Data processing: measuring – calibrating – or testing – Measurement system in a specific environment – Quality evaluation
Reexamination Certificate
2000-08-01
2002-11-12
Hilten, John S. (Department: 2863)
Data processing: measuring, calibrating, or testing
Measurement system in a specific environment
Quality evaluation
C700S099000
Reexamination Certificate
active
06480794
ABSTRACT:
BACKGROUND OF THE INVENTION
1. Field of the Invention
This invention relates to methods of allocating products to be tested to machines on a manufacturing line and more particularly to allocating products to the machines in view of capacity to handle products to be manufactured before testing. The invention teaches a method to minimize tool test time for testing factories.
2. Description of Related Art
In testing a factory, the unstable output data from the front-end of a manufacturing line results in frequent tester setup changes and severely late deliveries. But testing is often requested to catch-up to compensate for the delays which occurred on the front-end of the manufacturing line.
U.S. Pat. No. 5,880,960 of Lin et al. shows a method to balance WIP in a factory.
U.S. Pat. Nos. 5,841,677 of Yang et al., 5,826,238 of Chen et al., 5,818,716 of Chin et al., 5,825,650 of Wang and 5,548,326 of Tai et al. show methods for optimizing factory cycle time.
Disadvantages of current practices:
1) In the past, there has been no optimal methodology for allocating capacity before testing arranged a production schedule.
2) It has been difficult to prepare and to forecast the capability and capacity of the testers.
3) There has been neither an appropriate algorithm nor an appropriate procedure for measuring/setting accurate test times.
4) There has been no integrated system for utilizing all testers for testing of multiple FAB sites, referred to herein as the N-site, S-Site and T-site.
SUMMARY OF THE INVENTION
Objects of the system of this invention are as follows:
(1) A powerful/friendly web-based tool: Visible to calculate forecasted loading of both sites for testing at any time and feedback to Sales/CPD (Central Planning Department)/FAB(Fabrication Plant).
(2) An easy way to query test times quickly for any product.
(3) Integrate basic data & procedure between N-PC, N-Testing, S-PC, IE&OE, S-Testing, and TOM
FAB =
Fabrication Plant
N-PC =
N-site Production Control Department
N-Testing =
N-site Testing Department
S-PC =
S-Site Production Control Department
IE&OE =
Industrial Engineering and Operation Efficiency
Department
S-Testing =
S-site Testing Department
TOM =
Total Order Management Department
Tij =
Testing time for grid position ij is the time
for machine j to test product i
Xij =
Demand quantity dispatched for grid position ij
Ui =
variable which is used to decide “entering
basic variable”
Vj =
same as Ui
CP product =
Circuit Probing product
PB product =
same as CP product
FT product =
Final Test product
N
COL
=
number of columns of the transporting matrix.
N
LIN
=
number of lines of the transporting matrix.
V
NB
=
Non-Basic Variable
NV
B
=
the Number of a Basic Variable which has been
dispatched, some demand quantity
Non-basic =
A variable that is not basic is called a non
variable
basic variable
This invention comprises a testing method for an automatic capacity information integrated system, and more particularly, to a method for setting a standard test time and minimizing the total test time with respect to production scheduling in factories.
In accordance with this invention, a method and a system are provided for allocating products to be tested to machines on a manufacturing line by providing a standard test time, minimizing the total test time with respect to production scheduling in factories, and forming a supply demand matrix table for the products and the machines to which the products are to be allocated.
Preferably the method and system include steps and means for performing those steps, as follows:
provide the table with the testing time for the machine and the product corresponding to the location on the matrix table;
determine the grid location with the minimum testing time Tij;
provide the largest possible allocation of time from the corresponding machine at the corresponding position on the matrix table;
the system loops back to provide the largest possible allocation of remaining time from the corresponding machine,
the system continues to loop until no demand is left;
determine whether an optimum testing process is required, preferably by testing whether only one machine can test the product and no quantity is allocated to a machine and if YES branching to calculating of the utilization of each machine and if NO, then decide whether as follows
N
COL
+N
LIN
−1
=NV
B
(2)
where:
N
COL
=number of columns of the transporting matrix,
N
LIN
=number of lines of the transporting matrix,
NV
B
=number of basic variable,
if YES then proceeding to perform optimum testing, and
if NO the branching to calculate the utilization of each machine.
Preferably the method and system include as follows:
deciding whether the demand quantity is allocated to a dummy machine, if NO, then END the program and if YES, then proceed to decide whether
N
COL
+N
LIN
1
=NV
B
, and
if NO, then END the program if YES, then proceed,
eliminate the product row that can be tested at one machine,
eliminate the machine column that can test only one product,
eliminate the column or row which is irrelevant to a dummy machine with an allocated demand quantity.
Preferably the method and system include further as follows:
choosing the column or row that has the larger number of basic variables and assigning its U or V value to be zero (0), and
get all values of Ui and Vj by using every relation of the basic formula Tij=Ui+Vj,
calculate Tij−Ui+Vj of non-variable and largest absolute value in all negative numbers, and if there is a tie randomly choosing one.
decide whether (Tij−Ui−Vj) of all non-variables is larger than 0;
If YES, then the current solution is optimum and the procedure is stopped and if NO, then the system proceeds to calculating the utilization of each machine, and
calculate the utilization of each machine as follows:
average divide the product quantity on the “dummy” machine to every feasible machine wherein the dummy machine is designed to deal with a quantity which cannot be allocated to other machines that have been overloaded so in this step the system is employed to average dividing the overloaded quantity to every feasible machine for the purpose of calculating the final utilization of every machine,
calculate Machine Utilization (MU) by the formula as follow:
MU
=
[
Σ
(
Quan
alloc
*
Treq
)
efficiency
%
+
engineering time
]
(working day
*
machine quantity
*
available%)
(3)
DEFINITIONS
efficiency =
average
actual
test
time
theoretical
test
time
engineering time =
Time required to run a pilot run
Quan_alloc =
Quantity allocated
Treq =
required test time
Available capacity =
WD * MQ * AW
WD =
Working Day
MQ =
Machine Quantity
AW =
Available Working
day =
duration of working day in minutes
machine quantity =
number of machines
available % =
(1-down % - PM % - OFF % - monitor %)
decide whether the same machine can process both the CP product and the FT product and if the result of the decision is NO, then the system proceeds to block to END the program; and if the result of the decision in block is YES, then the system proceeds to process machine utilization for a CP product,
calculate the number of machines needed for CP adjusted for GP,
calculate the number of machines needed for FT adjusted for FT,
calculate the number of machines needed for a CP product,
calculate the number of machines needed for a FT product,
decide whether there is an insufficient number of machines or an excess number of machines by answering the question “Is utilization greater than or equal to 100%?”,
if the answer is YES, then the system determines that the machine shortage number equals as follows:
(utilization−1)*machine number,
if the answer is NO,
Chan Yi-Hsin
Hsieh Ming-Hsiu
Hsu Yi-Chin
Lai Fu-Kang
Liu Yao-Tung
Ackerman Stephen B.
Hilten John S.
Jones II Graham S.
Saile George O.
Sun Xiuqin
LandOfFree
Method for minimizing total test time for testing factories does not yet have a rating. At this time, there are no reviews or comments for this patent.
If you have personal experience with Method for minimizing total test time for testing factories, we encourage you to share that experience with our LandOfFree.com community. Your opinion is very important and Method for minimizing total test time for testing factories will most certainly appreciate the feedback.
Profile ID: LFUS-PAI-O-2918591