Policy-informed parametric multi-objective optimisation for neighbourhood land-use allocation


Tak M. D., Akay M., Gülümser A. A.

City and Environment Interactions, sa.30, ss.1-12, 2026 (Scopus)

Özet

Urban neighbourhoods increasingly face misalignments between demographic change, service provision, and legally defined planning standards. While computational optimisation methods have been widely applied to environmental and morphological performance, their integration with binding planning regulations remains limited. This study proposes a policy-informed, multi-objective optimisation framework that combines parametric modelling with the Non-Dominated Sorting Genetic Algorithm II (NSGA-II) to support neighbourhoodscale land-use allocation. The framework simultaneously minimises per-capita service area deficits relative to statutory standards and maximises walking accessibility to essential public services, using a slope-aware network-based accessibility model. Türkiye's Spatial Plan Legislation is embedded as a hard constraint, ensuring institutional feasibility. The approach is applied to Goktürk ¨ (Istanbul), where 10,000 alternatives are generated under two scenarios (current-trend growth and plan-based development), yielding 2343 Paretooptimal solutions. Results reveal systematic trade-offs between adequacy and accessibility: solutions that minimise service area deficits tend to concentrate facilities on larger plots, reducing spatial coverage, while accessibility-oriented solutions distribute facilities more broadly but leave residual area deficits. Across scenarios, the current-trend growth scenario yields 46% more service area than the plan-based scenario, yet underperforms by 8% in walking accessibility. Rather than prescribing a single optimal plan, the framework produces a decision space of regulation-compliant alternatives, supporting transparent and context-sensitive planning choices. The study contributes to computational urban science by extending multi-objective optimisation into the regulatory domain and positioning parametric methods as planning support systems that enhance transparency, enable participatory deliberation, and support adaptive urban governance.