Study on Parking Adaptability in Urban Complexes on Top of Subways Based on Shared Parking Spaces

Abstract

Urban complexes on top of subways as a function of intensive building groups, including residential, office, business, and other types nature of land use where parking time differences are obvious, can implement shared parking spaces, thereby reducing the index of parking allotment. Currently, the parking space allocation index for complexes is only a simple superposition of different land uses, resulting in an over-allocation of parking allotments, leading to a waste of land resources and a low utilization rate of parking allotments. Considering the factor of shared parking spaces, this paper conducted an in-depth analysis of the parking adaptability of urban complexes on top of subways and selected five urban complexes on top of subway stations in Wuhan to conduct a parking survey to analyze the parking demand characteristics. This study also investigated the parking behavior of parkers and analyzed the characteristics of parking behavior in urban complexes on top of subways as well as the current parking demand prediction methods and models, establishing a parking demand prediction model based on shared parking spaces and conducting an adaptability analysis. Finally, using five urban complexes in Wuhan as examples, the number of parking spaces demanded by urban complexes on top of subways in 2025 was predicted, and Wuhan Golden Harvest Fashion Plaza was used as an example to verify the feasibility and implementation ability of the theoretical and applied research in this paper.

1. Introduction

With rapid socioeconomic development and the continuous promotion of high-quality development processes, urban automobile ownership continues to increase in major cities across the country, and the “parking” problem has become a “pain point” in many cities [1]. Urban parking allotments play a crucial role in accommodating vehicles, and the reasonable determination of the number of parking allotment allocations can effectively alleviate urban parking problems [2]. Given the increasing scarcity of urban land resources, the government strongly supports the sustainable development of the “rail + property” mode [3].

At present, in the process of formulating the index of parking allotment in different cities in China, factors such as population size, economic development, and differences in the traffic flow generated by and attracted to various functional sites are considered [4]. However, these indexes are based on of the rate of demand for parking generated by sites of a single nature [5]. As urban complexes on top of subways contain a variety of land-use types, the number of parking spaces, if only in accordance with the index of parking allotment, will be a single number of parking spaces needed to add [6]. This approach not only results in a significant disparity between existing parking allotments and actual demand, but it also exacerbates conflicts between parking supply and demand [7]. It will also reduce the overall parking turnover rate, resulting in idle and wasteful parking facilities. Consequently, this paper conducts further research on parking adaptability for urban complexes on top of subways [8].

For the parking characterization study, Lau developed a daytime and nighttime parking demand prediction model through surveys of parking lots, parking characteristics, and driver parking behaviors to predict nighttime and daytime parking demand over a 15-year planning horizon, providing strategic guidance for the planning and management of parking facilities in Hong Kong [9]. Setiawati conducted a five-year parking demand prediction by recording vehicles entering and exiting the parking area on weekdays and holidays, as well as the time of day they are parked, and found that there will be a shortage of 68 parking spaces in the northern parking area and 17 parking spaces in the southern parking area over the next five years [10]. Li found the parking demand characterization indicators to be stable and representative across the range of index fluctuations through simultaneous statistical inference techniques [11]. Ali researched the situation of vehicles entering and exiting the parking lot of Malaysian hospitals, combining the duration of parking, parking occupancy rate, and other indexes, and found that the hospitals are mainly dominated by short-term parking [12]. Parma summarized the on-street and off-street parking characteristics of the business district by conducting a parking survey in the Delhi business district and combining the parking statistics including cumulative numbers of parking vehicles, parking occupancy, parking load level, average parking time, and peak parking index [13]. Chao proposed the concept of latent variable of parking selection, constructed a comprehensive model of SEM-BL for developers and drivers considering the combined effects of explicit and latent variables, and analyzed the influencing factors of the selection of three-dimensional parking facilities, and the results showed that the type of residential area, floor area, and floor area ratio have a key influence on the selection of parking [14]. Liu proposed a two-stage research framework to analyze the parking characteristics and optimize the parking allocation in highway service areas. A three-stage structure consisting of parking lot, parking area, and parking berths was established to systematically analyze the parking characteristics with parking saturation, parking turnover rate, and parking duration as the key indexes [15]. Das selected two roads, Central Road and Shyamaprasad Road in Silchar, where there is a shortage of parking spaces, and conducted a survey to determine the accumulation, occupancy, demand, and supply of parking spaces and modeled the demand for parking using linear regression analysis using the IBM SPSS Statistics 27 [16].

In the study of parking demand predictions, Swanson examined the impact of employment and parking supply on parking demand rates in the CBD and analyzed the relationship between employment and parking supply and parking demand rates [17]. Litman presented the problems with current parking planning and developed a parking management plan with benefits, finding that parking demand could be reduced by 20–40 percent with the implementation of the plan [18]. Jiang established a parking demand prediction model for comprehensive development zones by analyzing the parking choice characteristics of parkers and the geographic location of the project based on the consideration of spatial and temporal shared parking space factors and the traffic conditions of different areas [19]. Sahil examined the demand for parking at sites of different natures, such as flats, hotels, office buildings, and large shopping malls, and established the relationship between the generation of parking demand and the nature of the site by combining the data on parking during morning and evening peak hours at 11 sites of different natures [20]. Zhang compared the characteristics of existing parking demand prediction models then proposed an improved prediction model and used Handan City as an example to prove the feasibility of the improved model [21]. Paidi used three forecasting methods: long-term and short-term memory suitable for short-term forecasting, seasonal autoregressive comprehensive moving average with exogenous variables and integration-based forecasting to make short-term prediction of available parking spaces in open parking lots next to shopping centers [22]. Cheng conducted a cluster analysis of hotel parking demand on weekdays, and introduced an econometric time series model to predict short-term parking demand [23]. Duan discussed the contradiction between insufficient parking supply and rising parking demand and analyzed the related indicators of parking demand. Based on the ARMA model, the short-term parking demand was predicted and the effectiveness of parking demand was verified [24].

In the study of shared parking spaces, Chen studied the daily and weekly change characteristics of parking space usage, proposed the concept and division method of shared parking spaces time windows, determined the constraints, and gave the specific steps, and found that the change characteristics of parking space usage have regularity, stability, and obvious peak and trough periods [25]. Liu studied the shared capacity of parking garages on sites of different natures and in close proximity areas and proposed management measures for shared parking spaces to maximize the use of parking resources [26]. Jiang established a multi-objective optimization model to maximize user benefits and minimize parking search time by implementing shared private unused parking spaces in the region and proposed a parking demand matching scheme for the situation of vacant parking spaces [27]. Wang combined the theory of planning behavior and the benefit-risk perception model as a theoretical framework to explore the intention of shared parking spaces in different areas of cities from the perspective of private parking space owners [28]. Ji proposed a decision model for travelers’ parking choices based on cumulative prospect theory, developed an optimization model for shared parking space allocation that takes parking choice behavior into account, and suggested that managers could encourage travelers to accept shared parking by appropriately lowering parking fees [29]. Zhao proposed nine standard route recommendations for shared parking allocation and parking from four perspectives between shared parking, shared parking spaces, and destinations, and proposed a hierarchical process gray relational analysis method and an improved ant colony algorithm to solve the proposed parking allocation and recommend the optimal parking routes, respectively [30]. Xie proposed a shared parking allocation and guidance optimization framework for self-driving cars, establishing a rolling horizon parking allocation model embedded with an adjustment mechanism, and the results show that the parking allocation model significantly improves the station revenue and parking utilization rate and reduces the travel costs in the shared parking area [31]. Yuan constructed a multi-objective optimization model of parking space allocation and proposed a real-time allocation method of shared parking spaces based on deep reinforcement learning. The simulation results show that the average walking distance of users is reduced and the utilization rate of parking spaces is improved [32].

By analyzing the above literature, the following conclusions can be drawn: First of all, at present, research on shared parking spaces in China is in its infancy, and the focus of the research is mostly limited to the sharing of parking spaces, but the potential impact of peak parking hours on the sharing of parking spaces is neglected. Secondly, parking demand prediction was conducted without consideration of a single parking trip visiting sites of multiple natures, resulting in double counting of parking demand and without consideration of the interaction of sites of different natures. Finally, the research for parking demand prediction considered the shared theory, but this also resulted in the overlap effect under the influence of each factor, so further improvement is needed to explore the influencing factors of the parking demand prediction model.

Based on this, this paper analyzes the parking characteristics of urban complexes on top of subways based on the survey data from the perspective of shared spaces, taking into account time attributes, human subjective attributes, and the influencing factors of shared spaces. It also establishes a parking demand prediction model, verifies the applicability of the prediction model through case analysis, and gives a proposal of parking allotment indexes of urban complexes on top of subways.

2. Characterization of Parking in Urban Complexes on Top of Subways

2.1. Characterization of Parking at Complexes

An important factor in determining whether shared parking spaces can be implemented in urban complexes on top of subways lies in the parking demand time and parking behavior of parkers in various types of sites. A parking survey and a parking behavior survey are conducted for different types of urban complexes on top of subways to provide a data basis for the parking demand prediction model based on shared parking spaces.

2.1.1. Parking Survey for Urban Complexes on Top of Subways

• Method and content of parking survey.
• The parking survey of urban complexes on top of subways is carried out with a focus on two main aspects: parking demand situation and related information collection.
• Parking survey respondents.

Urban complexes on top of subways should meet these requirements: the total floor area of urban complexes on top of subways is more than 50,000 square meters, and the area of secondary functions accounts for more than 30% of the total floor area of the complex. Most of the urban complexes on top of subways in Wuhan contain three types of land use: residence use, office use, and business use, and there are shared parking spaces in these three types of land use, so we take these types of urban complexes on top of subways as a survey sample. Parking lots of the three types of urban complexes on top of subways are interconnected. In this paper, five urban complexes on top of subways in Wuhan are selected as survey samples, as shown in

Table 1. Survey sample.

Site Name	Site Type
	Residence	Office	Business
Chu River and Han Street	Residential Building	Ship International Office Building	CapitaLand 1818 Mall
Zhongjia village station	Min Dong Mansion	Min Dong Business Building	Min Dong International Building
Street crossing station	Qun Guang Shang Yuan	Qunguang Plaza Office Building	Chicony Plaza
Wuchang Railway Station	Residential Building	Tianlun Wanjin Office Building	Tianlun Wanjin Plaza
Yuejiazui Station	Donghu Ruiyuan	Golden Harvest Center Office Building	Golden Harvest Fashion Plaza

.

For the convenience of data statistics, the Chu River and Han Street, Zhongjia village station, Street crossing station, Wuchang Railway Station, and Yuejiazui Station will be numbered 1, 2, 3, 4 and 5 in the following text, and their land-use types will be recorded as Residence 1, Office 1, Business 1 … and so on.

2.1.2. Characterization of Parking Demand in Urban Complexes on Top of Subways

Indicator screening in the evaluation was conducted mainly by eliminating indicators that reflect overlapping information and by eliminating those that have less impact on the evaluation results. Indicator importance analysis consists of passengers and experts participating in the questionnaire survey, scoring to obtain the parameters of each indicator, and then using the critical value method to parameterize the statistics of each indicator to filter out the importance of each indicator.

As can be seen in Figure 1:

• Weekday parking demand for residential sites is concentrated at night, with peak hours concentrated before 7:00 and after 21:00. Throughout the day on non-weekdays, there are frequent fluctuations in the amount of parking demand, suggesting that it is related to the flexible weekend travel of some residents.
• There is a significant morning peak in weekday parking demand for office sites, and non-weekday parking demand is at a low peak throughout the day.
• The peak hours of parking demand for business sites on weekdays are mainly concentrated in the 12:00–14:00 and 17:00–19:00 time periods. On non-weekdays, parking demand gradually climbs from 9:00 a.m., and after 10:00 a.m., parking demand increases at a significantly faster rate, with slightly greater parking demand than on weekdays.

To summarize, the changes in parking demand over time are different for different types of land use, each with its own unique trend in parking demand. The average parking demand rates of the three different types of land uses are compared horizontally, and the length of the shaded area qualitatively characterizes the extent of the parking demand rate, as shown in Figure 2 and Figure 3.

As can be seen from Figure 2 and Figure 3, the parking demand rate of residential land shows an obvious “parabola with opening to the right” during the working day from 7:00 to 18:00, indicating that residential land can provide free parking spaces for other land in this time period. During the period from 8:00 to 17:00 on weekdays, the parking demand rate of office land shows a parabola with an opening to the left, indicating that other land may be needed to provide parking spaces during this period. And during the weekday’s 11:00–19:00 time period, the parking demand rate of business-type land use is lower, indicating that other land uses may be appropriately provided with vacant parking spaces during that time period. During non-weekdays, the parking demand rate for office sites is low, indicating that vacant parking can be provided for other sites, and the parking demand rate for residential sites is high, which may require other sites to provide parking, and the parking demand rate for business sites is high during the 11:00–20:00 time period, which may require other sites to provide parking.

2.1.3. Characterization of Parking Behavior in Urban Complexes on Top of Subways

In order to accurately analyze the parking behavior characteristics of urban complexes on top of subways, it is necessary to carry out a survey assessing the behavioral characteristics of parkers. The survey mainly includes three aspects of behavior: personal attributes, travel characteristics of parkers of different types of land, and parking characteristics. A total of 300 questionnaires were distributed to the parkers of urban complexes on top of subways, and 285 valid questionnaires were cumulatively collected, with an effective rate of 95%.

1.

Analysis of personal attributes

From the questionnaire survey, the personal attribute analysis as shown in Figure 4 is obtained.

The personal attributes of the survey respondents include gender, age, and monthly income, and the results of the survey are shown in Figure 4. According to the survey results, in terms of gender, the number of women accounted for 47% of the total number of survey samples, and the number of men accounted for 53%. Though the overall number of men surveyed is greater than that of women, the phenomenon is in line with the current social ratio of men and women driving. In terms of age, most of the parking people in the urban complexes on top of subways are 31–40 years old, of which the number of people under 40 years old accounted for 73% of the total, and the number of young people in the survey object accounted for a larger number of young people’s groups. In terms of monthly income, the vast majority of the level of income is in the range of 8000–10,000 yuan, of which the number of people whose monthly income is more than 8000 yuan accounted for 79%.

2.

Analysis of the purpose of travel

From the questionnaire survey, the analysis of travel purpose as shown in Figure 5 and Figure 6 is obtained.

As can be seen from Figure 5 and Figure 6, in terms of trip purposes, single-trip purpose personnel accounted for 66%, two-trip purpose personnel accounted for 25%, three-trip purpose personnel accounted for 7%, and four-trip purpose personnel accounted for only 2%, which proves that parkers of urban complexes on top of subways can accomplish multiple car trip purposes through one parking. Among these trip purposes, parking for going to work accounted for the highest percentage, with a proportion of 35%. This is followed by going home, with a share of 24%. Parking for shopping and official errands accounted for 11 percent and 10 percent, respectively.

3.

Vehicle parking time analysis

Based on the purpose of the survey respondents’ trips, they were categorized into four main groups, namely, office, residence, business, and other. Of these, office includes going to work and official errands; residential includes going home and visiting friends; and business includes dining, shopping, and cultural entertainment.

Through the questionnaire survey, the analysis of vehicle parking time as shown in

Table 2. Distribution of parking hours for sites of different natures.

	0–2	2–4	4–6	6–8	8–24
Business	48	32	5	2	1
Residence	2	5	7	23	56
Office	3	5	16	53	13
Others	3	6	2	2	1

and Figure 7 is obtained.

From Figure 7, it can be seen that users of parking lots of business land uses in urban complexes on top of subways mainly park for a short period of time, and the duration of parking is usually within 4 h. Users of office parking lots usually park for 6–8 h. Users of residential parking lots mainly park for long periods of time, with the length of parking concentrated at more than 8 h. Other site-type parkers typically park for less than 4 h. Compared to residential and office land uses, business land uses have shorter parking durations and higher parking turnover rates.

4.

Vehicle arrival and departure characteristics

From Figure 8, it can be seen that there are differences in the vehicle arrival and departure periods of parking operators for different types of sites in different time periods. The number of vehicle departures is greater than the number of vehicle arrivals in the 6:00–11:00 and 14:00–15:00 time slots, which is suitable for providing short-term shared parking spaces, while the rest of the time slots are not suitable for shared parking spaces. For office parking, the number of vehicle departures is greater than the number of vehicle arrivals from 17:00 to 7:00 the next day, so it is suitable to provide long-time shared parking spaces, and the rest of the day is not suitable for shared parking spaces; for business parking, the number of vehicle departures is greater than the number of vehicle arrivals from 14:00 to 16:00, so it is suitable to provide short-time shared parking spaces, and the number of vehicle departures is greater than the number of vehicle arrivals from 21:00 to 7:00 the next day, so it is suitable to provide long-time shared parking spaces, and the rest of the day is not suitable for shared parking spaces.

3. Parking Demand Prediction and Adaptability Study Based on Shared Parking Spaces

Through a parking survey of different types of land use in five urban complexes on top of subway stations in Wuhan, the parking demand characteristics of residential, office, and business land use on weekdays and non-weekdays were analyzed, and the average parking demand rates of the three types of land use at different times were obtained. Through the analysis of parking behavior, it was determined that there is shared parking in urban complexes on top of subway stations.

This chapter will establish a parking demand prediction model based on shared parking spaces, based on the parking data obtained and various factors affecting shared parking spaces, with the aim of providing a basis for revising the index of parking allotment for urban complexes on top of subway stations.

3.1. Mixed Land Use Parking Demand Prediction Method

The study on the parking demand prediction method for mixed land use mainly focuses on three aspects: land-use characteristics, socioeconomic activities, and vehicle travel characteristics.

The models related to land use are mainly based on the classification of urban construction land and study parking demand through factors such as land-use type and development level. The models mainly include three types: parking demand rate model, land-use type and traffic impact analysis model, and business land model. The models related to socioeconomic activities comprehensively consider the influence of multiple aspects, including the regional population, car ownership, and the average salary of employees, in order to accurately reflect the relationship between socioeconomic indicators and parking demand. The core idea of the parking demand prediction model for vehicle travel is to establish a functional relationship between travel intensity and parking demand based on the degree to which buildings of various functions in the region attract vehicle travel.

3.2. Parking Demand Prediction Model Based on Shared Parking Spaces

3.2.1. Modeling Ideas

By analyzing the applicability of each parking demand prediction model and combining it with the parking characteristics of urban complexes on top of subway stations, the concept of shared parking spaces was introduced, and the factors affecting shared parking spaces were quantitatively analyzed to establish a parking demand prediction model based on shared parking spaces. The modeling approach can be divided into three stages:

In the first stage, the research subjects are evaluated and analyzed to determine whether they meet the specific conditions for implementing shared parking spaces. If they do, the next step can be taken. If they do not, the standard index of parking allotment is used for regular parking allotment.

In the second stage, the geographic location of the area where the urban complexes on top of subway stations are located, the increasing number of automobiles, the proportion of public transportation trips, and the efficiency of shared parking spaces were considered as factors affecting shared parking spaces, and a quantitative analysis of various factors was conducted.

In the third stage, a parking demand prediction model for urban complexes on top of subway stations based on shared parking spaces is established based on the parking demand rate model and various factors affecting shared parking spaces.

Shared parking spaces are parking lots that are shared by multiple land uses with complementary parking demand. To achieve shared parking between different land uses, two basic conditions must be met: the parking time must be different, and the various land uses must share the same parking lot.

3.2.2. Analysis of Influencing Factors

1.

Location Factors

The differences in the functions of urban land lead to uneven economic, transportation, and population distributions which affect the demand for parking. The geographical advantages of a city (location potential) are related to employment, population, and land use. As urbanization has become increasingly important, it has given rise to location economics. In transportation research, location potential interacts with land, economy, and transportation systems to form a positive relationship between travel frequency and location potential.

The location potential map is shown in Figure 9, where accessibility refers to the ease of reaching one place from another. It is mainly related to the regional traffic conditions of the destination and can be quantified according to the road hierarchy structure and service level.

The formula for quantifying the road grade structure is shown in Equation (1).

$$K_i={\displaystyle \frac{n_{\alpha }{k}_{\alpha }}{N}}$$

(1)

$K_i=\mathrm{t}\mathrm{h}\mathrm{e} \mathrm{e}\mathrm{v}\mathrm{a}\mathrm{l}\mathrm{u}\mathrm{a}\mathrm{t}\mathrm{i}\mathrm{o}\mathrm{n} \mathrm{v}\mathrm{a}\mathrm{l}\mathrm{u}\mathrm{e} \mathrm{o}\mathrm{f} \mathrm{t}\mathrm{h}\mathrm{e} \mathrm{r}\mathrm{o}\mathrm{a}\mathrm{d} \mathrm{g}\mathrm{r}\mathrm{a}\mathrm{d}\mathrm{e} \mathrm{s}\mathrm{t}\mathrm{r}\mathrm{u}\mathrm{c}\mathrm{t}\mathrm{u}\mathrm{r}\mathrm{e};$

$n_{\alpha }=\mathrm{t}\mathrm{o}\mathrm{t}\mathrm{a}\mathrm{l} \mathrm{n}\mathrm{u}\mathrm{m}\mathrm{b}\mathrm{e}\mathrm{r} \mathrm{o}\mathrm{f} \mathrm{r}\mathrm{o}\mathrm{a}\mathrm{d} \mathrm{c}\mathrm{l}\mathrm{a}\mathrm{s}\mathrm{s}\mathrm{e}\mathrm{s}\mathsf{\alpha };$

$k_{\alpha }=\mathrm{V}\mathrm{a}\mathrm{l}\mathrm{u}\mathrm{e} \mathrm{f}\mathrm{o}\mathrm{r} \mathrm{R}\mathrm{o}\mathrm{a}\mathrm{d} \mathrm{C}\mathrm{l}\mathrm{a}\mathrm{s}\mathrm{s}\mathrm{i}\mathrm{f}\mathrm{i}\mathrm{c}\mathrm{a}\mathrm{t}\mathrm{i}\mathrm{o}\mathrm{n}\mathsf{\alpha };$

$N=\mathrm{T}\mathrm{o}\mathrm{t}\mathrm{a}\mathrm{l} \mathrm{n}\mathrm{u}\mathrm{m}\mathrm{b}\mathrm{e}\mathrm{r} \mathrm{o}\mathrm{f} \mathrm{r}\mathrm{o}\mathrm{a}\mathrm{d}\mathrm{s} \mathrm{i}\mathrm{n} \mathrm{t}\mathrm{h}\mathrm{e} \mathrm{a}\mathrm{r}\mathrm{e}\mathrm{a}.$

The formula for quantifying the level of service of a road is shown in Equation (2).

$$L=\sum _j{l}_j*F_i/m$$

(2)

$L=\mathrm{R}\mathrm{o}\mathrm{a}\mathrm{d} \mathrm{s}\mathrm{e}\mathrm{r}\mathrm{v}\mathrm{i}\mathrm{c}\mathrm{e} \mathrm{l}\mathrm{e}\mathrm{v}\mathrm{e}\mathrm{l};$

$l_j=\mathrm{L}\mathrm{e}\mathrm{n}\mathrm{g}\mathrm{t}\mathrm{h} \mathrm{o}\mathrm{f} \mathrm{r}\mathrm{o}\mathrm{a}\mathrm{d} \mathrm{j};$

$F_j=\mathrm{L}\mathrm{e}\mathrm{v}\mathrm{e}\mathrm{l} \mathrm{o}\mathrm{f} \mathrm{s}\mathrm{e}\mathrm{r}\mathrm{v}\mathrm{i}\mathrm{c}\mathrm{e} \mathrm{f}\mathrm{o}\mathrm{r} \mathrm{r}\mathrm{o}\mathrm{a}\mathrm{d} \mathrm{j};$

$m=\mathrm{T}\mathrm{o}\mathrm{t}\mathrm{a}\mathrm{l} \mathrm{l}\mathrm{e}\mathrm{n}\mathrm{g}\mathrm{t}\mathrm{h} \mathrm{o}\mathrm{f} \mathrm{r}\mathrm{o}\mathrm{a}\mathrm{d} \mathrm{s}\mathrm{e}\mathrm{c}\mathrm{t}\mathrm{i}\mathrm{o}\mathrm{n}\mathrm{s} \mathrm{i}\mathrm{n} \mathrm{t}\mathrm{h}\mathrm{e} \mathrm{a}\mathrm{r}\mathrm{e}\mathrm{a}.$

Based on the above quantitative analysis of road levels and service levels, the accessibility of urban complexes on top of subway stations can be calculated, as shown in Equation (3).

$$A_i=K_i*L$$

(3)

$A_i=\mathrm{U}\mathrm{r}\mathrm{b}\mathrm{a}\mathrm{n} \mathrm{c}\mathrm{o}\mathrm{m}\mathrm{p}\mathrm{l}\mathrm{e}\mathrm{x}\mathrm{e}\mathrm{s} \mathrm{o}\mathrm{n} \mathrm{t}\mathrm{o}\mathrm{p} \mathrm{o}\mathrm{f} \mathrm{s}\mathrm{u}\mathrm{b}\mathrm{w}\mathrm{a}\mathrm{y} \mathrm{t}\mathrm{r}\mathrm{a}\mathrm{n}\mathrm{s}\mathrm{p}\mathrm{o}\mathrm{r}\mathrm{t}\mathrm{a}\mathrm{t}\mathrm{i}\mathrm{o}\mathrm{n} \mathrm{a}\mathrm{c}\mathrm{c}\mathrm{e}\mathrm{s}\mathrm{s}\mathrm{i}\mathrm{b}\mathrm{i}\mathrm{l}\mathrm{i}\mathrm{t}\mathrm{y};$

$K_i=\mathrm{E}\mathrm{v}\mathrm{a}\mathrm{l}\mathrm{u}\mathrm{a}\mathrm{t}\mathrm{i}\mathrm{o}\mathrm{n} \mathrm{o}\mathrm{f} \mathrm{t}\mathrm{h}\mathrm{e} \mathrm{r}\mathrm{o}\mathrm{a}\mathrm{d} \mathrm{h}\mathrm{i}\mathrm{e}\mathrm{r}\mathrm{a}\mathrm{r}\mathrm{c}\mathrm{h}\mathrm{y} \mathrm{s}\mathrm{t}\mathrm{r}\mathrm{u}\mathrm{c}\mathrm{t}\mathrm{u}\mathrm{r}\mathrm{e};$

$L=\mathrm{R}\mathrm{o}\mathrm{a}\mathrm{d} \mathrm{s}\mathrm{e}\mathrm{r}\mathrm{v}\mathrm{i}\mathrm{c}\mathrm{e} \mathrm{l}\mathrm{e}\mathrm{v}\mathrm{e}\mathrm{l}.$

The comprehensive agglomeration scale refers to the comprehensive agglomeration degree of land use in an area. The calculation formula is shown in Equation (4).

$$M=qs$$

(4)

$M=\mathrm{C}\mathrm{o}\mathrm{m}\mathrm{p}\mathrm{r}\mathrm{e}\mathrm{h}\mathrm{e}\mathrm{n}\mathrm{s}\mathrm{i}\mathrm{v}\mathrm{e} \mathrm{a}\mathrm{g}\mathrm{g}\mathrm{r}\mathrm{e}\mathrm{g}\mathrm{a}\mathrm{t}\mathrm{i}\mathrm{o}\mathrm{n} \mathrm{s}\mathrm{c}\mathrm{a}\mathrm{l}\mathrm{e} \mathrm{f}\mathrm{a}\mathrm{c}\mathrm{t}\mathrm{o}\mathrm{r};$

$q=\mathrm{A}\mathrm{g}\mathrm{g}\mathrm{r}\mathrm{e}\mathrm{g}\mathrm{a}\mathrm{t}\mathrm{i}\mathrm{o}\mathrm{n} \mathrm{s}\mathrm{c}\mathrm{a}\mathrm{l}\mathrm{e} \mathrm{m}\mathrm{a}\mathrm{s}\mathrm{s} \mathrm{f}\mathrm{a}\mathrm{c}\mathrm{t}\mathrm{o}\mathrm{r};$

$s=\mathrm{A}\mathrm{g}\mathrm{g}\mathrm{r}\mathrm{e}\mathrm{g}\mathrm{a}\mathrm{t}\mathrm{i}\mathrm{o}\mathrm{n} \mathrm{s}\mathrm{c}\mathrm{a}\mathrm{l}\mathrm{e} \mathrm{f}\mathrm{a}\mathrm{c}\mathrm{t}\mathrm{o}\mathrm{r}.$

Therefore, the location potential model can comprehensively evaluate the location advantages of an area, specifically considering the transportation conditions and economic development conditions. Therefore, the location potential can be quantified by transportation accessibility and comprehensive agglomeration scale, as shown in Equation (5).

$$LP=A^x{M}^{\delta }$$

(5)

$A_i=\mathrm{U}\mathrm{r}\mathrm{b}\mathrm{a}\mathrm{n} \mathrm{c}\mathrm{o}\mathrm{m}\mathrm{p}\mathrm{l}\mathrm{e}\mathrm{x}\mathrm{e}\mathrm{s} \mathrm{o}\mathrm{n} \mathrm{t}\mathrm{o}\mathrm{p} \mathrm{o}\mathrm{f} \mathrm{s}\mathrm{u}\mathrm{b}\mathrm{w}\mathrm{a}\mathrm{y} \mathrm{t}\mathrm{r}\mathrm{a}\mathrm{n}\mathrm{s}\mathrm{p}\mathrm{o}\mathrm{r}\mathrm{t}\mathrm{a}\mathrm{t}\mathrm{i}\mathrm{o}\mathrm{n} \mathrm{a}\mathrm{c}\mathrm{c}\mathrm{e}\mathrm{s}\mathrm{s}\mathrm{i}\mathrm{b}\mathrm{i}\mathrm{l}\mathrm{i}\mathrm{t}\mathrm{y};$

$M=\mathrm{C}\mathrm{o}\mathrm{m}\mathrm{p}\mathrm{r}\mathrm{e}\mathrm{h}\mathrm{e}\mathrm{n}\mathrm{s}\mathrm{i}\mathrm{v}\mathrm{e} \mathrm{a}\mathrm{g}\mathrm{g}\mathrm{r}\mathrm{e}\mathrm{g}\mathrm{a}\mathrm{t}\mathrm{i}\mathrm{o}\mathrm{n} \mathrm{s}\mathrm{c}\mathrm{a}\mathrm{l}\mathrm{e} \mathrm{f}\mathrm{a}\mathrm{c}\mathrm{t}\mathrm{o}\mathrm{r};$

$LP=\mathrm{L}\mathrm{o}\mathrm{c}\mathrm{a}\mathrm{t}\mathrm{i}\mathrm{o}\mathrm{n} \mathrm{p}\mathrm{o}\mathrm{t}\mathrm{e}\mathrm{n}\mathrm{t}\mathrm{i}\mathrm{a}\mathrm{l}$;

$x=\mathrm{A}\mathrm{g}\mathrm{g}\mathrm{r}\mathrm{e}\mathrm{g}\mathrm{a}\mathrm{t}\mathrm{i}\mathrm{o}\mathrm{n} \mathrm{s}\mathrm{c}\mathrm{a}\mathrm{l}\mathrm{e} \mathrm{m}\mathrm{a}\mathrm{s}\mathrm{s} \mathrm{f}\mathrm{a}\mathrm{c}\mathrm{t}\mathrm{o}\mathrm{r};$

$\delta =\mathrm{C}\mathrm{o}\mathrm{m}\mathrm{p}\mathrm{r}\mathrm{e}\mathrm{h}\mathrm{e}\mathrm{n}\mathrm{s}\mathrm{i}\mathrm{v}\mathrm{e} \mathrm{a}\mathrm{g}\mathrm{g}\mathrm{r}\mathrm{e}\mathrm{g}\mathrm{a}\mathrm{t}\mathrm{i}\mathrm{o}\mathrm{n} \mathrm{s}\mathrm{c}\mathrm{a}\mathrm{l}\mathrm{e} \mathrm{f}\mathrm{a}\mathrm{c}\mathrm{t}\mathrm{o}\mathrm{r} \mathrm{c}\mathrm{o}\mathrm{r}\mathrm{r}\mathrm{e}\mathrm{c}\mathrm{t}\mathrm{i}\mathrm{o}\mathrm{n} \mathrm{c}\mathrm{o}\mathrm{e}\mathrm{f}\mathrm{f}\mathrm{i}\mathrm{c}\mathrm{i}\mathrm{e}\mathrm{n}\mathrm{t}.$

Therefore, the location potential model can comprehensively evaluate the location advantages of an area, specifically considering the transportation conditions and economic development conditions. Therefore, the location potential can be quantified by transportation accessibility and comprehensive agglomeration scale, as shown in Equation (6).

$$\lambda ={\displaystyle \frac{{LP}_i}{{LP}_j}}={\left({\displaystyle \frac{K_i{L}_i}{K_j{L}_j}}\right)}^x{\left({\displaystyle \frac{q_i{s}_i}{q_j{s}_j}}\right)}^{\delta }$$

(6)

$\lambda =\mathrm{C}\mathrm{o}\mathrm{r}\mathrm{r}\mathrm{e}\mathrm{c}\mathrm{t}\mathrm{i}\mathrm{o}\mathrm{n} \mathrm{f}\mathrm{a}\mathrm{c}\mathrm{t}\mathrm{o}\mathrm{r};$

$x=\mathrm{A}\mathrm{g}\mathrm{g}\mathrm{r}\mathrm{e}\mathrm{g}\mathrm{a}\mathrm{t}\mathrm{i}\mathrm{o}\mathrm{n} \mathrm{s}\mathrm{c}\mathrm{a}\mathrm{l}\mathrm{e} \mathrm{m}\mathrm{a}\mathrm{s}\mathrm{s} \mathrm{f}\mathrm{a}\mathrm{c}\mathrm{t}\mathrm{o}\mathrm{r};$

$\delta =\mathrm{C}\mathrm{o}\mathrm{m}\mathrm{p}\mathrm{r}\mathrm{e}\mathrm{h}\mathrm{e}\mathrm{n}\mathrm{s}\mathrm{i}\mathrm{v}\mathrm{e} \mathrm{a}\mathrm{g}\mathrm{g}\mathrm{r}\mathrm{e}\mathrm{g}\mathrm{a}\mathrm{t}\mathrm{i}\mathrm{o}\mathrm{n} \mathrm{s}\mathrm{c}\mathrm{a}\mathrm{l}\mathrm{e} \mathrm{f}\mathrm{a}\mathrm{c}\mathrm{t}\mathrm{o}\mathrm{r} \mathrm{c}\mathrm{o}\mathrm{r}\mathrm{r}\mathrm{e}\mathrm{c}\mathrm{t}\mathrm{i}\mathrm{o}\mathrm{n} \mathrm{c}\mathrm{o}\mathrm{e}\mathrm{f}\mathrm{f}\mathrm{i}\mathrm{c}\mathrm{i}\mathrm{e}\mathrm{n}\mathrm{t};$

$LP=\mathrm{L}\mathrm{o}\mathrm{c}\mathrm{a}\mathrm{t}\mathrm{i}\mathrm{o}\mathrm{n} \mathrm{p}\mathrm{o}\mathrm{t}\mathrm{e}\mathrm{n}\mathrm{t}\mathrm{i}\mathrm{a}\mathrm{l}$;

$q=\mathrm{A}\mathrm{g}\mathrm{g}\mathrm{r}\mathrm{e}\mathrm{g}\mathrm{a}\mathrm{t}\mathrm{i}\mathrm{o}\mathrm{n} \mathrm{s}\mathrm{c}\mathrm{a}\mathrm{l}\mathrm{e} \mathrm{m}\mathrm{a}\mathrm{s}\mathrm{s} \mathrm{f}\mathrm{a}\mathrm{c}\mathrm{t}\mathrm{o}\mathrm{r};$

$s=\mathrm{A}\mathrm{g}\mathrm{g}\mathrm{r}\mathrm{e}\mathrm{g}\mathrm{a}\mathrm{t}\mathrm{i}\mathrm{o}\mathrm{n} \mathrm{s}\mathrm{c}\mathrm{a}\mathrm{l}\mathrm{e} \mathrm{f}\mathrm{a}\mathrm{c}\mathrm{t}\mathrm{o}\mathrm{r};$

$K_i=\mathrm{E}\mathrm{v}\mathrm{a}\mathrm{l}\mathrm{u}\mathrm{a}\mathrm{t}\mathrm{i}\mathrm{o}\mathrm{n} \mathrm{o}\mathrm{f} \mathrm{t}\mathrm{h}\mathrm{e} \mathrm{r}\mathrm{o}\mathrm{a}\mathrm{d} \mathrm{h}\mathrm{i}\mathrm{e}\mathrm{r}\mathrm{a}\mathrm{r}\mathrm{c}\mathrm{h}\mathrm{y} \mathrm{s}\mathrm{t}\mathrm{r}\mathrm{u}\mathrm{c}\mathrm{t}\mathrm{u}\mathrm{r}\mathrm{e};$

$L=\mathrm{R}\mathrm{o}\mathrm{a}\mathrm{d} \mathrm{s}\mathrm{e}\mathrm{r}\mathrm{v}\mathrm{i}\mathrm{c}\mathrm{e} \mathrm{l}\mathrm{e}\mathrm{v}\mathrm{e}\mathrm{l}.$

2.

Growth in urban automobile ownership

The demand for parking increases with the level of urban motorization. According to GB/T 51149-2016 “Standard Code for Urban Parking Plan” [33], if the planning population size is greater than or equal to 500,000 people in the city, every increase of a small car requires an additional 1.1–1.3 parking spaces to meet its parking demand. For cities with a planned population size of 500,000 or more, for every additional car in the city, an additional 1.1–1.5 parking spaces are required to meet its parking demand. The increase in urban cars not only directly affects the total demand for parking but also affects the demand for parking during peak hours of the day, as shown in Figure 10.

This paper introduces the growth rate of urban vehicles to modify the model and uses the ratio of the vehicle ownership in the target year to the current vehicle ownership in Wuhan to determine the change coefficient. Historical data on vehicle ownership was obtained by consulting the Wuhan Statistical Yearbook, and the data was fitted using regression analysis.

3.

Proportion of public transport trips

The development of public transportation systems is changing people’s travel habits, causing some travelers to choose not to use private cars as their main means of transportation, thereby reducing parking demand. California, USA, took the lead in promoting the TOD model, and with its high level of service, it successfully attracted a large number of citizens to choose to travel by public transportation, thereby reducing the parking demand for buildings in the area by 12–15%. Similarly, according to parking surveys in Hong Kong, China, it was found that within a 300 m radius of subway stations, parking demand decreased by 5–20%. These cases show that as citizens increasingly choose public transportation as a means of transportation, the demand for parking is effectively reduced, which in turn can reduce the index of parking allotment for buildings.

Therefore, a 300-m radius is set around the urban complexes on top of subway stations, and the number of bus and subway stops that they pass through is counted. A formula for calculating the correction coefficient for the proportion of public transport trips is established, as shown in Formula (7).

$$T=(1-p{)}^k$$

(7)

$T=\mathrm{C}\mathrm{o}\mathrm{r}\mathrm{r}\mathrm{e}\mathrm{c}\mathrm{t}\mathrm{i}\mathrm{o}\mathrm{n} \mathrm{f}\mathrm{a}\mathrm{c}\mathrm{t}\mathrm{o}\mathrm{r} \mathrm{f}\mathrm{o}\mathrm{r} \mathrm{t}\mathrm{h}\mathrm{e} \mathrm{p}\mathrm{r}\mathrm{o}\mathrm{p}\mathrm{o}\mathrm{r}\mathrm{t}\mathrm{i}\mathrm{o}\mathrm{n} \mathrm{o}\mathrm{f} \mathrm{p}\mathrm{u}\mathrm{b}\mathrm{l}\mathrm{i}\mathrm{c} \mathrm{t}\mathrm{r}\mathrm{a}\mathrm{n}\mathrm{s}\mathrm{p}\mathrm{o}\mathrm{r}\mathrm{t} \mathrm{t}\mathrm{r}\mathrm{i}\mathrm{p}\mathrm{s};$

$p=\mathrm{A}\mathrm{v}\mathrm{e}\mathrm{r}\mathrm{a}\mathrm{g}\mathrm{e} \mathrm{a}\mathrm{n}\mathrm{n}\mathrm{u}\mathrm{a}\mathrm{l} \mathrm{g}\mathrm{r}\mathrm{o}\mathrm{w}\mathrm{t}\mathrm{h} \mathrm{r}\mathrm{a}\mathrm{t}\mathrm{e} \mathrm{o}\mathrm{f} \mathrm{p}\mathrm{u}\mathrm{b}\mathrm{l}\mathrm{i}\mathrm{c} \mathrm{t}\mathrm{r}\mathrm{a}\mathrm{n}\mathrm{s}\mathrm{p}\mathrm{o}\mathrm{r}\mathrm{t}\mathrm{a}\mathrm{t}\mathrm{i}\mathrm{o}\mathrm{n} \mathrm{t}\mathrm{r}\mathrm{a}\mathrm{v}\mathrm{e}\mathrm{l} \mathrm{i}\mathrm{n} \mathrm{p}\mathrm{a}\mathrm{s}\mathrm{t} \mathrm{f}\mathrm{i}\mathrm{v}\mathrm{e} \mathrm{y}\mathrm{e}\mathrm{a}\mathrm{r}\mathrm{s}$

$k=\mathrm{N}\mathrm{u}\mathrm{m}\mathrm{b}\mathrm{e}\mathrm{r} \mathrm{o}\mathrm{f} \mathrm{b}\mathrm{u}\mathrm{s} \mathrm{a}\mathrm{n}\mathrm{d} \mathrm{s}\mathrm{u}\mathrm{b}\mathrm{w}\mathrm{a}\mathrm{y} \mathrm{s}\mathrm{t}\mathrm{a}\mathrm{t}\mathrm{i}\mathrm{o}\mathrm{n}\mathrm{s} \mathrm{w}\mathrm{i}\mathrm{t}\mathrm{h}\mathrm{i}\mathrm{n} 300 \mathrm{m}.$

4.

Shared efficiency

Shared efficiency refers to the ability to share parking spaces. Since various types of land use share the same parking lot, the ability to share parking spaces varies due to the characteristics of the parking lot itself, making the size of the shared ability also different. In order to accurately measure shared efficiency, it is divided into subjective shared efficiency and objective shared efficiency.

Subjective shared efficiency refers to the shared capacity generated by the parking choice behavior of the parker. The parker will visit different types of land use when completing a parking behavior, which leads to the completion of multiple travel purposes with only one parking demand. For example, if the main purpose of the parker when visiting the urban complexes on top of subways is to go to work, and at the same time, the parker may walk to a restaurant for a meal. The parker has completed the purpose of dining while going to work and has visited two types of land use at the same time in one parking. However, when calculating the parking demand, the two types of land use will each include the parking demand generated by the traveler in their own parking allotment. This results in the parking demand being counted twice, leading to a deviation between the actual parking demand and the parking allotment. Therefore, the parking demand needs to be adjusted by a non-monopoly coefficient, which reduces the total parking demand of urban complexes on top of subway stations to a certain extent based on the artificial selection factors of the parker. The non-monopoly adjustment coefficient d corresponding to different land uses in urban complexes on top of subways is shown in

Table 3. Non-monopoly adjustment coefficient of various types of land use.

Types of Land Use	Office	Business	Residence
Non-monopoly adjustment coefficient	1.00	0.94	0.97

.

Objective shared efficiency refers to the shared capacity of different land uses due to the different parking demand periods. Because there is a phenomenon of “peak-shifting parking” between different land uses, once the land use is determined, its shared capacity is also determined and will not be affected by the external environment and human choice. The objective shared capacity is quantified by the time-varying coefficient, and the various land uses in the urban complexes on top of subways have different peak parking demand periods. The peak parking demand periods of the complexes are different from those of the single land uses. In order to more accurately predict the parking demand of the complex, the ratio of the parking demand generated by the peak hours of each single-purpose land use to the parking demand generated by the peak hours of the urban complexes on top of subways is defined as the time-varying coefficient. The calculation formula is shown in Formula (8).

$$H_i={\displaystyle \frac{d_i^T}{d_i^{T_i}}}$$

(8)

$H_i=i-\mathrm{c}\mathrm{l}\mathrm{a}\mathrm{s}\mathrm{s} \mathrm{n}\mathrm{a}\mathrm{t}\mathrm{u}\mathrm{r}\mathrm{e} \mathrm{l}\mathrm{a}\mathrm{n}\mathrm{d} \mathrm{u}\mathrm{s}\mathrm{e} \mathrm{t}\mathrm{i}\mathrm{m}\mathrm{e} \mathrm{c}\mathrm{h}\mathrm{a}\mathrm{n}\mathrm{g}\mathrm{e} \mathrm{c}\mathrm{o}\mathrm{e}\mathrm{f}\mathrm{f}\mathrm{i}\mathrm{c}\mathrm{i}\mathrm{e}\mathrm{n}\mathrm{t};$

$d_i^T=\mathrm{T}\mathrm{h}\mathrm{e} \mathrm{p}\mathrm{a}\mathrm{r}\mathrm{k}\mathrm{i}\mathrm{n}\mathrm{g} \mathrm{d}\mathrm{e}\mathrm{m}\mathrm{a}\mathrm{n}\mathrm{d} \mathrm{g}\mathrm{e}\mathrm{n}\mathrm{e}\mathrm{r}\mathrm{a}\mathrm{t}\mathrm{e}\mathrm{d} \mathrm{a}\mathrm{t} \mathrm{a}\mathrm{n}\mathrm{y} \mathrm{g}\mathrm{i}\mathrm{v}\mathrm{e}\mathrm{n} \mathrm{t}\mathrm{i}\mathrm{m}\mathrm{e} \mathrm{b}\mathrm{y} \mathrm{l}\mathrm{a}\mathrm{n}\mathrm{d} \mathrm{u}\mathrm{s}\mathrm{e} \mathrm{c}\mathrm{a}\mathrm{t}\mathrm{e}\mathrm{g}\mathrm{o}\mathrm{r}\mathrm{y} i;$

$T=\mathrm{P}\mathrm{e}\mathrm{a}\mathrm{k} \mathrm{p}\mathrm{a}\mathrm{r}\mathrm{k}\mathrm{i}\mathrm{n}\mathrm{g} \mathrm{t}\mathrm{i}\mathrm{m}\mathrm{e}\mathrm{s} \mathrm{f}\mathrm{o}\mathrm{r} \mathrm{t}\mathrm{h}\mathrm{e} \mathrm{e}\mathrm{n}\mathrm{t}\mathrm{i}\mathrm{r}\mathrm{e} \mathrm{c}\mathrm{o}\mathrm{m}\mathrm{p}\mathrm{l}\mathrm{e}\mathrm{x};$

$T_i=\mathrm{C}\mathrm{l}\mathrm{a}\mathrm{s}\mathrm{s}-i \mathrm{l}\mathrm{a}\mathrm{n}\mathrm{d} \mathrm{u}\mathrm{s}\mathrm{e} \mathrm{P}\mathrm{a}\mathrm{r}\mathrm{k}\mathrm{i}\mathrm{n}\mathrm{g} \mathrm{p}\mathrm{e}\mathrm{a}\mathrm{k} \mathrm{h}\mathrm{o}\mathrm{u}\mathrm{r}\mathrm{s}.$

The shared efficiency correction factor is established using subjective shared efficiency and objective shared efficiency, and the calculation formula is shown in Equation (9).

$$\sigma =d\times H_i$$

(9)

$d=\mathrm{T}\mathrm{h}\mathrm{e} \mathrm{p}\mathrm{a}\mathrm{r}\mathrm{k}\mathrm{i}\mathrm{n}\mathrm{g} \mathrm{d}\mathrm{e}\mathrm{m}\mathrm{a}\mathrm{n}\mathrm{d} \mathrm{g}\mathrm{e}\mathrm{n}\mathrm{e}\mathrm{r}\mathrm{a}\mathrm{t}\mathrm{e}\mathrm{d} \mathrm{a}\mathrm{t} \mathrm{a}\mathrm{n}\mathrm{y} \mathrm{g}\mathrm{i}\mathrm{v}\mathrm{e}\mathrm{n} \mathrm{t}\mathrm{i}\mathrm{m}\mathrm{e} \mathrm{b}\mathrm{y} \mathrm{l}\mathrm{a}\mathrm{n}\mathrm{d} \mathrm{u}\mathrm{s}\mathrm{e} \mathrm{c}\mathrm{a}\mathrm{t}\mathrm{e}\mathrm{g}\mathrm{o}\mathrm{r}\mathrm{y} i;$

$H_i=i-\mathrm{c}\mathrm{l}\mathrm{a}\mathrm{s}\mathrm{s} \mathrm{n}\mathrm{a}\mathrm{t}\mathrm{u}\mathrm{r}\mathrm{e} \mathrm{l}\mathrm{a}\mathrm{n}\mathrm{d} \mathrm{u}\mathrm{s}\mathrm{e} \mathrm{t}\mathrm{i}\mathrm{m}\mathrm{e} \mathrm{c}\mathrm{h}\mathrm{a}\mathrm{n}\mathrm{g}\mathrm{e} \mathrm{c}\mathrm{o}\mathrm{e}\mathrm{f}\mathrm{f}\mathrm{i}\mathrm{c}\mathrm{i}\mathrm{e}\mathrm{n}\mathrm{t}.$

3.2.3. Establishment of a Parking Demand Prediction Model Based on Shared Parking Spaces

In order to accurately predict the number of parking spaces required for urban complexes on top of subway stations in the target year, this paper improves the parking demand rate model. It quantifies the location potential correction coefficient, the urban car growth correction coefficient, the public transport travel ratio correction coefficient, and the shared efficiency correction coefficient. It also establishes a parking demand prediction model for urban complexes on top of subway stations based on shared parking spaces, as shown in Equation (10).

$$D=max\left[\sum _i{a}_{ij}\times R_{ij}\times \lambda \times \chi \times T\times \sigma \right]$$

(10)

$D$ = Parking demand for urban complexes on top of subway (number of parking spaces);

$a_{ij}=\mathrm{P}\mathrm{a}\mathrm{r}\mathrm{k}\mathrm{i}\mathrm{n}\mathrm{g} \mathrm{d}\mathrm{e}\mathrm{m}\mathrm{a}\mathrm{n}\mathrm{d} \mathrm{r}\mathrm{a}\mathrm{t}\mathrm{e}\left({\displaystyle \frac{\mathrm{u}\mathrm{n}\mathrm{i}\mathrm{t}\mathrm{s}}{100}}{\mathrm{m}}^2\right)\mathrm{f}\mathrm{o}\mathrm{r} \mathrm{l}\mathrm{a}\mathrm{n}\mathrm{d} \mathrm{o}\mathrm{f} \mathrm{c}\mathrm{a}\mathrm{t}\mathrm{e}\mathrm{g}\mathrm{o}\mathrm{r}\mathrm{y} \mathrm{i} \mathrm{a}\mathrm{t} \mathrm{t}\mathrm{i}\mathrm{m}\mathrm{e} \mathrm{j};$

$R_{ij}=\mathrm{i} \mathrm{D}\mathrm{i}\mathrm{s}\mathrm{t}\mathrm{r}\mathrm{i}\mathrm{c}\mathrm{t} \mathrm{j} \mathrm{T}\mathrm{y}\mathrm{p}\mathrm{e} \mathrm{o}\mathrm{f} \mathrm{l}\mathrm{a}\mathrm{n}\mathrm{d} \mathrm{u}\mathrm{s}\mathrm{e} \mathrm{B}\mathrm{u}\mathrm{i}\mathrm{l}\mathrm{d}\mathrm{i}\mathrm{n}\mathrm{g} \mathrm{s}\mathrm{c}\mathrm{a}\mathrm{l}\mathrm{e}\left(100 \mathrm{p}\mathrm{e}\mathrm{r} \mathrm{s}\mathrm{q}\mathrm{u}\mathrm{a}\mathrm{r}\mathrm{e} \mathrm{m}\mathrm{e}\mathrm{t}\mathrm{e}\mathrm{r}\right);$

$\lambda =\mathrm{L}\mathrm{o}\mathrm{c}\mathrm{a}\mathrm{t}\mathrm{i}\mathrm{o}\mathrm{n} \mathrm{p}\mathrm{o}\mathrm{t}\mathrm{e}\mathrm{n}\mathrm{t}\mathrm{i}\mathrm{a}\mathrm{l} \mathrm{c}\mathrm{o}\mathrm{r}\mathrm{r}\mathrm{e}\mathrm{c}\mathrm{t}\mathrm{i}\mathrm{o}\mathrm{n} \mathrm{f}\mathrm{a}\mathrm{c}\mathrm{t}\mathrm{o}\mathrm{r};$

$\chi =\mathrm{C}\mathrm{o}\mathrm{r}\mathrm{r}\mathrm{e}\mathrm{c}\mathrm{t}\mathrm{i}\mathrm{o}\mathrm{n} \mathrm{f}\mathrm{a}\mathrm{c}\mathrm{t}\mathrm{o}\mathrm{r} \mathrm{f}\mathrm{o}\mathrm{r} \mathrm{t}\mathrm{h}\mathrm{e} \mathrm{g}\mathrm{r}\mathrm{o}\mathrm{w}\mathrm{t}\mathrm{h} \mathrm{r}\mathrm{a}\mathrm{t}\mathrm{e} \mathrm{o}\mathrm{f} \mathrm{u}\mathrm{r}\mathrm{b}\mathrm{a}\mathrm{n} \mathrm{v}\mathrm{e}\mathrm{h}\mathrm{i}\mathrm{c}\mathrm{l}\mathrm{e}\mathrm{s};$

$T=\mathrm{C}\mathrm{o}\mathrm{r}\mathrm{r}\mathrm{e}\mathrm{c}\mathrm{t}\mathrm{i}\mathrm{o}\mathrm{n} \mathrm{f}\mathrm{a}\mathrm{c}\mathrm{t}\mathrm{o}\mathrm{r} \mathrm{f}\mathrm{o}\mathrm{r} \mathrm{t}\mathrm{h}\mathrm{e} \mathrm{p}\mathrm{r}\mathrm{o}\mathrm{p}\mathrm{o}\mathrm{r}\mathrm{t}\mathrm{i}\mathrm{o}\mathrm{n} \mathrm{o}\mathrm{f} \mathrm{p}\mathrm{u}\mathrm{b}\mathrm{l}\mathrm{i}\mathrm{c} \mathrm{t}\mathrm{r}\mathrm{a}\mathrm{n}\mathrm{s}\mathrm{p}\mathrm{o}\mathrm{r}\mathrm{t} \mathrm{t}\mathrm{r}\mathrm{i}\mathrm{p}\mathrm{s};$

$\sigma =\mathrm{S}\mathrm{h}\mathrm{a}\mathrm{r}\mathrm{e}\mathrm{d} \mathrm{e}\mathrm{f}\mathrm{f}\mathrm{i}\mathrm{c}\mathrm{i}\mathrm{e}\mathrm{n}\mathrm{c}\mathrm{y} \mathrm{c}\mathrm{o}\mathrm{r}\mathrm{r}\mathrm{e}\mathrm{c}\mathrm{t}\mathrm{i}\mathrm{o}\mathrm{n} \mathrm{f}\mathrm{a}\mathrm{c}\mathrm{t}\mathrm{o}\mathrm{r}.$

This model introduces the concept of shared parking spaces by analyzing the parking characteristics of urban complexes on top of subway stations, which makes up for the limitations of directly using the index of parking allotment to calculate the size of parking spaces.

3.3. Analysis of Parking Adaptability of Urban Complexes on Top of Subway

When urban complexes on top of subway stations consider implementing shared parking spaces, it helps reduce the supply of parking spaces and save land resources. Since the parking spaces that can be provided by various types of land use are different due to their own parking characteristics, the number of shared parking spaces between different types of land use will also be affected by their own parking spaces through the sharing of parking resources. In order to more intuitively show the impact of shared parking spaces on the parking demand of urban complexes on top of subway stations, the parking adaptability of urban complexes on top of subway stations is evaluated using the shared parking space utility. By establishing a shared utility model to quantify the degree of sharing, it provides a basis for the study of the parking adaptability of urban complexes on top of subways. The following Formula (11) is the established shared utility model.

$$\theta ={\displaystyle \frac{{\sum }_i{p}_i-D}{{\sum }_i{p}_i}}=1-{\displaystyle \frac{D}{{\sum }_i{p}_i}}$$

(11)

$d=$ Demand for parking after the implementation of shared parking spaces (units);

$\theta =\mathrm{S}\mathrm{h}\mathrm{a}\mathrm{r}\mathrm{e}\mathrm{d} \mathrm{u}\mathrm{t}\mathrm{i}\mathrm{l}\mathrm{i}\mathrm{t}\mathrm{y} \mathrm{c}\mathrm{o}\mathrm{e}\mathrm{f}\mathrm{f}\mathrm{i}\mathrm{c}\mathrm{i}\mathrm{e}\mathrm{n}\mathrm{t} \left(\mathrm{\%}\right);$

$P_i=$ Parking demand (units)calculated according to the index of parking allotment for conventional parking.

The parking demand of different land uses in urban complexes on top of subway stations is complementary in time and space. If the parking demand is calculated using the conventional index of parking allotment, the result will be quite different from the actual parking demand. The parking demand after the implementation of shared parking spaces is more in line with the parking demand of urban complexes on top of subway stations. As can be seen from the above model, the greater the shared utility coefficient, the stronger the complementarity of parking spaces between various land uses, and the higher the efficiency of shared parking spaces for various land uses in urban complexes on top of subway stations. The greater the demand for parking spaces, the more land resources will be saved.

3.4. Example Analysis

Taking Wuhan Jinhe Fashion Plaza as an example, the parking demand prediction model for urban complexes on top of subway stations based on shared parking spaces is applied to calculate the number of parking spaces required for each complex and analyze the degree of shared parking spaces to verify the reliability and accuracy of the model.

3.4.1. Jinhe Fashion Plaza Complex

The Jinhe Fashion Plaza complex is an urban complex on top of the subway in Wuchang District, Wuhan, with land use including residential, office, and business, as shown in Figure 11. The complex is located in the middle of Huanle Avenue, Wuhan Avenue, Yuejiazui Road, and Dongting Road, with convenient transportation. At the same time, the complex is directly connected to the K and J exits of Yuejiazui subway station, making it a well-functioning complex. The complex has a residential floor area of 91,756 square meters, an office floor area of 103,408 square meters, and a business floor area of 34,714 square meters.

3.4.2. Quantitative Analysis of Influencing Factors

1.

Analysis of project location factors

According to the urban road classification standard, different levels of roads correspond to different evaluation values. Urban expressways, as the main traffic arteries, have a value of 4; main roads, as roads connecting important nodes, have a value of 3; secondary roads, as auxiliary roads in the city, have a value of 2; and branch roads, as the smallest details of urban roads, have a value of 1. The Urban Road Traffic Planning and Design defines the structure of the urban road network, stipulating that the planned ratio between expressways, main roads, secondary roads, and branch roads is 1:2:3:8. From Formula (1), we can get the formula $K_{standard}=1.713$.

According to the capacity of the road, the traffic engineering divides the service level of urban roads into six grades, aiming to more accurately reflect the road traffic conditions. The six levels are A, B, C, D, E, and F, with corresponding values of 6, 5, 4, 3, 2, and 1. The service level standards for urban roads of each level are as follows: the service level of expressways should meet level C, the service level of urban main roads and secondary main roads should meet level D, and the service level of urban branch roads is divided into two categories. If there are few vehicles on the road and the traffic is smooth, it corresponds to level B, and the rest is level D. The formula $L_{standard}=6.013$ is obtained from Equation (2).

Based on the above, the road evaluation value of the area where the Jinhe Fashion Plaza complex is located is obtained, as shown in

Table 4. The road evaluation value.

Road Section Name	Length (m)	Road Grade	Service Level of Roads
Xudong street	295	3	3
Zhongbei road	237	3	3
Yuejiazui road	200	1	3
Dongting road	100	1	3

.

Through a quantitative analysis of the road level and service level of the Jinhe Fashion Plaza complex, the accessibility correction factor for the area where the Jinhe Fashion Plaza complex is located is obtained. As shown in

Table 5. Traffic accessibility correction coefficient of the area where Jinhe Fashion Plaza Complex is located.

Type	Evaluating Indicator
Road grade evaluation index	2
Overall service level of road sections	3
Traffic accessibility correction coefficient	0.58

.

In order to determine the location potential correction factor for the area where the Jinhe Fashion Plaza complex is located, in addition to quantifying the accessibility of transportation, it is also necessary to analyze the qualitative and quantitative factors of the scale of the complex. The quantitative factor is calculated by quantifying the size of the population in the area, while the qualitative factor is calculated by quantifying the degree of business activity and land development in the area. The Jinhe Fashion Plaza complex is located in the Xudong Street business district of Wuchang District, Wuhan. Due to its well-developed area and active business activities, it has obvious economic competitiveness. Therefore, when evaluating its agglomeration effect, only the qualitative factor is used as the sole evaluation criterion. The qualitative factor evaluation value is determined using the expert scoring method. The specific results are shown in

Table 6. Regional development.

Description	Very Bad	Bader	Bad	Good	Better	Very Good
Score	1	3	5	7	9	11

and

Table 7. Quality factor correction coefficient.

Types of Land Use	Residence	Office	Business
Qualitative factor	9	9	8

.

Referring to the research experience in Shanghai, the elasticity coefficient is set to $\chi =0.25$,$\delta =0.193$. Considering the development of the project location, the index of the quality factor is set to 6. According to Formula (6), the relative location potential correction coefficients for various types of land use in the Jinhe Fashion Plaza complex are calculated, as shown in

Table 8. Location potential correction coefficient of Jinhe Fashion Plaza.

Types of Land Use	Residence	Office	Business
Relative location potential coefficient	0.94	0.94	0.92

.

2.

Urban car growth rate correction factor

By consulting the Wuhan Statistical Yearbook, data on the number of vehicles and annual growth rates from 2012 to 2022 were collected, as shown in

Table 9. Car ownership in Wuhan in recent 10 years.

Year	Car Ownership (Ten Thousand Vehicles)	Annual Growth Rate (%)
2012	122.52	8.71
2013	135.09	10.26
2014	154.41	14.30
2015	181.65	17.64
2016	213.26	17.40
2017	240.09	12.58
2018	275.01	14.54
2019	312.59	13.67
2020	350.89	12.25
2021	381.19	8.63
2022	406.30	6.59

.

The data in Table 9 was processed using the IBM SPSS Statistics 27 statistical analysis software for regression analysis, and the results of the regression analysis are shown in

Table 10. Regression Index and Parameter Estimation of Car Ownership Model.

Equation	Model Regression Index					Parameter Estimation Value
	R2	F	df1	df2	Sig.	Constant	b1	b2	b3
linear	0.990	863.389	1	9	0.000	−60,558.08	30.149
logarithm	0.990	853.890	1	9	0.000	−462,447.3	60,806.574
inverse	0.989	844.514	1	9	0.000	61,055.073	−122,639,314
quadratic	0.990	873.011	1	9	0.000	−30,154.79		0.007
cube	0.990	882.759	1	9	0.000	−20,020.36			2.47 × 10−6
compound	0.992	1051.656	1	9	0.000	2.04 × 10−109	1.135
power	0.992	1064.431	1	9	0.000	0	255.732
S	0.992	1077.385	1	9	0.000	261.196	−515,835.16
grow	0.992	1051.656	1	9	0.000	−250.267	0.127
index	0.992	1051.656	1	9	0.000	2.04 × 10−109	0.127
Logistic	0.992	1051.656	1	9	0.000	4.85 × 10108	0.881

.

From the data in the above table, it can be seen that the model regression index R-square values of eleven equations, such as linear, logarithmic, etc., are similar, and the F-value of the inverse function is 844.514 as the lowest. Then it indicates that the inverse function has the best fit with the car ownership in Wuhan in the past ten years, so the modeling method of the inverse function is selected to predict the car ownership data in the target year, and the expression of the inverse function is shown in Equation (12).Then it shows that the inverse function has the best fitting effect with the car ownership in Wuhan in the past ten years. Therefore, the modeling method of the inverse function is selected to predict the car ownership data in the target year. The inverse function expression is shown in Equation (12).

$$y=\mathrm{61,055.073}-\mathrm{122,639,314}/x$$

(12)

$x=\mathrm{P}\mathrm{l}\mathrm{a}\mathrm{n}\mathrm{n}\mathrm{i}\mathrm{n}\mathrm{g} \mathrm{y}\mathrm{e}\mathrm{a}\mathrm{r};$

$y=\mathrm{V}\mathrm{e}\mathrm{h}\mathrm{i}\mathrm{c}\mathrm{l}\mathrm{e} \mathrm{o}\mathrm{w}\mathrm{n}\mathrm{e}\mathrm{r}\mathrm{s}\mathrm{h}\mathrm{i}\mathrm{p} \mathrm{i}\mathrm{n} \mathrm{t}\mathrm{h}\mathrm{e} \mathrm{t}\mathrm{a}\mathrm{r}\mathrm{g}\mathrm{e}\mathrm{t} \mathrm{y}\mathrm{e}\mathrm{a}\mathrm{r}.$

This paper takes 2025 as the target year, and the prediction based on the inverse function model is that the number of cars in Wuhan will be 4.9245 million in the target year. The correction coefficient for the degree of urban automobile growth was calculated as 1.21 based on the ratio between the target year’s car ownership in Wuhan and the current car ownership.

3.

Public transportation travel ratio correction factor

By consulting the annual report on transportation in Wuhan, the proportion of residents taking public transportation and the annual growth rate from 2017 to 2021 were collected, as shown in

Table 11. The proportion of public transport trips in Wuhan in recent five years.

Year	Proportion of Public Transport Trips (%)	Annual Growth Rate (%)
2017	15.5	3.82
2018	16.46	6.19
2019	18.42	11.91
2020	18.39	−0.16
2021	21.35	16.10

.

Using IBM SPSS Statistics 27 statistical analysis software to perform a regression analysis on the public transportation travel ratio data, the regression model list is shown in

Table 12. Regression Index and Parameter Estimation of Public Transport Travel Proportion Model Equation.

Equation	Model Regression Index					Parameter Estimation Value
	R2	F	df1	df2	Sig.	Constant	b1	b2	b3
linear	0.921	35.008	1	3	0.01	−2733.873	1.363
logarithm	0.921	34.975	1	3	0.01	−20,924.123	2751.796
inverse	0.921	34.943	1	3	0.01	2769.718	−5,555,667.921
quadratic	0.921	35.04	1	3	0.01	−1357.975	0	0
cube	0.921	35.072	1	3	0.01	−899.343	0	0	1.12 × 10−7
compound	0.935	43.152	1	3	0.007	2.38 × 10−65	1.078
power	0.935	43.123	1	3	0.007	0	151.684
S	0.935	43.095	1	3	0.007	154.566	−306,242.586
grow	0.935	43.152	1	3	0.007	−148.801	0.075
index	0.935	43.152	1	3	0.007	2.38 × 10−65	0.075
Logistic	0.935	43.152	1	3	0.007	4.20 × 1064	0.928

.

From the data in the above table, it can be seen that the model regression indexes of six functional equations such as composite, power, etc. have high and similar R-square values, while the F-value of the S function is the lowest at 43.095, which indicates that the S equation fits better with the proportion of public transportation trips of Wuhan in the past five years. Therefore, the S equation is used to make a prediction of the proportion of public transport trips in the target year. The S equation is expressed as Formula (13).

$$U=e^{154.566-\mathrm{306,242.586}/x}$$

(13)

$U=\mathrm{T}\mathrm{a}\mathrm{r}\mathrm{g}\mathrm{e}\mathrm{t} \mathrm{y}\mathrm{e}\mathrm{a}\mathrm{r} \mathrm{p}\mathrm{u}\mathrm{b}\mathrm{l}\mathrm{i}\mathrm{c} \mathrm{t}\mathrm{r}\mathrm{a}\mathrm{n}\mathrm{s}\mathrm{p}\mathrm{o}\mathrm{r}\mathrm{t} \mathrm{s}\mathrm{h}\mathrm{a}\mathrm{r}\mathrm{e} \mathrm{o}\mathrm{f} \mathrm{t}\mathrm{r}\mathrm{a}\mathrm{v}\mathrm{e}\mathrm{l};$

$x=\mathrm{T}\mathrm{a}\mathrm{r}\mathrm{g}\mathrm{e}\mathrm{t} \mathrm{y}\mathrm{e}\mathrm{a}\mathrm{r}.$

The S-equation can predict that the proportion of public transportation trips in Wuhan in the target year 2025 will be 28.1% and then substitute the results of this data into Equation (7), thus determining the correction factor of the proportion of public transportation trips in the target year to be 0.80.

3.4.3. Parking Allotment Calculation for the Jinhe Fashion Plaza Complex

According to the parking demand survey of the complex, the number of parking demands for different types of land use in the Jinhe Fashion Plaza complex on weekdays and non-weekdays is obtained.

Table 13. Parking demand rate and demand of different properties of Jinhe Fashion Plaza Complex on weekdays.

Time	Demand Rate			Demand
	Residence	Office	Business	Residence	Office	Business
7:00	0.85	0.27	0.1	780	279	35
8:00	0.67	0.7	0.18	615	724	62
9:00	0.52	0.83	0.23	477	858	80
10:00	0.38	0.85	0.25	349	879	87
11:00	0.4	0.85	0.34	367	879	118
12:00	0.48	0.8	0.46	440	827	160
13:00	0.5	0.77	0.43	459	796	149
14:00	0.42	0.79	0.37	385	817	128
15:00	0.39	0.8	0.35	358	827	121
16:00	0.39	0.73	0.36	358	755	125
17:00	0.5	0.45	0.41	459	465	142
18:00	0.73	0.27	0.48	670	279	167
19:00	0.82	0.21	0.5	752	217	174
20:00	0.88	0.13	0.33	807	134	115
21:00	0.9	0.04	0.12	826	41	42

shows the parking demand rate and demand quantity for different types of land use in the Jinhe Fashion Plaza complex on weekdays.

As shown in

Table 14. Parking demand rate and demand of different properties of Jinhe Fashion Plaza Complex on non-weekdays.

Time	Demand Rate			Demand
	Residence	Office	Business	Residence	Office	Business
7:00	0.9	0.12	0.05	826	124	17
8:00	0.85	0.14	0.1	780	145	35
9:00	0.88	0.19	0.18	807	196	62
10:00	0.79	0.2	0.28	725	207	97
11:00	0.77	0.22	0.45	707	227	208
12:00	0.8	0.18	0.6	734	186	219
13:00	0.81	0.2	0.63	743	207	226
14:00	0.76	0.24	0.65	697	248	226
15:00	0.71	0.21	0.65	651	217	236
16:00	0.68	0.16	0.68	596	124	243
17:00	0.65	0.12	0.7	551	114	260
18:00	0.6	0.11	0.75	578	114	285
19:00	0.63	0.11	0.82	642	93	132
20:00	0.7	0.09	0.38	807	21	59
21:00	0.88	0.02	0.17	826	248	285

, the parking demand rate and demand quantity of different properties of Jinhe Fashion Plaza Complex on non-weekdays.

Since the parking demand on weekdays at the Jinhe Fashion Plaza complex is much greater than on non-weekdays, the parking demand during peak hours on weekdays is used for parking demand prediction.

The parking demand of the Jinhe Fashion Plaza complex is predicted through a quantitative analysis of various influencing factors, as shown in

Table 15. Parking demand forecast of Jinhe Fashion Plaza Complex.

Types of Land Use	Residence	Office	Business
Peak hour parking demand	826	879	174
Location factor correction coefficient	0.94	0.94	0.92
Correction coefficient of urban automobile growth degree	1.21	1.21	1.21
Correction coefficient of public transport travel ratio	0.80	0.80	0.80
Parking demand forecast	752	800	155
aggregate	1707

.

Next, the shared efficiency is considered to calculate the parking demand of the Jinhe Fashion Plaza complex. With residential land as the main use, the peak parking demand of the Jinhe Fashion Plaza complex occurs on weekdays between 11:00 and 12:00. From Formula 8, the time variation coefficient is obtained, as shown in

Table 16. Time variation coefficient of different land use types.

Types of Land Use	Residence	Office	Business
Time variation coefficient	1.00	0.94	0.45

.

Combined with the non-monopoly adjustment coefficient to obtain the shared efficiency, and finally the shared parking demand for various types of land use. The shared parking demand for residential land is 730, the shared parking demand for office land is 752, and the shared parking demand for business land is 66. The total shared parking demand for the Jinhe Fashion Plaza complex is 1548. If the conventional index of parking allotment in Wuhan is used, the number of parking spaces in the Jinhe Fashion Plaza complex can be obtained as 2136. Compared to the conventional parking allotment, shared parking spaces can save 588 parking spaces, with a shared utility of 27.5%.

3.4.4. Analysis of the Shared Utility of Urban Complexes on Top of Subway Stations

Combined with the data from the parking demand survey of the other four urban complexes on top of subway stations, the parking demand of the urban complexes on top of subway stations in 2025 was calculated according to the conventional parking allotment index of Wuhan and according to the parking demand prediction model based on shared parking spaces. The parking demand before and after sharing was compared and analyzed to calculate the utility coefficient of sharing. The calculation results are shown in

Table 17. Shared utility of urban complexes on top of subway stations.

Sample Name	Number of Parking in Conventional Facilities	Parking Demand after Berth Sharing	Shared Utility Coefficient
CapitaLand 1818 Mall	2373	1626	31.5%
Min Dong International Building	1644	1126	28.7%
Chicony Plaza	2040	1416	30.6%
Tianlun Wanjin Plaza	1120	835	25.4%

.

Combined with Golden Harvest Fashion Plaza Complex and the remaining 4 urban complexes on top of subways, the parking demand is analyzed by comparing the number of regular allocated parking spaces with the parking space sharing. It is found that the parking demand is reduced to different degrees after the implementation of parking space sharing. Four urban complexes on top of subways reduce the parking demand by 747, 518, 624, and 285, respectively, after the sharing of parking spaces, and their sharing utility coefficients are as low as 25.4% and as high as 31.5%. The shared utility coefficient is 25.4% at the lowest and 31.4% at the highest. It shows that the implementation of parking sharing can reduce the parking allocation index of urban complexes on top of subways, which verifies the reliability of the model of shared parking.

4. Discussion

The research on parking spaces at home and abroad ignores the potential impact of parking peak hours on parking space sharing. Based on three land-use properties, this paper studies the impact of parking peak hours and flat peak hours on parking space sharing, and studies the parking space demand in different time periods. Previous studies did not consider the situation that a parking trip visited many kinds of land. In this paper, demand forecasting research was carried out based on three kinds of land properties: residential, office and, commercial, and five field surveys were conducted for each kind of land property. Although the previous research on parking demand prediction considered the sharing theory, it ignored the coincidence effect under the influence of various factors. Based on berth sharing, this paper studies the adaptability of metro-superstructure urban complexes, selects five metro-superstructure urban complexes in Wuhan to carry out parking surveys, and makes parking demand prediction and adaptability analysis considering location factors, public transport travel ratio, urban car growth rate, and sharing efficiency. In the next step, we can increase the number of samples and conduct berth sharing research in other cities other than Wuhan to make the research results more representative and more widely applicable. In the parking demand forecast based on berth sharing, more influencing factors such as the scale of the complex can be added to improve the accuracy and reliability of the model.

5. Conclusions

With the development of land-use patterns around subway stations in comprehensive development mode and agglomeration mode, urban complexes built on subways often gather a variety of formats such as business, office, and residential, and the development and construction of this mode has also become the mainstream of economic development. However, there are serious problems such as unbalanced parking supply and demand, low utilization rate of berths, and idle berths. Therefore, this paper studies the parking adaptability of the urban complex above the subway, and mainly obtains the following research results:

• Expounding the urban complex above the subway, including urban land classification standards and parking allocation indexes of comprehensive buildings, and analyzing the calculation method of the parking allocation index of the urban complex above the subway, which provides theoretical support for selecting survey samples and establishing a parking demand forecasting model based on berth sharing.
• The parking characteristics of the urban complex above the subway are analyzed and studied. Based on the investigation of parking in the urban complex above the subway, the parking demand characteristics and parking behavior characteristics of residential, office, and commercial land in different periods are analyzed, which provides basic data for the feasibility analysis of berth sharing and the parking demand prediction of the urban complex above the subway.
• Select the parking demand rate model as the basic model, quantify the influencing factors of parking sharing such as location factors, rising urban car numbers, public transport travel ratio, and sharing efficiency, build a parking demand forecasting model based on parking sharing, and analyze its adaptability.
• The parking demand prediction model based on berth sharing is applied to calculate the parking demand of the metro superstructure in 2025, and compared with the number of berths allocated by conventional parking, it is found that the parking demand decreases by 25.4–31.5% after the implementation of berth sharing. The reliability of the model is verified, and it also shows that the parking demand forecasting method based on berth sharing is more suitable for determining the number of parking spaces for urban complexes on top of subways.