CN103236382B - Method for determining process parallel rationalization of manufacturing process of plasma display screen - Google Patents

Abstract

The invention discloses a method for determining the process parallel rationalization of the manufacturing process of a plasma display screen. A comparison relation model of parameter set values of parallel equipment in the manufacturing process of the plasma display screen is established by utilizing the historical data of the manufacturing process; the influence degree of the parameter set values on the quality of the plasma display screen is reflected by utilizing yield difference between different set values under the same parameter between the parallel equipment; and finally, the method for determining the process parallel rationalization of the manufacturing process of the plasma display screen is formed. The method has the positive effects that the accuracy of determining the process parallel rationalization of the manufacturing process mattering the quality of the plasma display screen is promoted; by accumulation of the regulated new historical data, suggestions on the process parallel rationalization of the manufacturing process of the plasma display screen are found by a method which spirally reuses an analysis model; and meanwhile, the method cannot be related to the interruption of production of a production line and avoids unnecessary line stop loss.

Description

Plasma panel manufacturing processes walk abreast rationalize defining method

Technical field

The present invention relates to the Analysis on Data Mining method with plasma panel manufacture process characteristic, is core, by software development and solidification in this way.Formed a kind of plasma panel manufacturing processes walk abreast rationalize defining method.

Background technology

In plasma panel manufacture process, due to the reason of production capacity pressure (or technique convergence time is compact up and down), often same operation, arranges multiple stage identical device parallel production, has possessed the hardware condition comparing parallel process Rational Parameters.But how to carry out rationalization to determine, be in the past by the EXPERIMENTAL DESIGN of construction period and repeatedly checking determine.Along with production programming extension and 7*24 hour uninterrupted volume production, need traditional test design method by a large amount of manpower of waste and kinetic energy consuming cost if continue, also need production line to carry out stopping the debugging design that line (stopping volume production) carries out test parameters value simultaneously.This method reduces the production efficiency of volume production on the one hand, simultaneously by the manpower of at substantial and kinetic energy consuming cost.Therefore need a kind of new method dynamically carry out manufacturing processes walk abreast rationalize determination and optimization.

Along with the large quantities of volume production of plasma panel, face again data magnanimity situation.In plasma panel manufacture process volume production process, separate unit plasma panel process of producing product device parameter is included MES system storage in and is reached 1.2 ten thousand, every day more than data volume 10G, the parameter related to is more than 9000, in quantity, dimension and data generation speed have the large data characteristics of magnanimity, more in the urgent need to invent a kind of new method dynamically carry out manufacturing processes walk abreast rationalize determination and optimization.

The present invention should launch under this background.

Summary of the invention

The invention provides determination and the optimization method of LPT device parameter in a kind of plasma panel manufacture process, by utilizing manufacture process historical data, set up the relativity model of the pre-set parameter of LPT device in plasma panel manufacture process, utilize the yield difference between LPT device under identical parameters between different settings, reflect the influence degree of pre-set parameter article on plasma display screen quality, and find the regulating strategy of the apparatus parameter setting value promoting PDP display production yield, a kind of plasma panel manufacturing processes of final formation walk abreast rationalize defining method.

The technical solution adopted for the present invention to solve the technical problems is: a kind of plasma panel manufacturing processes walk abreast rationalize defining method, comprise the steps:

Step one, according to operation, the LPT device in operation to be divided into groups;

Step 2, the yield statistics of carrying out PDP display under approximate production status is respectively being ensured to the different LPT devices in every group;

Step 3, significance test is carried out to the different LPT devices in every group, calculate actual performance difference;

Step 4, the result of calculation interval of actual performance difference to be judged:

(1) if result of calculation interval comprises 0, then under explanation current set value, the actual performance difference of individual device is significantly not different, and the parameter of individual device remains unchanged;

(2) if result of calculation interval does not comprise 0, then under explanation current set value, the actual performance significant difference of individual device is different, needs the corresponding pre-set parameter pre-set parameter of equipment low for yield being adjusted to the high equipment of yield.

Compared with prior art, good effect of the present invention is: save time and resources costs spent by the mode by repetition test design and verification, improve determine concerning the manufacturing processes of plasma panel quality walk abreast rationalize accuracy; And by the new historical data accumulation after adjustment, carry out this analytical model method of screw type recycling to find out plasma panel and make process and to walk abreast conductive suggestion, manufacturing technology personnel are according to recommended value, carry out the setting of parallel process device parameter value, not only reach the object of lasting spiral lifting plasma panel yields, but also indirectly reduce the percent defective of manufacture process.Meanwhile, in the use procedure of method model, the breaks in production of production line can not be related to, avoid unnecessary line loss of stopping and lose.

Embodiment

Plasma panel manufacturing processes walk abreast rationalize a defining method, comprise the steps:

Step one, divided into groups by the LPT device in operation according to operation, the equipment in each group is the production equipment that function is identical under same operation, because also have identical parameter group.In the production process of PDP display, an equipment group comprises 2 to 3 LPT devices usually;

Step 2, for the above equipment group produced, the different LPT devices in each group are being ensured that the yield that (i.e. the identical period, kinetic parameter is similar) under approximate production status carries out PDP display is respectively added up;

Step 3, significance test is carried out to the different LPT devices in each group:

The yield of the PDP display on distinct device compares by the pre-set parameter for LPT device, and distinguish based on by the screen quantity of each LPT device and yield the LPT device possessing remarkable performance difference, concrete significance test method is as follows:

(1) variance is calculated as follows:

The approximate account form of list portion equipment variance is: wherein 1-e _ifor yield;

(2) cumulative variance is calculated

(3) actual performance difference is calculated as follows:

On the Confidence possessing (1-α), the actual performance difference of two equipment is:

wherein, the observation yield difference between d indication equipment, α is the yield expected, Z _α/2represent under current yield condition, be distributed as the probability of benchmark with current yield.

Step 4, the result of calculation interval of actual performance difference to be judged:

(1) if result of calculation interval comprises 0, then under explanation current set value, the actual performance difference of individual device is significantly not different, and the parameter of individual device remains unchanged;

(2) if result of calculation interval does not comprise 0, then under explanation current set value, the actual performance significant difference of individual device is different, needs the corresponding pre-set parameter pre-set parameter of equipment low for yield being adjusted to the high equipment of yield.

The inventive method illustrates as follows:

With two LPT device M ₁and M ₂for example, it is as follows to carry out significance test method:

Equipment M ₁yield (g in a measurement period ₁) be 85%, pass through M ₁screen quantity n ₁it is 30;

Equipment M ₂yield (g in a measurement period ₂) be 75%, pass through M ₂screen quantity n ₂it is 5000;

When shield quantity reach more than certain value (usually 30) time, yield be scattered in a normal distribution, that is:

E ₁=1-g ₁~ N (μ ₁, σ ₁), wherein, μ ₁for the average on M1, σ ₁for the variance on M1;

E ₂=1-g ₂~ N (μ ₂, σ ₂), wherein, μ ₂for the average on M2, σ ₂for the variance on M2;

The approximate account form of variance is:

For testing the statistical significance of two LPT device performance differences, i.e. real yield difference, needs the variance calculating accumulation

On the Confidence possessing (1-α), the actual performance difference of two kinds of equipment is:

wherein, the observation yield difference between d indication equipment, α is the yield expected, Z _α/2represent under current yield condition, be distributed as the probability of benchmark with current yield.

Wherein the value of Z can be learnt by following Confidence table:

1-α	0.99	0.98	0.95	0.90
					Z	2.58	2.33	1.96	1.65

This table is ready-made form under normal distribution by hypothesis yield.

Therefore will possess the Confidence of 95%, namely 1-α=0.95 is obtained by the upper table of inquiry: Z _α/2=1.96, the value of each computational element has in this example:

d＝|e ₁-e ₂|＝0.1

${\widehat{\mathrm{\σ}}}_t=\frac{0.15(1-0.15)}{30}+\frac{0.25(1-0.25)}{5000}=0.0043$

Under 95% Confidence, by Z _α/2=1.96 substitute into actual performance difference computing formula can obtain:

$d_t=0.100\±1.96\×\sqrt{0.0043}=0.100\±0.128,$

Find that above formula result of calculation interval comprises 0, although it is not two equipment are statistically variant at yield under can determining current set value, significantly different under actual performance difference, therefore do not have M ₂set point is to M ₁change.

Three equipment as above can compare wherein any two equipment, finally obtain that there is the equipment having optimum configurations most.

Claims (1)

1. plasma panel manufacturing processes walk abreast rationalize a defining method, it is characterized in that: comprise the steps:

Step one, according to operation, the LPT device in operation to be divided into groups;

Step 2, the yield statistics of carrying out PDP display under approximate production status is respectively being ensured to the different LPT devices in every group;

Step 3, carry out significance test to the different LPT devices in every group, calculate actual performance difference, concrete grammar is as follows:

(1) variance of single device is calculated;

(2) cumulative variance is calculated

(3) actual performance difference is calculated as follows:

in formula, the observation yield difference between d indication equipment, α is the yield expected, Z _α/2represent under current yield condition, be distributed as the probability of benchmark with current yield;

Step 4, the result of calculation interval of actual performance difference to be judged:

(1) if result of calculation interval comprises 0, then under explanation current set value, the actual performance difference of individual device is significantly not different, and the parameter of individual device remains unchanged;

(2) if result of calculation interval does not comprise 0, then under explanation current set value, the actual performance significant difference of individual device is different, needs the corresponding pre-set parameter pre-set parameter of equipment low for yield being adjusted to the high equipment of yield.