular, an interesting topic concerns assembly systems with uncertain lead times, for which the main difficulty comes from the inter-dependence of components inventories. Another promising issue, which is also presented, relates to supply planning under simultaneously demand and lead time uncertainties, which is certainly of great interest for both the academic and industrial communities.     

Planning and execution through variable resolution planning

Generating sequences of actions–plans–for robots using Automated Planning in stochastic and dynamic environments has been shown to be a difficult task with high computational complexity. These plans are composed of actions whose execution might fail due to different reasons. In many cases, if the execution of an action fails, it prevents the execution of some (or all) of the remainder actions in the plan. Therefore, in most real-world scenarios computing a complete and sound (valid) plan at each (re-)planning step is not worth the computational resources and time required to generate the plan. This is specially true given the high probability of plan execution failure. Besides, in many real-world environments, plans must be generated fast, both at the start of the execution and after every execution failure. In this paper, we present Variable Resolution Planning which uses Automated Planning to quickly compute a reasonable (not necessarily sound) plan. Our approach computes an abstract representation–removing some information from the planning task–which is used once a search depth of $k$ steps has been reached. Thus, our approach generates a plan where the first $k$ actions are applicable if the domain is stationary and deterministic, while the rest of the plan might not be necessarily applicable. The advantages of this approach are that it: is faster than regular full-fledged planning (both in the probabilistic or deterministic settings); does not spend much time on the far future actions that probably will not be executed, since in most cases it will need to replan before executing the end of the plan; and takes into account some information of the far future, as an improvement over pure reactive systems. We present experimental results on different robotics domains that simulate tasks on stochastic environments.     

A generalized Petri net modeling approach for the control of re-entrant flow semiconductor wafer fabrication

Re-entrant flow manufacturing lines, such as occur in semiconductor wafer fabrication, are characterized by a product routing that consists of multiple visits to a workstation or group of workstations during the manufacturing process. In this paper, a modeling approach is based on the use of generalized Petri nets for a re-entrant flow manufacturing line is presented. Specifically, three Petri net models representing a re-entrant flow line with three work centers and six machines are modeled. How these models may be used to represent a variety of queuing disciplines and work release policies is discussed.     

Simultaneous resource portfolio planning under demand and technology uncertainty in the semiconductor testing industry

One of the most challenging issues for the semiconductor testing industry is how to deal with capacity planning and resource allocation simultaneously under demand and technology uncertainty. In addition, capacity planners require a tradeoff among the costs of resources with different processing technologies, while simultaneously considering resources to manufacture products. The need for exploring better solutions further increases the complexity of the problem. This study focuses on the decisions pertaining to (i) the simultaneous resource portfolio/investment and allocation plan accounting for the hedging tradeoff between the expected profit and risk, (ii) the most profitable orders from pending ones in each time bucket under demand and technology uncertainty, (iii) the algorithm to efficiently solve the stochastic and mixed integer programming problem. Due to the high computational complexity of the problem, this study develops a constraint-satisfaction based genetic algorithm, in conjunction with a chromosome-repair mechanism and sampling procedure, to resolve the above issues simultaneously. The experimental results indicate that the proposed mathematical model can accurately represent the resource portfolio planning problem of the semiconductor testing industry, and the solution algorithm can solve the problem efficiently.     

Optimizing class-constrained wafer-to-order allocation in semiconductor back-end production

This paper studies the problem of allocating semiconductor wafers to customer orders with the objective of minimizing the overallocation prior to assembly. It is an important problem for back-end semiconductor manufacturing as overallocation may have severe impact on operational performance due to excess inventory and unnecessarily occupied manufacturing equipment. In practice, a wafer can contain dies from several different die classes, making the wafer-allocation problem more challenging. As a novel contribution of this work, we explicitly consider the existence of multiple die classes on a wafer in the wafer-allocation problem. An integer linear programming formulation of the class-constrained wafer allocation problem is provided. The formulation is further extended to be more flexible by allowing the dies from different classes on the same wafer to be allocated to distinct customer orders. A real-world case study from the back-end assembly and test facility of a semiconductor manufacturer is presented. Experiments with real-world data show that the proposed method significantly reduces the overallocation performance in current practice and allows planners to quantify the value of flexibility in wafer allocation.     

MES在炼钢-中厚板生产计划信息管理中的应用

MES是一个能精确调度、发送、跟踪、监控车间生产信息和过程,且能够及时测量和报告其实时性能的制造执行系统。本文重点讨论了济钢三炼钢-中厚板一体化中的生产计划信息的采集、传递和管理以及MES系统在其中的作用。     

A multi-plant production planning model considering non- repeated setup and aperiodic shipment

A new model for multi-plant production planning is developed. As the important actual features of some manufacturers, non-repeated setup and aperiodic shipment are appropriately introduced into the multi-plant production planning model and the corresponding constraints are accurately linearized. The new model is also applicable in the case of periodic shipment or backorder prohibition. Its effectiveness is examined by an instance which simulates many real characteristics. The experimental results indicate that the new model achieves the optimal profit. The sensitivity of unit setup cost and unit shipment cost is analyzed. The significance of backorder and the limitation of shipment at a time are discussed in detail.