Scholarlink

The lack of visual interfaces for developing fuzzy logic in Python is a common problem, as it limited efficient access to the simulation and creation of these systems. This work presents the development of a graphical framework based on the UPAFuzzySystems library in version 0.2.4, oriented towards the design, simulation, and analysis of type 1 fuzzy inference systems. This implementation is carried out using the Tkinter library, and the framework offers complete access to functionalities such as the definition of universes, membership functions, rule generation, and analysis of fuzzy inference systems. This implementation is carried out using the Tkinter library. The framework offers full access to functions such as the definition of universes, membership functions, generation of “If-Then” rules, defuzzification, among other functions. Through the use of libraries such as NumPy and Matplotlib, the system allows fuzzy sets to be represented graphically, as well as output surfaces to be visualized in two and three dimensions. To evaluate the effectiveness of the system, a comparison is made of the time required to create a system from zero, both using the framework and through traditional programming. The evaluation was complemented by the application of the SUS (System Usability Scale) scale, applied to technical users, undergraduate students, people with little programming experience, and some users with previous experience using the UPAFuzzySystems library in its pragmatic form. Obtaining an average score of 70.18, which indicates acceptable usability. Compared to traditional console use, the framework enabled modeling tasks to be completed 50% to 75% faster, confirming the effectiveness of a well-designed interface in technical environments. This work constitutes a versatile tool for teaching, experimenting, and implementing fuzzy systems. We propose extending the framework from models to other approaches such as FLSSMITH Takagi-Sugeno, and even to type 2 fuzzy inference systems, with the aim of broadening its functional reach.

错误：搜索内容不能为空，请输入英文关键词
错误：关键词超出字数限制，请精简
高级检索

Development of a Visual Framework for Type-I Mamdani Systems Based on UPAFuzzySystems

  • 会议论文

  • 发表日期：

    2026年

查看全文
  • David Alonso Carranza Escobar,
  • Martín Montes Rivera,
  • José Eder Guzmán Mendoza,
  • Alberto Ochoa Zezzatti,
  • Julio César Ponce Gallegos

摘要

The lack of visual interfaces for developing fuzzy logic in Python is a common problem, as it limited efficient access to the simulation and creation of these systems. This work presents the development of a graphical framework based on the UPAFuzzySystems library in version 0.2.4, oriented towards the design, simulation, and analysis of type 1 fuzzy inference systems. This implementation is carried out using the Tkinter library, and the framework offers complete access to functionalities such as the definition of universes, membership functions, rule generation, and analysis of fuzzy inference systems. This implementation is carried out using the Tkinter library. The framework offers full access to functions such as the definition of universes, membership functions, generation of “If-Then” rules, defuzzification, among other functions. Through the use of libraries such as NumPy and Matplotlib, the system allows fuzzy sets to be represented graphically, as well as output surfaces to be visualized in two and three dimensions. To evaluate the effectiveness of the system, a comparison is made of the time required to create a system from zero, both using the framework and through traditional programming. The evaluation was complemented by the application of the SUS (System Usability Scale) scale, applied to technical users, undergraduate students, people with little programming experience, and some users with previous experience using the UPAFuzzySystems library in its pragmatic form. Obtaining an average score of 70.18, which indicates acceptable usability. Compared to traditional console use, the framework enabled modeling tasks to be completed 50% to 75% faster, confirming the effectiveness of a well-designed interface in technical environments. This work constitutes a versatile tool for teaching, experimenting, and implementing fuzzy systems. We propose extending the framework from models to other approaches such as FLSSMITH Takagi-Sugeno, and even to type 2 fuzzy inference systems, with the aim of broadening its functional reach.