Latent effects decision analysis

Details Latent effects decision analysis Latent effects decision analysis

2000-09-28

2004-08-24

Khatri, Anil (Department: 2121)

Data processing: artificial intelligence

Fuzzy logic hardware

Fuzzy inference processing

C706S002000, C706S052000, C706S048000

Reexamination Certificate

active

06782372

ABSTRACT:

CROSS REFERENCE TO RELATED APPLICATIONS
(Not Applicable)
BACKGROUND OF THE INVENTION
It is conventional in decision analysis to weigh a variety of diverse factors and combine these factors either directly or through proportional logic to derive metrics. For example, a student's grade may be based 20% on homework scores, 10% on class participation, 10% on classroom behavior, and 60% on test scores. Such a grade would be
G
=
∑
i
4
⁢
w
i
⁢
x
i
1
where w
i
=the individual weights; x
i
=the individual grades; and
∑
l
n
⁢
w
i
=
1.
This linear analysis has been used for many years in education and many other fields; however, it and other known analysis systems may not be satisfactory for weighing factors that are uncertain and which do not combine linearly or independently. Surety analysis and the associated decision analysis for such conventional systems are basically applicable to only experience-measurable or physical-model-derived data. However, most practical analyses, including high-consequence system surety analysis, must also utilize subjectivity.
It is a challenging task to systematically (preferably mathematically) assess subjective factors such as organizational factors as to their effects on system operation. Furthermore, these are generally latent factors that influence critical system operation over time. For instance, if a grade is supposed to measure the student's knowledge in a subject, other factors relating to the school, the teacher, the administration, and the student's home life also have an effect on the student's knowledge, but these factors are not accounted for in the traditional system.
There has been considerable prior effort on analytically incorporating engineering judgment into analysis systems. For example, Dempster-Shafer theory establishes a framework in which frequentist probability and Bayesian incorporation of new data are subsets. Although these methodologies both allow judgment, neither derives results that can explicitly indicate the relative amounts of subjective judgment and measurable data in the results.
For example, the safety status of an airline operation might depend on measuring factors such as the implications of accident/incident statistics, maintenance personnel/pilot competence and experience, scheduling pressures, and the safety “culture” of the organization. Many of the potential metrics on such parameters are difficult (and generally uncertain) to determine. A top-down analytical approach requires more than individual parameter assessment which is used in some conventional tabular schemes. Furthermore, aggregation of the parameters into an overall metric requires a methodology that can account for nonlinearities and dependence. Twice as many attributes is unlikely to be twice as beneficial, and scheduling pressures could influence accident/incident statistics.
SUMMARY OF THE INVENTION
It is an object of this invention is to provide a new risk management decision-aid methodology.
It is also an object of this invention to aggregate data that are inputs (both objective and subjective) to each subsystem by a “soft aggregation” process.
It is a further object to track uncertainty whereby inputs may be entered with uncertainty measures and output uncertainty may be correspondingly determined.
It is another object to provide a technique to account for dependence between or among inputs (which is by far the most common case as opposed to independent inputs).
It is also an object of the invention to account for “importance” (contribution of each input to the output value(s)) and “sensitivity” (potential for improvement in the output due to each input(s)).
And it is an object of the invention to be able to determine “trends” for each input and output over time.
It is also an object of this invention to display inputs as they are entered, outputs as they are aggregated for overall contribution or “early alerts,” in a manner that is unusually natural for humans to understand and work with (instrumentation-like displays portrayed on a computer screen)
To achieve the foregoing and other objects, and in accordance with the purpose of the present invention, as embodied and broadly described herein, the present invention may comprise a method of nonlinear decision analysis, comprising providing a plurality of inputs having either positive or negative values, and determining an output representative of a decision as a function of the inputs, where the output approaches a first value asymptotically as the value of positive inputs increases, and approaches a second value asymptotically as the value of negative inputs increases. The invention further includes instrumentation-like gage displays (giving a human-friendly interface and display of uncertainty range) for displaying inputs and outputs, an architecture of subsystem modules that are interconnected to reflect latent effects, and a “soft aggregation” combination of input data to reflect nonlinear accumulation of effects.
Additional objects, advantages, and novel features of the invention will become apparent to those skilled in the art upon examination of the following description or may be learned by practice of the invention. The objects and advantages of the invention may be realized and attained as particularly pointed out in the appended claims.


REFERENCES:
patent: 5014220 (1991-05-01), McMann et al.
patent: 5682309 (1997-10-01), Bartusiak et al.
patent: 5754738 (1998-05-01), Saucedo et al.
patent: 5774121 (1998-06-01), Stiegler
patent: 5963447 (1999-10-01), Kohn et al.
patent: 5999923 (1999-12-01), Kowalski et al.
patent: 6088689 (2000-07-01), Kohn et al.
patent: 6172966 (2001-01-01), Taillibert et al.
patent: 6397202 (2002-05-01), Higgins et al.
patent: 6611735 (2003-08-01), Henly et al.
patent: 2001/0053991 (2001-12-01), Bonabeau
patent: 2002/0165837 (2002-11-01), Zhang et al.
patent: 2003/0115175 (2003-06-01), Baatz et al.
patent: 406096055 (1994-04-01), None
Evans et al., “Multicriteria Optimization of Simulation Models”, Proceedings of the 1991 Winter Simulation Conference, pp. 894-900, 1991.*
Cooper, J.A., “Non-Traditional Performance Analysis”, Proceedings of the 2nd International Conference on Intelligent Processing and manufacturing of Materials, vol. 1, pp. 191-194, Jul. 1999.*
O'Rorke, P., “Qualitative Decision Analysis: Comparing Alternative Approaches”, NEC Research Index, Oct. 1995.*
Slagle et al., “An Expert System for a Resource Allocation Problem”, Communications of the ACM, vol. 28, No. 9, Sep. 1985.*
Stanley et al., “A Cost-Benefit Decision Model: Analysis, Comparison, and Selection of Data Management Systems”, ACM Transactions n Database Systems, vol. 12, No. 3, pp. 472-520, Sep. 1987.*
Cooper et al., “Improved Safety Analysis Through Enhanced Mathematical Structures”, 1997 IEEE International Conference on Systems, Man, and Cybernetics, vol. 2, pp. 1656-1661, Oct. 1997.*
Cooper, J.A., “The Application of Fuzzy Mathematics in Safety Analysis*”, Proceedings of the 5th IEEE International Conferenc on Fuzzy Systems, vol. 2, pp. 1002-1006, Sep. 1996.*
Cooper et al., “The Application of New Mathematical Structures to Safety Analysis”, The 1998 IEEE International Conference o Fuzzy Systems Proceedings, vol. 1, pp. 692-697, May 1998.*
Covan et al., “Predictable Safety in the Control of High Consequence Systems”, 3rd IEEE International Conference on High-Assurance Systems Engineering Symposium, pp. 200-204, Nov. 1998.*
Cheng et al., “Generation of High Quality Non-Robust Tests for Path Delay Faults”, 31st ACM/IEEE Design Automation Conference, 1994, pp. 365-369.*
Chillarege et al., “The Effect of System Workload on Error Latency: An Experimental Study”, 1985, pp. 69-77.

Affiliated with

Also associated with

Rating

Latent effects decision analysis does not yet have a rating. At this time, there are no reviews or comments for this patent.

If you have personal experience with Latent effects decision analysis, we encourage you to share that experience with our LandOfFree.com community. Your opinion is very important and Latent effects decision analysis will most certainly appreciate the feedback.

Rate now

Comments { 0 }

Profile ID: LFUS-PAI-O-3304493