Development of a complex network-based integrated multi-layer urban growth and optimisation model for efficient urban traffic network

Nowadays, most cities and urban areas increasingly pay attention to issues relating to urban growth, while studies on the urban traffic network structure is important to increase the efficiency of urban traffic networks. Analysis of the basic properties of the current urban transportation network can be used to optimise the urban design process. However, the analyses that integrate traffic network and land-use in light of their co-evolution process are still very limited, and there is a need to link the rail and the street network analysis and to examine the influence of multi-layer networks on the structures. This study can give a clear indication for the urban traffic network design and growth. The goal of this study is to apply the network analysis through simulation and optimisation methods to obtain a more efficient urban traffic network. The study adopts single-layer and multi-layer network representation methods to capture traffic networks’ basic characteristics, properties and the growth trends. The complexity and impacts of evolving multi-layer network and the co-evolution model are discussed, to address the multi-layer network and land-use based co-evolution process. Network growth methods, robustness methods and some multi-objective methods are used to optimise both the single-layer and multi-layer urban traffic networks. This study proposed the complex network based integrated multi-layer urban growth and optimisation model (CNIMUGOM) to explain the co-evolution process of the land-use, urban traffic network and population, and to obtain an efficient network structures. The complexity of Kuala Lumpur street networks and worldwide rail networks (sampled from 45 cities) are presented to show their basic structural properties to guide the future traffic network design. The Public Urban Rail Transit Networks of Kuala Lumpur (PURTNoKL) is taken as a sample case to capture the network growth process and the change trends of a single-layer network more clearly. The relationships between traffic multi-layer networks and land-uses were discussed, and their co-evolution model is established. The study found that the rail network growth triggers the variation of urban traffic network structure and community partition; the network diameter and the average shortest path length are decreasing dramatically, while the highest decreasing rate of accessibility is around 12%. The Relative Neighbourhood Graph (RNG) and Gabriel Graph (GG) are introduced to represent the multilayer traffic network. Considering the cooperation strength and average operation speed ratio in different layers, the impacts of these variable parameters to the network structures are analysed. The optimisation methods are proposed based on the maximisation of network performance from the growth and robustness. These nodes with the biggest shortest path lengths can be treated as the important and potential nodes for the future development of the single-layer network. Then these methods are expanded to multi-layer networks, and are tested using their coupling networks. For those nodes with biggest cluster centrality and closeness centrality can be treated as important and potential nodes. Then the multi-objective methods based analysis indicated that for a single-layer urban traffic network, the scale-free based networks could support more traffic flow than other coupling networks. When considering the multi-layer urban traffic networks, small-world based networks can support more traffic flow than other forms of network. After considering the traffic network structure and traffic flow constraints, the small-world based multi-layer network co-evolution model is proposed. The findings of the study afford to improve the current land-use and traffic integrated models. Based on the traffic network and urban land-use coevolution process, the proposed CNIMUGOM model can save the traffic network construction investment, reduce the travel cost and make the urban traffic network more efficient. Based on the simulation, the proposed network can increase the network efficiency around 30%, and the total traffic flow amount is decreased around 30%. This study contributes to the current literature of complex network theory by gaining additional insights on multi-layer networks related studies and the modelling applications. With the use of the network growth measurement method, the multi-layer network co-evolution model, the network optimisation method and CNIMUGOM, urban planners and designers can provide a better design network structures and optimise the urban traffic networks for rail and road networks efficiency