Method and system for enhancing the yield in semiconductor manufacturing

Abstract

Roughly described, a manufacturing process is enhanced by using TCAD and TCAD-derived models. A TCAD simulation model of the process is developed, which predicts, in dependence upon a plurality of process input parameters, a value for a performance parameter of a product to be manufactured using the process. Estimated, predicted or desired values for a calculated subset of the parameters (including either process input parameters or product performance parameters or both), are determined in dependence upon the process model, and further in dependence upon actual, estimated or desired values for a different subset of the parameters (again either process input parameters or product performance parameters or both). The determination is preferably made using a process compact model of the process, itself developed in dependence upon the simulation model.

Claims

1. A method for enhancing yield in an integrated circuit manufacturing process, comprising the steps of: providing a computer system having a processor and memory and programmed with a simulation model of an integrated circuit manufacturing process, the simulation model when executed by the computer system predicting, in dependence upon a plurality of process input parameters, a value for a performance parameter of an integrated circuit product to be manufactured using the manufacturing process; in a particular run of the manufacturing process, actually fabricating a particular integrated circuit product; measuring an actual value for the product performance parameter from the particular integrated circuit product and providing the actual value to the computer system; the computer system estimating, in dependence upon the simulation model and in dependence upon the actual value for the product performance parameter, past values for a calculated subset of the process input parameters that actually existed for the particular run prior to said estimating, the estimated values for the calculated subset of the process input parameters being usable for yield management, the estimated values including at least one non-measured or non-measurable parameter of the product; and adjusting the integrated circuit manufacturing process in dependence upon the estimated past values for the calculated subset of the process input parameters that actually existed for the particular run prior to said estimating.

2. A method according to claim 1, wherein the step of the computer system estimating comprises the step of the computer system estimating values for the calculated subset of the process input parameters in dependence upon actual values for a particular subset of the process input parameters, comprising the step of measuring the values in the particular subset of the process input parameters.

3. A method according to claim 1, wherein the step of the computer system estimating comprises the step of the computer system estimating values for the calculated subset of the process input parameters in dependence upon estimated values for a particular subset of the process input parameters, comprising the step of the computer system estimating the values in the particular subset of process input parameters in dependence upon the simulation model, and in dependence upon at least one member of the group consisting of actual values for a subset of the process input parameters, estimated values for a subset of the process input parameters, and an actual value for the product performance parameter.

4. A method according to claim 1, comprising the step of developing a process compact model of the manufacturing process in dependence upon the simulation model, the process compact model calculating a value for the performance parameter in dependence upon the plurality of process input parameters, and wherein the step of the computer system estimating comprises the step of the computer system estimating values for the calculated subset of the process input parameters in dependence upon the process compact model.

5. A method according to claim 4, wherein the simulation model predicts, in dependence upon the plurality of process input parameters, a value for an additional performance parameter of the product, and wherein the process compact model calculates a value for the additional performance parameter in dependence upon the plurality of process input parameters.

6. A method according to claim 1, wherein the step of the computer system estimating comprises the step of the computer system using an optimization algorithm to estimate the values for the calculated subset of the process input parameters.

7. A method according to claim 1, wherein the simulation model predicts, in dependence upon the plurality of input process parameters, a value for an additional performance parameter of the product, and wherein the step of the computer system estimating is performed in dependence upon the additional performance parameter.

8. A method according to claim 1, wherein the simulation model predicts the value for the product performance parameter using a numerical solution of partial differential equations in space or time or both.

9. A manufacturing process system for enhancing yield in an integrated circuit manufacturing process, comprising: a computer system having a processor and memory and programmed with a simulation model of an integrated circuit manufacturing process, the simulation model, when executed on the computer system predicting, in dependence upon a plurality of process input parameters, a value for a performance parameter of an integrated circuit product to be manufactured using the manufacturing process; means for determining an actual value for the product performance parameter of a particular product actually fabricated during a particular run of the integrated circuit manufacturing process by measuring the value for the product performance parameter from the particular product; the computer system being programmed to estimate, in dependence upon the simulation model and in dependence upon the actual value for the product performance parameter, past values for a calculated subset of the process input parameters that actually existed for the particular run prior to said estimating, the estimated values for the calculated subset of the process input parameters being usable for yield management, the estimated values including at least one non-measured or non-measurable parameter of the product; and means for adjusting the integrated circuit manufacturing process in dependence upon the estimated past values for the calculated subset of the process input parameters that actually existed for the particular run prior to said estimating.

10. A system according to claim 9, wherein the computer system is programmed to estimate the values for the calculated subset of the process input parameters in dependence upon measured actual values for a particular subset of the process input parameters.

11. A system according to claim 9, wherein the computer system is programmed to estimate the values for the calculated subset of the process input parameters in dependence upon estimated values for a particular subset of the process input parameters, wherein the computer system is programmed to estimate the values in the particular subset of process input parameters in dependence upon the simulation model, and in dependence upon at least one member of the group consisting of actual values for a subset of the process input parameters, estimated values for a subset of the process input parameters, and an actual value for the product performance parameter.

12. A system according to claim 9, wherein the simulation model comprises a process compact model of the manufacturing process, the process compact model calculating a value for the performance parameter in dependence upon the plurality of process input parameters, and wherein the computer system is programmed to estimate the values for the calculated subset of the process input parameters in dependence upon the process compact model.

13. A system according to claim 9, wherein the simulation model predicts, in dependence upon the plurality of process input parameters, a value for an additional performance parameter of the product, and wherein the process compact model calculates a value for the additional performance parameter in dependence upon the plurality of process input parameters.

14. A system according to claim 9, wherein computer system is programmed to estimate the values for the calculated subset of the process input parameters using an optimization algorithm.

15. A system according to claim 9, wherein the simulation model predicts, in dependence upon the plurality of input process parameters, a value for an additional performance parameter of the product, and wherein the computer system is programmed to estimate the values for the calculated subset of the process input parameters in dependence upon the additional performance parameter.

16. A system according to claim 9, wherein the simulation model predicts the value for the product performance parameter using a numerical solution of partial differential equations in space or time or both.

Description

BRIEF DESCRIPTION OF THE DRAWINGS

(1) The invention will be described with respect to particular embodiments thereof, and reference will be made to the drawings, in which:

(2) FIG. 1 is a flowchart of steps that can be used in creating a TCAD Process Compact Model in accordance with aspects of the invention.

(3) FIGS. 2-4 are flowcharts illustrating different use cases for aspects of the invention.

DETAILED DESCRIPTION

(4) The following description is presented to enable any person skilled in the art to make and use the invention, and is provided in the context of a particular application and its requirements. Various modifications to the disclosed embodiments will be readily apparent to those skilled in the art, and the general principles defined herein may be applied to other embodiments and applications without departing from the spirit and scope of the present invention. Thus, the present invention is not intended to be limited to the embodiments shown, but is to be accorded the widest scope consistent with the principles and features disclosed herein.

(5) In FIG. 1, a PCM 122 is shown together with its process parameter inputs and its product performance parameter outputs. The PCM can be used in a variety of different ways to enhance the manufacturing process. In various different use scenarios, actual or desired values are provided for some of the parameters, and actual or desired values are determined using the PCM for one or more of the remaining parameters. The provided parameters are not limited to the process input parameters, nor are the determined parameters limited to the product performance parameters. In many of the most useful scenarios, in fact, the provided parameters include one or more of the product performance parameters, and the PCM is used to determine one or more of the process input parameters which either were not measured, or are not susceptible to measurement. In these cases, optimization algorithms or data mining techniques can be used with the PCM to determine actual or desired values for the unknown parameters. The use of a PCM instead of a full TCAD process model to execute these algorithms and techniques renders them practical.

(6) Three example use cases will now be described in order to illustrate various ways in which a PCM can be used to enhance a manufacturing process.

(7) Use Case No. 1: Generation of Data for Performance Prediction or Yield Management, Both Simulated Data Equivalent to Measured Device/product Performance Data, and Data Predicting Device/product Performance Data That Is Not Measured.

(8) TCAD allows the user to complement and compare measured data from the production process with the corresponding simulated data. FIG. 2 is a flowchart illustrating how a PCM can be used with measured process data to predict a device characteristic by simulation.

(9) In step 210, actual process input parameter values are obtained and accumulated. Step 210 results in values p.sub.1, . . . , p.sub.n characterizing characteristical features of the process steps, such as film thickness, gate critical dimension, implant angle and dose, anneal temperature, etc. As used herein, an actual value for a parameter is one that exists and is either measured, or is assumed based on equipment settings. An estimated value is also one that exists, but it is estimated through the use of TCAD or a TCAD-derived model. The terms actual and estimated distinguish over values which do not yet exist, such as desired and not-yet-determined values. A not-yet-determined value is one that is not yet in existence, typically because the process steps it would characterize have not yet been performed or the partially or fully completed product it would characterize does not yet exist. If not-yet-determined values are required, they can be predicted by TCAD or a TCAD-derived model, based on provided, actual, assumed or estimated, process input parameter values. For certain purposes not-yet-determined values can also be assumed because there is little doubt about what the value will be. One example of an assumed not-yet-determined value is the value of a process parameter used in a process step that has not yet occurred, but for which the equipment that will be used typically produces only one value. Another example is a product performance characteristic predicted by TCAD or a TCAD-derived model. The terms actualand estimated also distinguish over example values, such as might be used when exploring example process input parameters in an effort to identify desired values for them which are predicted to yield desired product performance parameters, or such as might be used when creating a data set.

(10) In step 212, the PCM is evaluated for each of the actual process input parameter values, resulting in predicted values r.sub.1, . . . , r.sub.m. The predicted values r.sub.1, . . . , r.sub.m are accumulated and made available in a data-mining system such as a yield management system for subsequent comparison or correlation with actual data obtained from a production lot. In particular, in step 214, actual and/or estimated values are obtained from an actual production lot or lots, and in step 216, the difference between predicted data and actual or estimated data can be used in data mining techniques used in production for finding process deviations, process drift, or changing influences by process parameters that are not captured in the TCAD Process Model or TCAD Process Compact Model.

(11) Example: Referring to the TCAD Process Compact Model described above, gate oxide thickness and gate critical dimension are typically measured for each product wafer. For halo implant tilt, halo implant dose and the final RTP temperature, the equipment setting values in the corresponding process steps can be used. The device performance data V.sub.T, I.sub.ON, and I.sub.OFF can all be predicted by the TCAD Process Compact Model and be measured in-line. The comparison of the predicted device performance and measured device performance, as an example, can be used to analyze whether any variations in product performance are likely to be caused by related variations in gate oxide thickness and gate critical dimension, or whether variations or drifts in halo implant dose, halo implant tilt, or final RTP temperature or a variation of other process parameters are likely causes for the inconsistency between predicted and observed device performance. Furthermore, using the actual in-process data, it is possible to predict and therefore monitor non-measured product performance such as inverter switching speed. The quality of the predicted product performance data is likely to move synchronously with the quality measure obtained by comparing the predicted device performance data with the measured device performance data.

(12) A particular advantage of using data from TCAD Process Model and TCAD Process Compact Model is that, by nature, these models are not subject to noise, uncertainties and drift as occur in the manufacturing process. The combination and correlation of measured data with simulated data is therefore particularly valuable.

(13) In addition to providing simulated product performance data that replicate measured product performance data, TCAD Process Models and TCAD Process Compact Models also allow to analyze, regulate and optimize device characteristics that are conventionally not measured in manufacturing because such measurement is fundamentally not possible, difficult or uneconomical. A parameter is considered herein to be commercially incapable of measurement if either it cannot be measured, or it can be measured but is deemed not feasible or economical to do so in a commercial context. Examples may be the transient characteristics, noise characteristics, high-frequency characteristics, the performance of small circuits with particularly high sensitivity to process parameters, or similar. Notably, such additional data for predicted product performance can be used in conventional yield managements systems. On the basis of the predicted product performance the yield of manufacturing can be analyzed and improved, with respect to other performance criteria than the conventionally measured and available ones.

(14) Use Case No. 2: Inverse Analysis

(15) With the use of the present invention non-measured or non-measurable parameters, e.g., implant doses, implant tilt angles, gate shape characteristics, layer thicknesses, or similar, can be calculated with the simulation software, using other measurements and indirectly estimating the parameters of interest. This software can be a TCAD simulation, a TCAD Process Model and/or a TCAD Process Compact Model, but is preferably a TCAD Process Compact Model. The non-measurable parameters are reconstructed by inverse analysis of a TCAD Process Model or TCAD Process Compact Model from measurements of device characteristics of semiconductor devices and/or test structures. Inverse analysis uses an optimization algorithm in which the input parameters whose values are known through measurements and output parameters that are known through measurements are provided (taken as given) and the unknown input parameters to the process compact model, which represent the non-measured parameters, are optimized by changing them in a mathematical optimization algorithm until the residual error becomes minimal. FIG. 3 is a flowchart illustrating major steps in the Inverse Analysis use case.

(16) In step 310, actual values are obtained and accumulated for some of the process input parameters and product performance parameters. These values can be obtained, for example, by measuring on product wafers, or by assumption from the process equipment settings. This results in values p.sub.1, p.sub.2, . . . , p.sub.k characterizing characteristical features of the process steps such as film thickness, gate critical dimension, implant angle and dose, anneal temperature, etc. This also results in values rm.sub.1, rm.sub.2, . . . , rm.sub.m, characterizing product performance.

(17) In step 312, using a TCAD Process Compact Model with functions r.sub.i(p.sub.1, p.sub.2, . . . , p.sub.k, p.sub.k+1, p.sub.k+2, . . . , N) where i=1, . . . , m, the most likely actual values are determined and accumulated for unknown process input parameters p.sub.k+1, p.sub.k+2, . . . , p.sub.n. This is performed by selecting weights w.sub.i (i=1, . . . , m) for the individual device performance characteristics, denoting their relative importance or quality of measurement, and minimizing the difference length of the residual error vector w.sub.i*(r.sub.irm.sub.i) (i=1, . . . , m). Typical multidimensional optimization algorithms can be used, such as the Newton-method, nonlinear simplex method, simulated annealing, genetic algorithms or similar. A beneficial part of the algorithm is the compensation for any systematic differences between measured values r.sub.i.sub._.sub.m and simulated values r.sub.i.sub._.sub.s for functions r.sub.i(p.sub.1, . . . , p.sub.n). The differences typically arise through imperfect calibration. While the imperfect calibration has an effect on the absolute values for r.sub.i, it has only a small effect on the sensitivity of the function r.sub.i(p.sub.1, . . . , p.sub.n). It is therefore sufficient to 1) estimate the simulated nominal value r.sub.i.sub._.sub.n of the function r.sub.i(p.sub.1, . . . , p.sub.n) from evaluating the function for nominal values p.sub.1.sub._.sub.n, . . . , p.sub.n.sub._.sub.n and 2) and determining the averages r.sub.i.sub._.sub.m.sub._.sub.ave or medians r.sub.i.sub._.sub.m.sub._.sub.median of a number of measured values from the measured values r.sub.i.sub._.sub.m. The reconstruction is then made for the compensated value r.sub.i.sub._.sub.m(r.sub.i.sub._.sub.m.sub._.sub.medianr.sub.i.sub._.sub.n) or r.sub.i.sub._.sub.m(r.sub.i.sub._.sub.m.sub._.sub.aver.sub.i.sub._.sub.n), respectively, if subtraction is used for compensation. Alternatively, the compensated value r.sub.i.sub._.sub.m/r.sub.i.sub._.sub.m.sub._.sub.median*r.sub.i.sub._.sub.n or r.sub.i.sub._.sub.m/r.sub.i.sub._.sub.m.sub._.sub.ave*r.sub.i.sub._.sub.n can be used, respectively, if multiplication is used for compensation.

(18) Example: Using on the TCAD Process Compact Model from the examples above, it is possible to attribute any of the differences between predicted and observed product performance data V.sub.T, I.sub.ON, I.sub.OFF to changes in halo implant dose, halo implant tilt, either each by itself, assuming others with assumed values, in combinations of two, or all together. These process parameters are usually not measured, but assumed. Yet, they may be subject to drift or to variation across a wafer. Using the reconstructed (estimated) values for these parameters, it is in turn possible through evaluation of the TCAD Process Model or TCAD Process Compact Model with measured gate oxide thickness, gate critical dimension, reconstructed halo tilt and dose, and RTP temperature from the equipment setting, to predict the product performance characteristic of inverter switching speed.

(19) If the input parameters of the TCAD Process Model or TCAD Process Compact Model comprise all relevant sources of variation in the process parameters and the measurements (of process parameters or responses), the resulting estimate can be considered as being correct. If other sources of variation or noise in the manufacturing process and the measurements exist, the estimation of the unknown parameter is less correct, since the unknown additional variability is attributed to the process parameters under study. Since the resulting variability is larger than the real variability, the estimate serves as a useful upper limit.

(20) Notably, such additional reconstructed data for parameters can be used in conventional yield managements systems. On the basis of the reconstructed or estimated non-measured or non-measurable parameters the yield of manufacturing can be improved.

(21) Use Case No. 3: Feed-Forward Process Enhancement

(22) According to the third use case, semiconductor manufacturing processes are optimized during one or more manufacturing step by means of TCAD. FIG. 4 is a flowchart illustrating a method for enhancing the yield in semiconductor manufacturing, wherein semiconductor products are manufactured in a plurality of sequential manufacturing steps.

(23) In step 410, one or more steps of a semiconductor fabrication process are performed, yielding a partially completed or intermediate product. As used herein, the term process step can include other process sub-steps, which are themselves considered herein to be steps in their own right.

(24) In step 412, actual values are obtained for process input parameters already established in the intermediate product. The values can be obtained by measurement or by assumption based on process equipment settings. Alternatively, estimated values can be obtained for some or all of the process input parameters, for example using inverse analysis as described above. This results in values p.sub.1, p.sub.2, . . . , p.sub.k characterizing characteristic features of the process steps such as film thickness, gate critical dimension, implant angle and dose, anneal temperature, etc.

(25) In step 414, targeted (desired) product performance is characterized with desired values rt.sub.1, rt.sub.2, . . . , rt.sub.m.

(26) In step 416, using a TCAD Process Compact Model with functions r.sub.i(p.sub.1, p.sub.2, . . . , p.sub.k, p.sub.k+1, p.sub.k+2, . . . , p.sub.n) where i=1, . . . , m, and values p.sub.1, . . . , p.sub.k for the known process input parameters, desired characteristic values are determined for yet to be performed process steps p.sub.k+1, p.sub.k+2, . . . , p.sub.n. This is performed by selecting weights w.sub.i (i=1, . . . , m) for the individual device performance characteristics, denoting their relative importance or quality of measurement, and minimizing the difference length of the residual error vector w.sub.i*(r.sub.irt.sub.i) (i=1, . . . , m). Typical multidimensional optimization algorithms can be used, such as the Newton-method, nonlinear simplex method, simulated annealing, genetic algorithms or similar. A beneficial part of the algorithm is the compensation for any systematic differences between measured values r.sub.i.sub._.sub.m and simulated values r.sub.i.sub._.sub.s for functions r.sub.i(p.sub.1, . . . , p.sub.n). The differences typically arise through imperfect calibration. While the imperfect calibration has an effect on the absolute values for r.sub.i, it has only a small effect on the sensitivity of the function r.sub.i(p.sub.1, . . . , p.sub.n). It is therefore sufficient to 1) estimate the simulated nominal value r.sub.i of the function r.sub.i(p.sub.1, . . . , p.sub.n) from evaluating the function for nominal values p.sub.i.sub._.sub.n, . . . , p.sub.n.sub._.sub.n, and 2) and determining the averages r.sub.i.sub._.sub.m.sub._.sub.ave or medians r.sub.i.sub._.sub.m.sub._.sub.median of a number of measured values from the measured values r.sub.i.sub._.sub.m. The determination of optimum parameters is then made for the compensated target value r.sub.i.sub._.sub.m(r.sub.i.sub._.sub.m.sub._.sub.medianr.sub.i.sub._.sub.n) or r.sub.i.sub._.sub.m(r.sub.i.sub._.sub.m.sub._.sub.aver.sub.i.sub._.sub.n), respectively, if subtraction is used for compensation. Alternatively, the compensated target value r.sub.i.sub._.sub.m/r.sub.i.sub._.sub.m.sub._.sub.median*r.sub.i.sub._.sub.n or r.sub.i.sub._.sub.m/r.sub.i.sub._.sub.m.sub._.sub.ave*r.sub.i.sub._.sub.n can be used, respectively, if multiplication is used for compensation.

(27) In step 418, subsequent manufacturing steps are performed with p.sub.k+1, . . . , p.sub.n that result from the optimization algorithm, thereby forming a final product or another intermediate product.

(28) Depending on the product it is possible that the steps 412-418 are to be repeated several times until the semiconductor product is finished. The intermediate product is a partially finished product.

(29) The method can be performed in semiconductor manufacturing on a lot level, on a wafer level or on a die level, sometimes also called reticle level. On every level, enhancement of the manufacturing yield is possible, i.e., on a lot level if wafers are manufactured by a lot, on a wafer level if wafers are manufactured in single-wafer processing, and, in both cases, on a die level during die-level processing steps. Examples for die-level processing steps are the exposure of a individual die on the wafer in the stepper, or the annealing of an individual die on the wafer in a laser annealing step.

(30) Example: Using the TCAD Process Compact Model described above, it is possible to partially process a wafer, measure gate oxide thickness and gate critical dimension in one or several locations of the product wafer, determine V.sub.T, I.sub.ON, I.sub.OFF as well as the inverter switching speed that are most desirable, assume that the final RTP temperature will not be changed, and, using the described analysis algorithm, determine the most desirable setting for dose and tilt angle of the halo implant step. If measurements are made in more than one location on the wafer, we can determine the best setting from the average of the measurements or from an average that is weighted by the area corresponding to a particular measurement.

(31) As can be seen, a TCAD PCM can be used in a variety of scenarios for the improvement of manufacturing processes. Many of the scenarios can be summarized using the PCM 122 in FIG. 1, as shown together with its process parameter inputs and its product performance parameter outputs. In some scenarios there are values provided for one or more of the process input parameters, whereas in other scenarios there are values provided for one or more of the product performance parameters. In still other scenarios there are values provided for both process input parameters and product performance parameters. The provided values may be in various ones of the scenarios either actual values (measured or assumed based on other knowledge) of process input parameters or product performance parameters, estimated values of process input parameters or product performance parameters based on indirect analysis methods, not-yet-determined values of process input parameters which have not yet occurred, or desired values of product performance parameters. In addition, in some scenarios the PCM 122 is used in the determination of one or more desired, estimated or predicted values for process input parameters, whereas in other scenarios the PCM 122 is used in the determination of desired, estimated or predicted values for product performance parameters.

(32) Using this conceptual framework, the following table provides a summary comparison of certain specific example use cases.

(33) TABLE-US-00003 BASED ON USE PCM TO DETERMINE USE CASE EXAMPLE THESE VALUES THESE VALUES Performance Prediction Actual, estimated and/or not- Predicted values for product for partially completed yet-determined but assumed performance parameters products values for process input parameters Performance Prediction Actual or estimated values for Estimated values for product for completed products process input parameters performance parameters Performance Prediction Not-yet-determined but Predicted values for product for hypothetical products assumed values for process performance parameters input parameters Inverse Analysis Actual or estimated values for Estimated values for Others of Some Process Input the process input parameters Parameters and actual values for product performance parameters Feed Forward for partially Actual, estimated or not-yet- Desired values for others of the completed products determined but assumed process input parameters values for some process input parameters, and desired values for product performance parameters

(34) The method according to the present invention is preferably related to the manufacturing of semiconductor devices, semiconductor test structures, and circuits, especially yield-critical circuits. It is applicable to all semiconductor devices that can be simulated by TCAD. In particular, the manufacturing of transistors can be improved significantly.

(35) Through the use of TCAD models the electrical performance characteristics of the semiconductor devices, a semiconductor test structure or circuit, especially a yield-critical circuit, is optimized. For instance the threshold voltage, drive current or leakage current of a transistor can be regulated and optimized for a particular application. The TCAD model helps to find optimized manufacturing parameters on the basis of parameters measured during the manufacturing process.

(36) TCAD Process Models and TCAD Process Compact Models also allow to optimize for device characteristics that are conventionally not measured in manufacturing because such measurement is fundamentally not possible, difficult or too uneconomical. Examples are the transient characteristics, noise characteristics, high-frequency characteristics, the performance of small circuits with particularly high sensitivity to process parameters, or similar. Notably, such additional data for predicted product performance can be used in conventional yield managements systems. On the basis of the predicted product performance the yield of manufacturing can be improved.

(37) The improvements over conventional process control methods include the use of TCAD for process modeling, the TCAD Process Model being more comprehensive and accurate than other, simplified models, and in the use of TCAD Process Compact Models, which are sufficiently fast, robust accurate and embeddable (meaning that they can easily be integrated into other manufacturing software environments for process control and yield management) to allow deployment and use in a manufacturing environment.

(38) A computer program product comprising software code portions for performing a method according to the method of the present invention when run on a computer system having a processor and memory, can be provided. With such a computer program product it is possible to perform the method according to the present invention on different manufacturing locations.

(39) As used herein, a given event or information item is responsive to a predecessor event or information item if the predecessor event or information item influenced the given event or information item. If there is an intervening processing step or time period, the given event or information item can still be responsive to the predecessor event or information item. If the intervening processing step combines more than one event or information item, the result of the step is considered responsive to each of the event or information item. If the given event or information item is the same as the predecessor event or information item, this is merely a degenerate case in which the given event or information item is still considered to be responsive to the predecessor event or information item. Dependency of a given event or information item upon another event or information item is defined similarly.

(40) The foregoing description of preferred embodiments of the present invention has been provided for the purposes of illustration and description. It is not intended to be exhaustive or to limit the invention to the precise forms disclosed. Obviously, many modifications and variations will be apparent to practitioners skilled in this art. In particular, and without limitation, any and all variations described, suggested or incorporated by reference in the Background section of this patent application are specifically incorporated by reference into the description herein of embodiments of the invention. The embodiments described herein were chosen and described in order to best explain the principles of the invention and its practical application, thereby enabling others skilled in the art to understand the invention for various embodiments and with various modifications as are suited to the particular use contemplated. It is intended that the scope of the invention be defined by the following claims and their equivalents.