TY - JOUR
T1 - Advanced artificial intelligence system by intuitionistic fuzzy Γ-subring for automotive robotic manufacturing
AU - Firouzkouhi, Narjes
AU - Amini, Abbas
AU - Nazari, Marziyeh
AU - Alkhatib, Fadi
AU - Bordbar, Hashem
AU - Cheng, Chun
AU - Davvaz, Bijan
AU - Rashidi, Maria
PY - 2023
Y1 - 2023
N2 - In recent years, robotic engineering has been enriched with Artificial Intelligence (AI) technology, preparing the industries to enter the Industry 4.0 era. The powerful neoteric paradigm of AI can serve automotive industries (as one of the largest sectors in the world), to inevitably change their outdated manufacturing strategies. These industrial sectors are increasingly encountering mega data that inevitably carry uncertainty, for which the available methodologies are not capable to deal with that efficiently. To theoretically resolve this gap, a generalized intuitionistic fuzzy set (IFS) theory is proposed here as an efficient, fast, and flexible method. Based on the membership and non-membership degrees, multi-aspect Γ -systems is developed to model the complex real systems. Inspired by multi-attribute Γ -systems and IFS approach, a novel mathematical concept namely intuitionistic fuzzy Γ -subring (IFΓ R) method, is developed to establish an AI platform for robotic automotive manufacturing. Significant characteristics of IFΓ R are developed, including the overlapping of elements with IFΓ R property is IFΓ R, also image and inverse image of elements with IFΓ R property are IFΓ R under Γ -ring homomorphism. Additionally, the connection between upper and lower bound level cuts and image/inverse image property are parametrically discussed. With the effect of surjective homomorphism on upper and lower level cuts, there would be equivalent upper and lower level cuts of image/inverse image in IFΓ R environment. The developed notion of IFΓ I is obtained as the generalization of Γ -ideal under Γ -ring R along with the resultant fundamental properties of IFΓ I, where the overlapping/intersection family of IFΓ Is is proved to be IFΓ I. Also, the upper and lower bound level cuts of elements with IFΓ I property are Γ -ideals. Finally, the proposed IFΓ R method is utilized for automotive AI systems (AAIS) by means of mathematical algebraic notions of Γ -ring, IFS, Γ -ring isomorphism, and upper and lower bound levels. The developed methodology is validated using real dataset of industrial robots in supply chain and then, the elements are characterized in terms of metric overall factory effectiveness. With a systematic pattern of Γ -ring structure, the IFΓ R model is accomplished on elements, and the intercomponent correspondence of AAIS is established with the Γ -ring isomorphism. Based on QC (quality criteria) and non-QC indexes, as the derivation of upper and lower bound level cuts, the analysis of parameters (robots) is simplified for the identification of effective and compatible components in AAIS. The generalized IFS-based method for complex systems has a potential to be used in different AI platforms.
AB - In recent years, robotic engineering has been enriched with Artificial Intelligence (AI) technology, preparing the industries to enter the Industry 4.0 era. The powerful neoteric paradigm of AI can serve automotive industries (as one of the largest sectors in the world), to inevitably change their outdated manufacturing strategies. These industrial sectors are increasingly encountering mega data that inevitably carry uncertainty, for which the available methodologies are not capable to deal with that efficiently. To theoretically resolve this gap, a generalized intuitionistic fuzzy set (IFS) theory is proposed here as an efficient, fast, and flexible method. Based on the membership and non-membership degrees, multi-aspect Γ -systems is developed to model the complex real systems. Inspired by multi-attribute Γ -systems and IFS approach, a novel mathematical concept namely intuitionistic fuzzy Γ -subring (IFΓ R) method, is developed to establish an AI platform for robotic automotive manufacturing. Significant characteristics of IFΓ R are developed, including the overlapping of elements with IFΓ R property is IFΓ R, also image and inverse image of elements with IFΓ R property are IFΓ R under Γ -ring homomorphism. Additionally, the connection between upper and lower bound level cuts and image/inverse image property are parametrically discussed. With the effect of surjective homomorphism on upper and lower level cuts, there would be equivalent upper and lower level cuts of image/inverse image in IFΓ R environment. The developed notion of IFΓ I is obtained as the generalization of Γ -ideal under Γ -ring R along with the resultant fundamental properties of IFΓ I, where the overlapping/intersection family of IFΓ Is is proved to be IFΓ I. Also, the upper and lower bound level cuts of elements with IFΓ I property are Γ -ideals. Finally, the proposed IFΓ R method is utilized for automotive AI systems (AAIS) by means of mathematical algebraic notions of Γ -ring, IFS, Γ -ring isomorphism, and upper and lower bound levels. The developed methodology is validated using real dataset of industrial robots in supply chain and then, the elements are characterized in terms of metric overall factory effectiveness. With a systematic pattern of Γ -ring structure, the IFΓ R model is accomplished on elements, and the intercomponent correspondence of AAIS is established with the Γ -ring isomorphism. Based on QC (quality criteria) and non-QC indexes, as the derivation of upper and lower bound level cuts, the analysis of parameters (robots) is simplified for the identification of effective and compatible components in AAIS. The generalized IFS-based method for complex systems has a potential to be used in different AI platforms.
UR - https://hdl.handle.net/1959.7/uws:69272
M3 - Article
SN - 1573-7462
SN - 0269-2821
VL - 56
SP - 9639
EP - 9664
JO - Artificial Intelligence Review
JF - Artificial Intelligence Review
IS - 9
ER -