Double weighted combat data quality evaluation method based on CVF optimized FAHP.

Area of Science:

• Military simulation
• Data quality assessment
• Computational intelligence

Background:

• Accurate combat data quality assessment is crucial for effective multi-agent combat simulations.
• Current evaluation methods lack the necessary accuracy to adequately support these exercises.
• This deficiency hinders the reliability and utility of simulation outcomes.

Purpose of the Study:

• To propose an advanced method for evaluating combat data quality in multi-agent simulations.
• To enhance the accuracy and reliability of combat data assessment.
• To provide better data support for military simulation exercises and drills.

Main Methods:

• Developed a three-tiered evaluation framework for combat data quality indicators.
• Optimized Fuzzy Analytic Hierarchy Process (FAHP) weights using the Satisfaction Consistency Approach.
• Constructed a Comparative Value Function (CVF) to derive second-tier weights for a double-weighted evaluation.

Main Results:

• The proposed Double-Weighted FAHP optimized by CVF method significantly reduced the mean squared error to 5.35.
• This represents a notable improvement compared to FAHP, interval intuitionistic fuzzy methods, and artificial neural networks.
• The results indicate the method's output is closer to actual standard values.

Conclusions:

• The developed method offers a more accurate evaluation of multi-agent combat data quality.
• This enhanced accuracy provides robust data support for future combat simulation exercises.
• The findings contribute to improving the effectiveness of military drills and strategic planning.