Optimizing Failure Modes and Effects Analysis with Fuzzy

Multiattribute Grey Theory and DEA

SAFIYE TURGAY

Department of Industrial Engineering

Sakarya University

54187, Esentepe Campus Serdivan-Sakarya,

TURKEY

and continual attributes which are related to failure modes and their influence on the system, while cost analysis

done in DEA offers the idea of efficiency solutions that are optimal. By blending both control strategies of

FMEGT and DEA within an integrated framework, FMEA analysis is able to reach greater

effectiveness. Serving as a case study we do so in a series of specific tests and simulations, the approach

proposed successfully analyzes critical failure modes, risk factors, and resource allocation. The results indicate

that the suggested integrated way acts as a facilitator of decision-making by minimizing risk and making

system wise reliability in complex industrial plants.

Key-Words: - Failure Modes and Effects Analysis (FMEA); Fuzzy Logic; Multi-attribute Evaluation; Grey

Theory; Data Envelopment Analysis (DEA); Risk Assessment; Uncertainty; Optimization; Decision Making;

Reliability Analysis

complex systems or process. This technique brings

out the causes and outcomes of failure in a logical

manner that leads to premonitory risk management

and assists organizations to achieve the highest

degree of reliability and safety in respect of their

products or operations. In other words, conventional

FMEA methods are often confronted with problems

due to the fact that they don't perform adequately

when immediate uncertainties and multiple

attributes concerning failure modes and their

this possibility, the study gives the response coming

collating the different components of FMEA and

DEA, the proposed study aims to overcome the

challenges, which were obvious in case of

conventional FMEA techniques to effectively

address complex industrial environments [4], [5],

[6].

By making use of the connections between the fuzzy

logic, multi-attribute assessment, grey theory, and

efficiency analysis, the proposed methodology

provides a universal solution to increasing the

management procedures of various business

industries {7], [8], [9].

The present research develops a gray-related fuzzy

set model, where data envelopment analysis used to

rank alternatives in a more objective way. Section 2

discusses the DEA, which is relevant to this

work. In part 3, the gray-related technique is

described. According to section 4, the ranking of

failure modes by the severity assessment in a

hypothetical FMEA analysis is demonstrated

method is an effective tool for risk assessment and

management that commonly used in diverse

industries, such as manufacturing and engineering,

healthcare and aerospace. Several research works

have been carried out over the years with the

objective of optimizing and streamlining FMEA

procedures so as to address the challenges presented

by more and more complex systems and

processes. In addition to that, the adoption of

advanced methods like Fuzzy Multiattribute Grey

Fuzzy logic has been widely used in FMEA for

uncertainties towards system behavior, failure

modes, and their effects.

The assumption is that F is the group of failure

modes that addressed by the Failure Modes and

Effects Analysis (FMEA). Therefore, E would

correspond to the collection of effects

In which means the weight assigned to each

input variable used for determining the efficiency of

failure mode f, 

membership degree of severity, occurrence, and

detectability of failure mode f corresponding to

assigned attributes.

3.1.3 Integration of FMGT and DEA

delivered by DEA. Develop clear, implementable

solutions and recommendations for the decision-

makers, who will need to consider the overall

picture of the risk factors including their

interconnections.

Which α represents the rating coefficient

established in the assigned assessment (a mixture of

The proposed mathematical model combines Fuzzy

Multicriteria Grey theory and Data Envelopment

Analysis to optimize Failure Modes an Effects

Analysis, which provides a holistic approach of risk

assessment as well as management for complex

systems and processes.

FMEA is a technique perform risk analysis and error

avoidance in planning and development procedure

of products and manufacturing. In other words, the

consequences, as well as effects on the system. We

then need to classify the errors according to their

contribution into the error rate. This research was

aimed to forecast the types of errors and their

impact, error risks, applying possible means to

avoid error-prone situations and the hybridization of

these results in order to produce quality products. At

the stage of estimating the incidence situations of

number to get the decision matrix

The column vectors of the decision matrix A with

calculated as a height might be formally considered

(19)

(21)

established option apparently selected with the

largest score.

There are the different types of these weight

normalization constraints. To illustrate this, assume

that there an inequality constraint that makes the

weight vector to belong to the unit simplex. This

results in the following in equations (22-25),

(23)

The normalisation musts always result into too

many efficient (non-dominated) which are usually

scheme with the leader's decision [28]. Adopting the

assurance region scheme like in DEA studies, the

following model (8) can efficiently be developed so

that the applicable as cone constraints.

makes possible errors of the risky type sneaking in

examination.

- Unveils hypothetical system defects and

interdependencies with subsystems. Use our

automated writing assistant to find out more!

identification the most important activities.

- Rank action for maintenance optimization.

In this study, based on the information received

from experts, the types of faults encountered in

Failure Mode and Effects Analysis is an analysis

prevent failure before it occurs.

effects data.

The raw data is expressed in fuzzy intervals as

shown in Table 2. These data are standardized

(fulfilling Step 3 of the previous section), with

attractive values higher than low. Each criterion is

now on a common 0-1 scale, where 0 represents the

worst imaginable gain in a criterion and 1.00

Conversely, fuzzy weights assigned by the decision

maker or decision-making group reflecting the

relative importance of the criteria. Then DEA to

identify used in a set of efficient sites.

The next step of the gray-related method (Step 4) is

to obtain reference number sequences based on the

optimal weighted interval number value for each

alternative. Since we assumed in the initial analysis

that all weights are in the range [0, 1], the reference

number vector will be the maximum left range value

the lowest risk level are S9- Adapter Fault, S2-

Pinning Fault, S4- General Faults.

Table 7. DEA solutions for all weights.

These averages comprised the basis of the grey-

related method, which was designed to analyze the

performance of alternatives in accordance with the

alternatives corresponding to the one that is

evaluated in each case.

Here ξi(k) is utility information explicitly indicate in

step 2, we changed the contrarian index to the

positive index. Identified efficient collectives can be

ranked and the most effective ones can be

chosen. On the other hand, the criteria to settle a

preorder relation within the set of non-dominated

alternatives are equally baseless.

certainty region DEA scores are shown in Table 8.

In the cost-benefit analysis, the optimal weights are

now all positive showing that all four of the

standards are employed in the evaluation process. In

Table 9, the DEA scores of the identified non-

dominated solutions had been used, the ranking

order of the sites would have been: S3-S5-S6-S7;

Error. Table 10 presents average rates of the all

columns. Hence, those values are located in a larger

area of figures with a higher sensitivity. On top of