Scribd Logo
Upload

Welcome to Scribd!

  • 56 views

    Fuzzy Logic: Assignment

    Uploaded by

    Chirag Gautam
    The document describes a fuzzy logic model for supplier selection. It discusses how fuzzy logic can capture the subjectivity involved in expressing preferences compared to strict numeric scales. The model involves 5 steps: 1) identifying selection criteria, 2) assessing supplier performance, 3) using fuzzy membership functions to determine supplier scores, 4) converting scores to rankings, 5) selecting the highest ranked supplier. However, the model only provides rankings and does not indicate differences between suppliers or how changes in criteria would impact scores. The document then proposes improvements to the fuzzy logic model. It describes adding 3 phases: 1) fuzzification to convert crisp inputs to fuzzy values, 2) an inference mechanism using IF-THEN rules to determine outputs, 3

    Copyright:

    Attribution Non-Commercial (BY-NC)
    Download as PDF, TXT or read online from Scribd

    Fuzzy Logic: Assignment

    Uploaded by

    Chirag Gautam
    56 views8 pages
    The document describes a fuzzy logic model for supplier selection. It discusses how fuzzy logic can capture the subjectivity involved in expressing preferences compared to strict numeric scales. The model involves 5 steps: 1) identifying selection criteria, 2) assessing supplier performance, 3) using fuzzy membership functions to determine supplier scores, 4) converting scores to rankings, 5) selecting the highest ranked supplier. However, the model only provides rankings and does not indicate differences between suppliers or how changes in criteria would impact scores. The document then proposes improvements to the fuzzy logic model. It describes adding 3 phases: 1) fuzzification to convert crisp inputs to fuzzy values, 2) an inference mechanism using IF-THEN rules to determine outputs, 3

    Original Description:

    Original Title

    Fuzzy Logic

    Copyright

    © Attribution Non-Commercial (BY-NC)

    Available Formats

    PDF, TXT or read online from Scribd

    Did you find this document useful?

    Is this content inappropriate?

    The document describes a fuzzy logic model for supplier selection. It discusses how fuzzy logic can capture the subjectivity involved in expressing preferences compared to strict numeric scales. The model involves 5 steps: 1) identifying selection criteria, 2) assessing supplier performance, 3) using fuzzy membership functions to determine supplier scores, 4) converting scores to rankings, 5) selecting the highest ranked supplier. However, the model only provides rankings and does not indicate differences between suppliers or how changes in criteria would impact scores. The document then proposes improvements to the fuzzy logic model. It describes adding 3 phases: 1) fuzzification to convert crisp inputs to fuzzy values, 2) an inference mechanism using IF-THEN rules to determine outputs, 3

    Copyright:

    Attribution Non-Commercial (BY-NC)
    Download as PDF, TXT or read online from Scribd
    Download as pdf or txt
    56 views8 pages

    Fuzzy Logic: Assignment

    Uploaded by

    Chirag Gautam
    The document describes a fuzzy logic model for supplier selection. It discusses how fuzzy logic can capture the subjectivity involved in expressing preferences compared to strict numeric scales. The model involves 5 steps: 1) identifying selection criteria, 2) assessing supplier performance, 3) using fuzzy membership functions to determine supplier scores, 4) converting scores to rankings, 5) selecting the highest ranked supplier. However, the model only provides rankings and does not indicate differences between suppliers or how changes in criteria would impact scores. The document then proposes improvements to the fuzzy logic model. It describes adding 3 phases: 1) fuzzification to convert crisp inputs to fuzzy values, 2) an inference mechanism using IF-THEN rules to determine outputs, 3

    Copyright:

    Attribution Non-Commercial (BY-NC)
    Download as PDF, TXT or read online from Scribd
    Download as pdf or txt
    of 8

    Fuzzy Logic

    ASSIGNMENT

    Submitted by: Chirag Gautam Roll No. 48 IM- 17

    1. Describe the fuzzy logic used in the supplier selection model and how far is it effective? Ans: In the supplier selection model used here the rankings of the potential suppliers are assigned based on two factors: the importance weight of the attributes, and suppliers performance with respect to these attributes. Both of these factors are decision maker-specific and solicited from the individuals. Often elicitation process is conducted by asking the decision makers to express their preferences in pure numeric scales. The main difficulty with such an elicitation procedure is that the subjectivity and imprecision associated with perceptions are lost by forcing the decision makers to use numeric scales. Fuzzy logic captures the subjectivity involved in expressing individual preferences and allows decision makers preferences to be expressed in linguistic terms.

    Steps:
    i) ii) iii)

    Identifying the appropriate selection criteria. Perception of the suppliers performance with respect to these criteria. Using fuzzy membership functions and fuzzy mathematical operators a fuzzy score is determined for each supplier.

    iv) v)

    Fuzzy scores are then converted to crisp values through defuzzification process to make the rankings. Supplier with the highest ranking is selected.

    Step i) The selection criterias Delivery Compliance with due date, Fill rate, lead time, Flexibility Service Reliability, empathy, Responsiveness, Assurance Product Product range, new product availability, Recycled materials, Ergonomic features Quality Quality Control rejection rate, Customer rejection rate Cost Purchase price, logistics

    Step ii)

    Step iii) Define two fuzzy membership functions: one for assessment of the attribute weights and one for performance ratings of the suppliers. The importance of each supplier selection criterion is evaluated by a question with the answer set of low importance, moderate importance, high importance, and very high importance. These values correspond to fuzzy numbers on the numeric scale 0-1. The performance of a supplier with respect to each criterion is evaluated by a question with the answer set of excellent, very good, good, and poor. These values correspond to fuzzy numbers on the numeric scale 010. Using the appropriate fuzzy membership functions, convert the linguistic terms into fuzzy weights.

    The weights are calculated by multiplying the fuzzy scores of each criterion with its sub criteria. For example w2 the importance weight for quality with respect to customer rejection rate is computed by multiplying the importance weight of quality (VH) by importance weight of the customer rejection rate (M). The procedure is followed for rest.

    The linguistic supplier performance ratings are translated into fuzzy performance ratings and the suppliers fuzzy scores are calculated by multiplying fuzzy performance ratings matrix by fuzzy importance weights matrix.

    Where ssj = Supplier fuzzy score rji = Supplier fuzzy performance ratings matrix wi = Importance weight matrix

    Step v) Convert the aggregate fuzzy scores into crisp scores using the Center of Area (COA) defuzzification approach. Step vi) Rank the suppliers according to their crisp scores. The supplier that has the highest overall crisp score is assigned the highest ranking and will be selected.

    The proposed approach of using Fuzzy Logic (FL) in the case study whilst able to capture the subjectivity involved in expressing individual preferences and circumvents the need for rigorous mathematical modeling, it just provides the ranking of the three suppliers A, B, and C. It provides a quantitative ranking system without having to trade off one for the other. The scale used to determine the crisp score does not say much about how close, or far, they are from each other. There is no way of identifying how the different factors affect the final score. For example what happens if the importance of one of the factors is changed?

    2. Modify and improve the model with fuzzy logic techniques. Inference mechanism in Fuzzy logic is implemented in three phases Phase-1: Fuzzification phase (converting crisp input value into fuzzy value). Phase-2: Inference phase (computing fuzzy output value by the fuzzy rules base). Phase-3: Defuzzification phase (converting fuzzy output value into crisp value).

    Fuzzy logic system

    Delivery Compliance with due date, Fill rate, lead time, Flexibility Service Reliability, empathy, Responsiveness, Assurance Product Product range, new product availability, Recycled materials, Ergonomic features Quality Quality Control rejection rate, Customer rejection rate Cost - Purchase price, logistics

    i)

    Fuzzification: Fuzzification means mapping an input value into fuzzy membership functions, thus

    converting the crisp input value into fuzzy value. Delivery Fuzzy 1 2 3 Variable Slow Medium Fast Range 0 - 10 5 15 10 - 20

    Service Fuzzy 1 2 3 Variable Cheap Optimum Good Range 08 4 12 10 - 16

    Product Fuzzy 1 2 3 Variable Low Medium High Range 0 - 10 5 15 10 - 20

    Quality Fuzzy 1 2 3 Variable Poor Acceptable Good Range 0 - 20 10 30 20 - 40

    Cost Fuzzy 1 2 3 Variable Less Medium High Range 0 - 10 5 15 10 - 20

    Supplier Fuzzy 1 2 3 Variable Reject Under consider Accept Range 06 39 6 - 12

    Rule construction: Fuzzy logic system makes decisions and generates output values based on knowledge provided by the designer in the form of IF condition-THEN- action_-rules. The rule base specifies qualitatively how the output parameter overall rating of the suppler proposal is determined for various instances of the input parameters of cost, delivery, quality, product and Service and output parameter is suppler selection. Example rule Rule 1: IF Cost is less AND Quality is poor AND Delivery is slow AND Product is low AND Serivce is low THEN Price Matching is high.

    When Product is low AND Serivce is low Cost Less Quality Poor Reject Reject Reject Medium High

    Acceptable

    Accept

    Reject

    Reject Under consideration

    Good

    Accept

    Under consideration

    There are 5 factors with 3 levels therefore 243 rules are possible with various combinations.

    Defuzzification When the inference process is complete, the resulting data for each output of the fuzzy classification system are a collection of fuzzy sets or a single, aggregate fuzzy set. The process of computing a single number that best represents the outcome of the fuzzy set evaluation is called defuzzification. Defuzzification can be performed by deriving the centre of gravity method.

    The logic used provides a decision support for the ranking of suppliers and also explaining the tradeoffs between the criteria. --------------------------------------------------------------------------------------------------------------------------------------------

    You might also like