Green infrastructure (GI) has a decisive influence on the quality of citizen’s life and species diversity even though it composed of small and fragmented green spaces, thus the planning of GI is essential in various public renewal projects. However, little is known what is the best GI plan for the whole region, where composing of multiple sites. This is very complex NP-hard problem because regional-scale GI plan should maximize the total greenery benefit from multiple sites showing different characteristic in various aspect, while minimize implementation cost. There are so many different cases to select, thus we cannot sure which is the best plan. Furthermore, it is required to consider changing vegetation structure over time. Thus, we applied a memetic algorithms to this problem because it can improve the performance and reduce the computational time by using reference points in optimizing process. Therefore, in this study, we developed planning model that determines sites for greening and its appropriate greenery type to maximize benefit while minimizing cost with vegetation dynamic in certain period, using memetic optimization approach, Shuffled Frog Leaping Algorithms (SFLA). And then, proposed model is applied to the Seo-cho, Seoul in Republic of Korea. In this problem, we considered relative importance in ecological and visual aspects as benefit of greenery, and also considered implementing costs including planting, land purchasing, and management costs
Results/Conclusions
As a results, we simulated a range of greening plans maximizing benefits of greenery at the each cost level. We find that all plans show reasonable composition of vegetation types from a target period and users’ own criteria perspectives (ecological and visual perspectives). To achieve sustainable greening plans, decisions on location and vegetation type should be connected with the spatial-temporal dynamics of benefits and costs. Recently, some authors suggested optimal locations for GI components in regional scale based on quantitative methodologies, however, temporal dynamics have not yet been considered. Incorporating time scale into the spatial optimization problem is complex, but its effectiveness and efficiency can be enhanced by employing SFLA. Also we expected that this model can support the decision process for the urban greening or co-design between various stakeholders by suggesting detailed options (a kind of map) with performances (benefit and cost).
Saneinejad, S., Moonen, P., Carmeliet, J.(2014). Comparative Assessment of Various Heats Island Mitigation Measures, Building and Environment, 73, 162-170.