Multi objective Optimization for Sustainable Urban Land-Use Planning Under Ecological Constraints

Abstract Achieving sustainability and environmental quality, public safety and economic development are complex trade-offs in urban land-use planning. These goals are typically planned separately, which reduces their ability to put together solutions that are both balanced and sustainable. This study suggests a framework for multi-objective optimisation, which combines data-driven predictive modelling, feature sensitivity analysis and evolutionary optimization, for sustainable urban land use. A surrogate model called A Histogram Based Gradient Boosting (HistGB) is created to explore the nonlinear relationships between 15 sustainability indicators, and the hyper parameters of HistGB are optimized by Novel Hybrid Kepler Optimization Algorithm (NOHKOA). The Dynamic Trade-off Coefficient Selection Multi-Objective (DTCSMO) algorithm is then used to optimize conflicting sustainability, environmental burden and urban risk goals while integrating continuous and discrete planning variables. The coefficient of determination for the test data reaches 0.9877 and the mean absolute error (MAE) is 0.0152 and the root mean squared error (RMSE) is 0.0193 for the HistGB model. For feature sensitivity analysis, the most sensitive features are green cover (mean change in coefficient of determination: 0.8196) and renewable energy use (mean change in coefficient of determination: 0.3584). The results of the ablation experiment also show the significance of the environmental indicator, and the test coefficient of determination drops to 0.1542 when the environmental indicator is removed. The DTCSMO algorithm can produce 120 Pareto-optimal solutions, allowing decision makers to consider different trade-offs between the planning goals. For the evaluation of the characteristics of the solutions a comparison with the multi-objective particle swarm optimization algorithm is additionally provided. In summary, the proposed framework provides a data-driven method for urban land-use planning that is accurate, interprets the data through sensitivity analysis and incorporates multiple objectives in decision making.

Authors

Institutions

Publication Details

Journal
Research Square
Published
2026-11-13
DOI
https://doi.org/10.21203/rs.3.rs-11505560/v1
Primary Topic
Advanced Multi-Objective Optimization Algorithms
Type
preprint
Controls
|||
ALL TIME
JAN
FEB
MAR
APR
MAY
JUN
JUL
AUG
SEP
OCT
NOV
preprint

Multi objective Optimization for Sustainable Urban Land-Use Planning Under Ecological Constraints

Zhao Lina, 晶 崔, Yiyun Wang
Research Square
Advanced Multi-Objective Optimization Algorithms
preprint

Multi objective Optimization for Sustainable Urban Land-Use Planning Under Ecological Constraints

Zhao Lina, 晶 崔, Yiyun Wang
preprint en

Abstract

Abstract Achieving sustainability and environmental quality, public safety and economic development are complex trade-offs in urban land-use planning. These goals are typically planned separately, which reduces their ability to put together solutions that are both balanced and sustainable. This study suggests a framework for multi-objective optimisation, which combines data-driven predictive modelling, feature sensitivity analysis and evolutionary optimization, for sustainable urban land use. A surrogate model called A Histogram Based Gradient Boosting (HistGB) is created to explore the nonlinear relationships between 15 sustainability indicators, and the hyper parameters of HistGB are optimized by Novel Hybrid Kepler Optimization Algorithm (NOHKOA). The Dynamic Trade-off Coefficient Selection Multi-Objective (DTCSMO) algorithm is then used to optimize conflicting sustainability, environmental burden and urban risk goals while integrating continuous and discrete planning variables. The coefficient of determination for the test data reaches 0.9877 and the mean absolute error (MAE) is 0.0152 and the root mean squared error (RMSE) is 0.0193 for the HistGB model. For feature sensitivity analysis, the most sensitive features are green cover (mean change in coefficient of determination: 0.8196) and renewable energy use (mean change in coefficient of determination: 0.3584). The results of the ablation experiment also show the significance of the environmental indicator, and the test coefficient of determination drops to 0.1542 when the environmental indicator is removed. The DTCSMO algorithm can produce 120 Pareto-optimal solutions, allowing decision makers to consider different trade-offs between the planning goals. For the evaluation of the characteristics of the solutions a comparison with the multi-objective particle swarm optimization algorithm is additionally provided. In summary, the proposed framework provides a data-driven method for urban land-use planning that is accurate, interprets the data through sensitivity analysis and incorporates multiple objectives in decision making.

Research Square
Hebei University of Engineering (CN)
Sustainable cities and communities
Advanced Multi-Objective Optimization Algorithms
AI Navigator

Ask Laika to Summarize, Analyze, and Connect papers live on the map.

Summarize Papers & Methodologies

Extract key findings, datasets, and comparative methods across publications.

Benchmark Rankings & Visual Analytics

Rank top research institutions, authors, funders, topics, and journals by Field-Weighted Citation Impact (FWCI) and paper volume with instant charts.

Connect Distant Disciplines

Bridge topological clusters on the map to find hidden collaborative intersections.

Laika access coming soon