Monotone Fuzzy Systems
Associate Professor Dr Tay Kai Meng
Data science niche: Foundations of Data Science
INTRODUCTION
Associate Professor Dr Tay Kai Meng
Data science niche: Foundations of Data Science
INTRODUCTION
This project focus on constructing monotone fuzzy systems, from data and/or fuzzy if-then rules from human experts. This include to derive parametric conditions for a Fuzzy Inference System to be monotone, as well as, devising various approach to deal with data and/or fuzzy if-then rules which is inconsistent with the derived parametric conditions. The use of monotone fuzzy systems in various real-world applications, including, n-Ary aggregation, FMEA, risk analytic, and education assessment are illustrated. We have also proposed a number of new theorems, hypothesis and approaches pertaining to monotone fuzzy systems. This includes a new hypothesis stating that monotone data does not always produce monotone fuzzy rules. We also proposed MIFIS, monotone fuzzy rule relabeling, monotone fuzzy rule interpolations and etc.
OBJECTIVES
To propose a generic method to simplify the fuzzy logic-based failure mode and effect analysis (FMEA) methodology by reducing the number of rules that needs to be provided by FMEA users for the fuzzy risk priority number (RPN) modeling process.
METHODOLOGI
The fuzzy RPN approach typically requires a large number of rules, and it is a tedious task to obtain a full set of rules. The larger the number of rules provided by the users, the better the prediction accuracy of the fuzzy RPN model. As the number of rules required increases, ease of use of the model decreases since the users have to provide a lot of information/rules for the modeling process. A guided rules reduction system (GRRS) is thus proposed to regulate the number of rules required during the fuzzy RPN modeling process. The effectiveness of the proposed GRRS is investigated using three real-world case studies in a semiconductor manufacturing process.
1. Define the scale Table of Severity, Occurrence, and Detect.
2. Studies intent, purpose, goal, objective of a product/process. Generally, it is identified by interaction among components/process flow diagram followed by task analysis.
3. Identify potential failures of product/process; this includes problems, concerns, and opportunity of improvement
4. Identify consequence of failures to other components/next processes, operation, customers and government regulations.
5. Identify the potential root cause of potential failures.
6. First level method/procedure to detect/prevent failures of product/process.
7. Severity rating: rank the seriousness of the effect of the potential failures.
8. Occurrence rating: estimation of the frequency for a potential cause of failures.
9. Detect rating: likelihood of the process control to detect a specific root cause of a failure.
10. RPN calculation: product of the three inputs rating; severity, occurrence, detect.
11. Correction. back to (2) if available.
12. End
RESULTS
In this paper, we argued that not all the rules are actually required in the fuzzy RPN model. Eliminating some of the rules does not necessarily lead to a significant change in the model output. However, some of the rules are vitally important and cannot be ignored. The proposed GRRS is able to provide guidelines to the users which rules are required and which can be eliminated. By employing the GRRS, the users do not need to provide all the rules, but only the important ones when constructing the fuzzy RPN model. The results obtained from the case studies demonstrate that the proposed GRRS is able to reduce the number of rules required and, at the same time, to maintain the ability of the Fuzzy RPN model to produce predictions that are in agreement with experts’ knowledge in risk evaluation, ranking, and prioritization tasks.
OBJECTIVES
To propose a generic method to simplify the fuzzy logic-based failure mode and effect analysis (FMEA) methodology by reducing the number of rules that needs to be provided by FMEA users for the fuzzy risk priority number (RPN) modeling process.
METHODOLOGI
The fuzzy RPN approach typically requires a large number of rules, and it is a tedious task to obtain a full set of rules. The larger the number of rules provided by the users, the better the prediction accuracy of the fuzzy RPN model. As the number of rules required increases, ease of use of the model decreases since the users have to provide a lot of information/rules for the modeling process. A guided rules reduction system (GRRS) is thus proposed to regulate the number of rules required during the fuzzy RPN modeling process. The effectiveness of the proposed GRRS is investigated using three real-world case studies in a semiconductor manufacturing process.
1. Define the scale Table of Severity, Occurrence, and Detect.
2. Studies intent, purpose, goal, objective of a product/process. Generally, it is identified by interaction among components/process flow diagram followed by task analysis.
3. Identify potential failures of product/process; this includes problems, concerns, and opportunity of improvement
4. Identify consequence of failures to other components/next processes, operation, customers and government regulations.
5. Identify the potential root cause of potential failures.
6. First level method/procedure to detect/prevent failures of product/process.
7. Severity rating: rank the seriousness of the effect of the potential failures.
8. Occurrence rating: estimation of the frequency for a potential cause of failures.
9. Detect rating: likelihood of the process control to detect a specific root cause of a failure.
10. RPN calculation: product of the three inputs rating; severity, occurrence, detect.
11. Correction. back to (2) if available.
12. End
RESULTS
In this paper, we argued that not all the rules are actually required in the fuzzy RPN model. Eliminating some of the rules does not necessarily lead to a significant change in the model output. However, some of the rules are vitally important and cannot be ignored. The proposed GRRS is able to provide guidelines to the users which rules are required and which can be eliminated. By employing the GRRS, the users do not need to provide all the rules, but only the important ones when constructing the fuzzy RPN model. The results obtained from the case studies demonstrate that the proposed GRRS is able to reduce the number of rules required and, at the same time, to maintain the ability of the Fuzzy RPN model to produce predictions that are in agreement with experts’ knowledge in risk evaluation, ranking, and prioritization tasks.
REFERENCES
Kerk, Yi Wen, Kai Meng Tay, and Chee Peng Lim. "Monotone Fuzzy Rule Interpolation for Practical Modeling of the Zero-Order TSK Fuzzy Inference System." IEEE Transactions on Fuzzy Systems 30.5 (2022): 1248-1259.
Kerk, Yi Wen, Chin Ying Teh, Kai Meng Tay, and Chee Peng Lim "Parametric conditions for a monotone TSK fuzzy inference system to be an n-Ary aggregation function" IEEE Transactions on Fuzzy Systems 29.7 (2020): 1864-1873.
Kerk, Yi Wen, Kai Meng Tay, and Chee Peng Lim. "Monotone interval fuzzy inference systems" IEEE Transactions on Fuzzy Systems 27.11 (2019): 2255-2264.
Teh, Chin Ying, Yi Wen Kerk, Kai Meng Tay, and Chee Peng Lim. "On modeling of data-driven monotone zero-order TSK fuzzy inference systems using a system identification framework" IEEE Transactions on Fuzzy Systems 26.6 (2018): 3860-3874.
Pang, Lie Meng, Kai Meng Tay, and Chee Peng Lim. "Monotone fuzzy rule relabeling for the zero-order TSK fuzzy inference system" IEEE Transactions on Fuzzy Systems 24.6 (2016): 1455-1463.
Jee, Tze Ling, Kai Meng Tay, and Chee Peng Lim. "A new two-stage fuzzy inference system-based approach to prioritize failures in failure mode and effect analysis" IEEE Transactions on Reliability 64.3 (2015): 869-877.
Fuzzy FMEA procedure (Tay and Lim, 2006, doi.org/10.1108/02656710610688202)
Pang, Lie Meng, Kai Meng Tay, and Chee Peng Lim. "Monotone fuzzy rule relabeling for the zero-order TSK fuzzy inference system" IEEE Transactions on Fuzzy Systems 24.6 (2016): 1455-1463.
Jee, Tze Ling, Kai Meng Tay, and Chee Peng Lim. "A new two-stage fuzzy inference system-based approach to prioritize failures in failure mode and effect analysis" IEEE Transactions on Reliability 64.3 (2015): 869-877.
Fuzzy FMEA procedure (Tay and Lim, 2006, doi.org/10.1108/02656710610688202)
