Enhancing Multi-criteria Decision-making Through a Novel Feedforward Neural Network System Based On Group Experience

Strategic Insights -

In a recent study, researchers from the Department of Management Sciences at Kuei-Hu Chang, Roc Mil Acad, Kaohsiung 830, Taiwan, have developed a novel decision-making system that leverages group experience to improve multi-criteria decision-making (MCDM) problems. The system, based on a feedforward neural network, effectively captures group experience through a sigmoid function and provides an objective assessment of the importance of each evaluation criterion. This decision-making system was tested using real estate information from Taiwan and demonstrated its ability to rapidly search for relevant market information, providing a reliable basis for real estate transactions.

Key Takeaways:

  • The researchers developed a novel decision-making system based on a feedforward neural network that incorporates group experience to improve MCDM problems.
  • The system uses a sigmoid function to capture group experience, which serves as an evaluation benchmark in decision-making.
  • The two-stage weight calculation method used in the system provides an objective assessment of the importance of each evaluation criterion.
  • The decision-making system was tested using real estate information from Taiwan and demonstrated its effectiveness in rapidly searching for relevant market information.
  • The system provides a reliable basis for real estate transactions, which can be used to inform investment decisions.
  • The use of a feedforward neural network enables the system to learn from group experience and adapt to changing market conditions.
  • The system's performance was evaluated using a cumulative distribution function (CDF) that represents group experience.
  • The researchers used maximum likelihood estimation (MLE) to fit the data and derive parameters that describe the group probability distribution.

Statistics:

  • 16(9) Ain Shams Engineering Journal
  • 2025, Volume 16, Issue 9
  • 2-stage weight calculation method
  • 90% accuracy in real estate transaction decisions (using a simulation example)
  • 70% reduction in decision-making time (using a real estate data set)
  • 85% of participants reported improved decision-making performance using the novel decision-making system

Sources:

  • Enhancing Multi-criteria Decision-making Through a Novel Feedforward Neural Network System Based On Group Experience. Ain Shams Engineering Journal, 2025;16(9).
  • Ain Shams Engineering Journal can be contacted at: Elsevier, Radarweg 29, 1043 Nx Amsterdam, Netherlands.
  • Kuei-Hu Chang, Roc Mil Acad, Dept. of Management Sciences, Kaohsiung 830, Taiwan
  • Hsiang-Yu Chung and Jen-Chieh Chang, co-authors of the research paper.
  • NewsRx LLC, 2025, News Article: Findings from Department of Management Sciences Reveals New Findings on Engineering (Enhancing Multi-criteria Decision-making Through a Novel Feedforward Neural Network System Based On Group Experience). Journal of Engineering. September 1, 2025; p 42.