Integrated Transport

Abstract

Urban rail transit system in China has been rapidly constructed in response to the effects of urbanization, such as severe urban congestion and excessive air pollution. The sustainable land use planning (i.e. transit-oriented development, TOD) around the subway stations is important for the rail transit system because of its long-term influence on travel demand. However, there are limited studies that focus on the station level TOD planning. In this context, the aim of this study is to propose a multi-objective pro- gramming model that integrates transport and land use design for station-level TOD planning. In this study, one subway station in Beijing City is taken as the case, considering the unique features of urban development (e.g. high density and diversity), five objectives are taken to account in our model, including rail transit ridership, compactness, accessibility, conflict degree, and environmental effects. Meanwhile, an improved immune-genetic based algorithm is designed to obtain the optimal solutions under alternative land use schemes. The model results show that the proposed algorithm is superior to conventional genetic algorithms. This study is hoped to provide sustainable station-level planning for urban planning decision-makers.

Introduction

If properly designed, mass rail transit (MRT) systems can significantly enhance urban transportation service by attracting passengers from other more expensive modes of transportation and reducing their dependence on private vehicles. MRT systems have been identified as an efficient way to ease urban traffic congestion and energy consumption, reduce the environmental effects of transportation, and promote economic growth (Knowles, 1996; Bhattacharjee and Goetz, 2012; Huang et al., 2016). Numerous megacities worldwide have begun establishing MRT systems, particularly in fast-growing developing nations. In main- land China, 95 rail transit lines were in operation in 22 cities by 2014. The Beijing subway system included 18 lines and 334 stations with 554 km of MRT network by 2015 (Ma et al., 2013, 2017a, 2017b). Numerous cities have plans of expanding and upgrading their existing MRT systems, and a few cities have plans of building new systems. Notably, the performance of an MRT system in terms of coping with rapid urbanization and the increasing number of private vehicles is largely dependent on the land use layout in the serviced areas and the efficiency of relevant resource allocations (Lin and Gau, 2006). Therefore, rather than separately consider MRT systems and relevant land use, a good planning strategy should integrate both aspects into the transit-oriented develop- ment (TOD) aspect of the urban spatial structure planning (Ding et al., 2017a). TOD devises urban development plans based on public transportation systems to enhance the sustainability of MRT systems, land use efficiency, and traffic operation effectiveness.

Numerous studies have explored TOD theories and strategies for region- and station-level planning. The planning strategies of TOD generally include raising transit ridership, diversifying land use types, and increasing the use of mass transit (Bernick and Cervero, 1997). In the past decade, built environment has been considered a critical factor in TOD planning to improve MRT sustainability by combining economic and social factors (Chiou et al., 2013; Cruz and Katz-Gerro, 2016; Wey et al., 2016). Several detailed TOD planning methods have been explored according to their relevant strategies and principles, such as analytical mathematical models, to assist planners. The purpose of establishing analytical TOD models is to obtain alternative schemes of land use plans around subway sta- tions, including locations, types, and area sizes. Several studies have established different TOD models for land use planning (Lin and Gau, 2006; Lin and Li, 2008). Specifically, a China TOD model was firstly established for China cities by Li et al. (2010) and was tested using the Shenzhen MRT station as a case study. However, despite these preliminary studies, the TOD model is still insufficient and requires further exploration. First, the TOD models in the current literature can be modified and extended into a station-level TOD, especially when integrating land use and transport. Second, the solution algorithm of the model can be improved to generate better results. Third, few studies have focused on the application of the TOD model for cities in China. Although the TOD model of China has been established by Li et al. (2010), the model should still be adjusted and modified for application in other Chinese cities to accommodate high population density and high land use density.

In the context of China’s urbanization, this paper presents a station-level TOD planning model with five objective functions: maximizing rail transit ridership, compactness, and accessibility, and minimizing conflict degree and environmental effects. The primary contributions of this study are three-fold: (1) The multi- objective programming model integrates both land use and trans- portation factors, while previous studies commonly neglect the contribution  of  transportation  for  station-level  planning  (Ding et al., 2016). The index of accessibility in this study is used to measure the influence of transportation. (2) Taking a real station area in Beijing (the Nan Shao Station area) as an example, a case study is analyzed to identify feasible solutions for this model. The model proposed in this study extends the application of TOD in China. (3) An immune-genetic based algorithm (IGA) is improved to search for non-dominated solutions for the TOD planning model, producing a set of alternative land use schemes. The results are analyzed and compared with those of the cumulative genetic al- gorithm (CGA) (Xiong and Schneider, 1995; Lin and Feng, 2003) to verify the superiority of the improved IGA in this study in solving the proposed model.

The rest of the paper is organized as follows. Section 2 provides a review of the literature on TOD planning. Section 3 describes the framework of the TOD planning model. Section 4 presents the de- tails of the TOD plan model. Furthermore, the improved IGA is briefly described in terms of the preparation for solving the multi- objective TOD planning model. In Section 6, a case study is analyzed to verify the effectiveness of the model. The conclusion is provided in the last section.

Conclusions

In this study, we developed a TOD planning model to provide references for the rapid development of the rail transit in China.

Based on the conditions of China, this TOD model was established on a multi-objective program model, which considers five objec- tives and two decision variables. The model not only focuses on the efficiency of the land use, such as the ridership and compactness, but also the living equality and environment problems in the planned area. Based on the previous method for solving the TOD model, an improved IGA approach was applied to generate alter- native non-dominated solutions for the proposed model. The model can output alternative land use plan schemes with  the layout and the land use density around the studied subway station. Meanwhile, a case study was used in this study to test the effec- tiveness of the proposed TOD model and solution method. This case study verified that the proposed multi-objective program model is applicable to Chinese megacities. Urban decision-makers can select alternative schemes from this model based on their preferences.

In future work, there are some directions that should be further extended. First, the objective functions can be formulated into nonlinear equations to capture the dynamic and complicated re- lations between different variables. Second, with the development of  bicycle-sharing  systems,  a  subway  station  may  affect  larger regions. The bicycle-sharing system can make residents live little far from the station to take subway. Therefore, the model can be improved to take into account the effect of bicycle-sharing systems. Third, the traffic volumes on roads around the station in the pro- posed model only consider peak-hour volumes. The off-peak traffic volumes should be simultaneously analyzed in the future study. Finally, while the parameters of this model are assumed to be static, this model can be extended to incorporate parameter uncertainties (e.g., with a grey or fuzzy programming model). For using grey or fuzzy methods to establish complex model in future study, it is a challenge work to identify the range or probability distributions of parameters since some data may be difficult to collect in practice. With the application of the big data technology in urban planning, parameters in model can be collected by mining data from different types and sources. For example, the trip generation rate can be estimated by analyzing the smart card data and the mobile sensors data. Therefore, the data-driven method may be a valuable tool for TOD planning in the future work.