A Method for Evaluating the Green Economic Efficiency of Resource-Based Cities Based on Neural Network Improved DEA...

Area of Science:

• Environmental Science
• Economics
• Urban Planning

Background:

• Traditional Data Envelopment Analysis (DEA) has limitations in ranking effective units, impacting empirical accuracy for green economic efficiency.
• Resource-based cities face challenges in evaluating and improving their eco-efficiency due to complex interdependencies.

Purpose of the Study:

• To improve the evaluation method for green economic efficiency in resource-based cities using an enhanced DEA model.
• To analyze temporal trends, spatial disparities, and efficiency differences in eco-efficiency.
• To provide targeted policy recommendations for sustainable urban development.

Main Methods:

• Utilizing a Backpropagation (BP) neural network to improve the traditional Data Envelopment Analysis (DEA) model, creating the BP-DEA model.
• Calculating optimal inputs and outputs via DEA, then fitting their relationship with a BP neural network for output prediction.
• Combining the BP-DEA model with the Malmquist index for comparative analysis and decomposition of eco-efficiency and total factor productivity changes.

Main Results:

• The BP-DEA model accurately identifies performance gaps and improvement potential for individual indicators.
• It quantifies necessary adjustments for each indicator and determines appropriate rewards or penalties for cities.
• Analysis revealed temporal evolution, spatial differences, and efficiency variations in eco-efficiency across provinces.

Conclusions:

• The BP-DEA model offers a robust framework for detailed eco-efficiency assessment and comparative analysis.
• Findings provide actionable insights into the drivers of total factor productivity changes.
• The study proposes targeted urban development strategies based on comprehensive eco-efficiency evaluations across urbanization, ecology, and economic dimensions.