Chapter 3 Current Conditions of Curb Parking in Taipei
4.2 Research Data
This paper incorporates different sources of data to acquire both spatial information and parking rates of curb parking. However, these data are highly heterogeneous as they were produced individually. We, therefore, integrate the data residing in different sources by map overlay analysis in ArcGIS. The following paragraph lays out the data utilized in this research by three categories: the spatial information, the data of parking dynamics, and other attributes data such as parking rates and land use zoning.
4.2.1.1 Spatial Information of Curb Parking
Based on our research design, the spatial information of curb parking is threefold: first, the traffic zone, then, the road segment and finally, the exact coordinate of each space.
We employ Taipei Parking Supply and Demand Analysis (臺北市停車供需調查報告) to obtain the information regarding traffic zone and Taipei Curb Parking Map Layer (臺北市路邊停車格位圖) to gain both road segment data and the exact coordinate of
curb spaces. The following shows the details of each source of data.
1. Taipei Parking Supply and Demand Analysis (臺北市停車供需調查報告)
Taipei Parking Supply and Demand Analysis is an annual report carried out by the
Department of Transportation each year that covers half of the Taipei City. During the even-numbered years (i.e., 2012, 2014 and 2016), the survey pays attention to six northern districts — Datung, Zhongshan, Neihu, Songshan, Shilin and Beitou; whereas, in the odd-numbered years (i.e., 2013, 2015 and 2017), the government shifts the focus
‧
to the other six southern districts — Wanhua, Daan, Sinyi, Nangang, Zhongzheng, and Wenshan. However, rather than looking into parking dynamics just by districts, the unit of analysis in this survey is the traffic zone; they are more detailed divisions under a district. Traffic zones reflect local parking patterns because they were constructed on the basis of nearby land-use types, traffic flows, road width, the center of the neighborhood and so forth. After their establishment, each traffic zone is assigned with a type of land use such as the residential area in the city center, the commercial area with night markets on the outskirts and so forth. There were 374 traffic zones in 2015 in the six northern districts; whereas, four traffic zones in Beitou district are missing in the map. While in 2016, the six southern districts comprised 312 traffic zones with no data missing. Therefore, our research focus on those 684 traffic zones with comprehensive information.
Table 4.1 Number of Traffic Zones in Each District
Source: Taipei Parking Supply and Six Southern Districts Six Northern Districts
‧ 國
立 政 治 大 學
‧
N a tio na
l C h engchi U ni ve rs it y
Figure 4.2 Traffic Zones in Each District
Six Southern Districts Six Northern Districts
Traffic Zones
‧
Taipei Curb Parking Map is an open data issued irregularly from the Department of
Transportation of Taipei. We retrieved the map on 29th June 2018, and there were 267,761 curb spaces in the city. The data comprised eight data fields including PKID, PKTYPE, PKTYPE1, ZONE, PKKIND, PKROAD, GEOM, and GEOM4326; see Table 4.2 for more detailed information. Although PKTYPE depicts 23 different types of on-street parking, we focus on curb parking for compact car (小型車)11. Because different modes of transport result from different purposes of travel, embracing all kinds of parking is prone to diverging our attention to land. We, therefore, focus on one of the most common types of curb spaces, the compact car, and attempt to see how pricing may influence the land use efficiency of curb parking.
Table 4.2 Data Fields in Taipei Curb Parking Map Layer
Notation Explanation Notes
PKID Serial Number of Curb Parking
PKTYPE Types of use of curb parking in code Ex: 01-23
PKTYPE1 Types of use of curb parking with explanation Ex: Compact car, mid-sized car, full-sized car, motorcycle, disabled space PKKIND Types of curb parking Ex: Parallel parking, angle parking,
and perpendicular parking.
ZONE The located district Twelve administrative districts in
Taipei
PKROAD The located road segment Ex: Section 5, Civic Boulevard GEOM The coordinate of curb space
GEOM4326 The coordinate of curb space Refer to The World Geodetic System 1984 (WGS84)
‧ 國
立 政 治 大 學
‧
N a tio na
l C h engchi U ni ve rs it y
Table 4.3 briefly summarizes the spatial attributes of curb parking. Until June 2018, there are 50,608 curb spaces for compact car and 2,290 among which are without data of the district attribute. Hence, as shown in Table 4.3, there are only 48,318 curb spaces in total.
In column 2, it is evident that the distribution of curb parking is uneven around the city.
Each district should have approximately 4,027 parking spaces on average to balance between each other. Although most districts range within the allowance of plus and minus 20 percent, Neihu district over-scored the average by more than 95 percent, accounting for 16.3 percent of curb spaces in the city. On the contrary, the number of curb spaces in Datong District takes up merely 3.6 percent, which is less than half of the average. Despite the uneven distribution pattern of curb parking, the average size per curb space is somewhat equivalent (see in column 4). One curb space takes up about 10.21 square meter of land on average.
Figure 4.3 Spatial Distribution of Curb Spaces in Daan District
Curb Spaces Traffic Zones
‧
Table 4.3 The Distribution Area of Curb Parking Spaces
District
Source: Taipei City Department of Transportation.
‧
4.2.1.2 Parking Supply and Demand
Parking Supply and Demand Survey (臺北市停車供需調查報告) conducts two types
of analysis: the parking census on every road segments during weekdays and the sample survey on 5 percent of road segments during weekends. The data used here are from the parking census on 684 traffic zones in Taipei, and we retrieve only those for the compact car and at peak time.
This survey contains the data on parking supply and demand. The data for parking supply can be subdivided into off-street parking and street parking. Under the on-street parking supply, there are two classifications: the marked parking spaces and the unmarked parking spots. Marked parking spaces are the so-called curb spaces that we have extensively discussed; whereas, the unmarked parking spaces specify those available spots along the white12 or yellow13 lane at the curbside.
Figure 4.4 The Types of Parking Spaces Include in the Parking Supply
12 According to Article 183 of the Traffic Sign Markings Installation Regulations (道路交通標誌標線 號誌設置規則), the 15-centimeter white edge lane is the boundary of pavement. Based on Traffic Safety Regulation (道路交通安全規則), automobiles are allowed to parked outside the pavement edge temporarily. All vehicles must be parked in the direction of the flow of traffic with the right-hand wheel parallel and within 40 centimeters to the edge of the roadway.
13 According to Article 168 of the Traffic Sign Markings Installation Regulations, parking on yellow lanes is prohibited from 7:00 am – 8:00 pm daily. In other words, parking on yellow lanes is permitted from 8:00 pm – 7:00 am.
‧ 國
立 政 治 大 學
‧
N a tio na
l C h engchi U ni ve rs it y
The data for parking demand collected in this survey is the number of cars parked in on-street and off-street parking. Same as the parking supply, there are two types of marked or unmarked parking spaces under the on-street classification. However, within the type of unmarked parking spaces, here, we get additional data on the number of cars illegally parked along the curbside. Figure 4.5 shows the items of parking demand that are investigated in this survey.
Figure 4.5 The Types of Parking Spaces Include in the Parking Demand Source: Taipei Parking Supply and Demand Analysis (2016 & 2017)
In addition to the data description, we lay out in-depth discussion on parking data below.
First, we observe a severe shortage of on-street parking even when there is an overall parking surplus in Taipei City. Table 4.4 describes the parking supply and demand for each type of parking spaces. The quantity supplied of parking spaces in Taipei city is 1.08 times (719,395 ÷ 664,278) the quantity demanded (see row 8). Among which, the quantity supplied of off-street parking is 1.28 times (651,718 ÷ 508,015) more than the number of cars parked on them. However, this parking surplus does not alleviate the inefficiency of on-street parking. The quantity demanded of on-street parking is 2.31 times (156,263 ÷ 67,677) more than the current quantity supplied (see
‧
parking at the curbside. It is interesting to think over the fact that how can unmarked parking spaces succeed to accommodate 5.75 times ((84,439 + 28,246) ÷ 19,607) more cars than its original capacity (see row 6). This bizarre phenomenon implies that off-street parking is not a close substitute of on-street parking. The blind replacement of on-street parking by off-street parking is never persuasive enough to tackle the curbside chaos at its roots. A more thorough strategy such as pricing to clear the market should be considered.
Table 4.4 The Overview of Parking Supply and Demand
Parking Supply Parking Demand
Number of
Source: Taipei Parking Supply and Demand Analysis (2016 & 2017)
Furthermore, we discovered that different types of land use generate different patterns of parking behaviors in Table 4.5. Curb parking fees, for instance, are higher in the city center in comparison to the outskirts. The phenomenon of illegal parking is rampant especially in commercial areas with shopping center in that these places happen to see more temporary parking for drop-off and pick-up. For the demand to supply ratio of on-street parking, there is not much difference between curb spaces in the city center or on the outskirts. The average demand to supply ratio is 0.92, and the on-street parking in the center is one percent more likely to be parked.
‧
Table 4.5 The Classification of Land Use Types
Type of land use Curb Parking
Fee (NT$)
Center (Residential Area) 28.36 0.96 39.3
Center (Commercial Area with Shopping Center) 35.50 0.90 44.8 Center (Commercial Area with Night Market) 37.91 1.01 51.8 Center (Mixed Residential and Commercial Area) 34.69 0.96 38.2
Center (Cultural District) 30.71 0.90 15.9
Center (Hospital) 32.26 1.00 37.7
Outskirts (Residential Area) 17.31 0.90 30.4
Outskirts (Commercial Area with Shopping Center) 24.37 0.84 85.1 Outskirts (Commercial Area with Night Market) 31.02 0.89 36.7 Outskirts (Mixed Residential and Commercial Area) 20.73 0.93 33.5
Outskirts (Cultural District) 18.18 0.92 25.6
Outskirts (Hospital) 20.95 0.96 30.4
Outskirts (Industrial Area) 29.41 0.94 39.1
Outskirts (Public Facility) 33.86 0.81 12.3
* Curb parking refers to the on-street “marked” parking spaces.
* DSR is the ratio of parking demand to parking supply.
Sources: Taipei Parking Supply and Demand Survey (2016, 2017); Taipei City On-street Parking Fee (June 2018).
‧
Since the release of Taipei City Ten-Year Parking Plan in 2015, the government has dedicated to extending the coverage of curb parking fees in Taipei City. The data for curb parking fees, thus, change rapidly. This research employs the latest update of on-street parking fee (各行政區路邊汽車收費資訊) on 13th June 2018 and complements some of the missing data with the most recent announcement from Taipei City Parking Management and Development Office. We retrieve the pricing information for 2043 road segments in total. By sorting out which road segment each curb space is located on, we acquire the individual rate of on-street parking. Next, we delineate three main attributes of curb parking fees below: the level of the parking rate, the calculation method, and the billing time of parking fees.
1. Level of Parking Rates and the Calculation Methods
According to Taipei City Parking Management and Development Office, parking fees are set under considerations of area characteristics, the traffic and the billing time.
Taipei city government reviews the effect of implementation from time to time taking the affordability of the public and the parking turnover into account. Based on the pricing information of 2,043 road segments we acquired, we summarize five different levels of parking rates and three different methods for fee calculation. First, these five levels of rates are NT$20, NT$30, NT$40, NT$50, and NT$60; they are calculated in combination with three ways of calculation: per time, per hour or per hour progressively.
Table 4.6 shows that road segments which are charged per hour take the majority at 76.7 percent; and among which, 49 percent of road segments charge the fees of NT$20
‧
an hour. These road segments are those located in the alleys of business areas or on the arteries in residential districts. For alleys in the residential areas, parking fees reflect the daily needs and the affordability of nearby residents; and thus the fee is often charged per time. On the opposite, road segments with parking fees from NT$40 to NT$60, are likely to be the main arteries in business areas with heavy traffic; curb spaces on which are demanding and often occupied. In Table 4.7, we discovered that the share of curb spaces charged per time is surprisingly high in Beitou (86.29 percent) and Wenshan (73.27 percent) districts. With the same reasoning mentioned above, these two districts should be the typical residential districts.
Table 4.6 The Calculation Methods of Parking Fee
Parking Fee Charged per time Charged per hour Total
20 0.20% 37.63% 37.84%
Source: Taipei City Parking Management and Development Office
Note that in Table 4.6, road segments that are charged per hour or per time take up merely 93.47 percent together, because the other 6.53 percent of road segments are unable to classify. Curb spaces on these roads are charged in a combination of different rates and calculation methods. A more explicit explanation of this idea is presented in Table 4.8. Thirteen different combinations of parking fees are carried out in Taipei City.
Due to the variety of rates and calculation methods, how to measure different pricing
‧
segments are charged per hour, we attempt to transform parking rates that are charged per time and per hour progressively into per hour. To carry out these transformations, we need additional information about parking duration of those curb spaces.
For curb spaces that are priced per hour progressively, we refer to a press release14 from the Department of Transportation. After the price reform, the parking duration on twelve hourly-charged road segments in Wenshan district reduced from 6 hours to 2 hours, and in Tianmu neighborhood and Neihu district, the hourly-charged road segments both decrease from 3.5 hours to 1.5 hours. On average, we take 2.5 hours as the parking duration of hourly-charged curb spaces. Therefore, the hourly rate of progressive pricing is the average parking fee charged in 2.5 hours.
However, since the real-time parking sensors only work for road segments charged hourly, there is no official data for curb parking charged per time. Therefore, we take a piece of research on parking behavior in Neihu and Wenshan Districts (Wu and Lin, 2016) as a reference. This paper surveyed on nine road segments in Neihu and seven road segments in Wenshan. We obtain the changes in parking duration for road segments that are charged per time before and after the 2015 price reform. For instance, the average parking duration before the policy implementation is 9.66 hours on surveyed segments in Neihu and 8.9 hours in Wenshan; and after the reform is carried out, the average parking duration declined to 3.46 hours and 5.22 hours respectively.
We failed to know the average parking duration of the entire district as this research
14臺北市全市汽車收費辦理情形說明
https://www.dot.gov.taipei/News_Content.aspx?n=D739A9F6B5C0AB95&sms=72544237BBE4C5F6
&s=0BFD9103EB138003
‧
was just a small sampling, but we get a hint of parking duration between 3 hours to 5 hours generally. Hence in this paper, we adopt the average 4 hours as the parking duration for road segments charged per time. That is, we divide the parking fee which is charged per time by 4 to obtain the hourly rate.
Table 4.7 Different Calculation Methods of Parking Fee in Each District
District Number of
‧
Table 4.8 Different Combinations of Parking Fees Types Parking Fees
A Monday-Friday: NT$50 /time; Saturday, Sunday, and holidays: NT$60 /hour
B
NT$40 for the first hour; NT$50 for the second and the third hour;
NT$ 60 for more than four hours.
C
NT$30 for the first hour; NT$40 for the second and the third hour;
NT$ 50 for more the fourth and fifth hour; NT$60 for more than six hours.
D NT$40 within the first two hours; NT$50 for more than two hours.
E Monday-Friday: NT$30 /hour; Saturday: NT$20 /hour F Monday-Friday: NT$30 /hour; Saturday: NT$40 /hour G Monday-Friday: NT$40 /hour; Saturday: NT$30 /hour
H Monday-Friday: NT$20 /hour (9-17); Saturday: NT$50 /hour (8-18) J NT$30 within the first two hours; NT$40 for more than two hours.
K Monday-Friday: NT$20 /hour; Saturday, Sunday, and holidays: NT$30 /hour L Monday-Friday: NT$30 /hour; Saturday: NT$50 /time
M Monday-Friday: NT$20 /hour; Saturday: NT$50 /time N Monday-Friday: NT$20 /hour; Saturday: NT$30 /hour Source: Taipei City Parking Management and Development Office
‧
On-street parking can be charged in the different days and time periods. Table 4.9 shows five different combinations of day to charge for parking fees. Most road segments (62.6%) are charged six days a week from Monday to Saturday; while, another 35.54 percent of road segments get to pay during the weekdays. Both row 4 and 5 present the road segments that must pay the fee seven days, but there is a slight difference between the two of them. Road segments included in row 4 are charged for the different rates between the weekdays and weekends; whereas, the counterparts in row 5 pay the same fee every day.
Table 4.9 Days to Pay Curb Parking Fees
Day to pay parking fees Percentage of road segments
Monday to Friday 35.54%
Monday to Saturday 62.60%
Monday to Friday; Saturday to Sunday 0.20%
Monday to Sunday 1.62%
Saturday, Sunday and holidays 0.05%
Total 100%
Source: Taipei City Parking Management and Development Office
In addition to the difference in days, Table 4.10 shows a variety of meter hours.
Different meter hours often reflect distinctive neighborhood characteristics. For example, all curb parking in Nangang District have the meter hours starting from 9 a.m.
till 5 p.m. Because during the office hour, Nangang software park brings in significant commuters for business and work; curb spaces demand a higher turnover in response to the heavier traffic. By contrast, after the work from 5 p.m. to 9 a.m., the outflow of
‧
charge for the nearby residents. Take Zhongshan district as another example, there are 45 percent of road segments charged from 7 a.m. to 8 p.m. These segments are mostly the central arteries in downtown such as Zhongshan North Road (中山北路), Minsheng East Road (民生東路), Songjiang Road (松江路) and so forth.
With the analysis above, the time scope of this study focuses on Monday to Friday and from 9 a.m. to 5 p.m. We choose this duration because a large number of curb spaces are charged in this fashion. We, thus, secure the majority of data in the subsequent research.
Table 4.10 Meter Hours of Curb Parking
No. Start from End at Percentage of
Source: Taipei City Parking Management and Development Office
‧ 國
立 政 治 大 學
‧
N a tio na
l C h engchi U ni ve rs it y
4.2.2 Data Processing
This paper employs Taipei Curb Parking Map Layer (臺北市路邊停車格位圖) to explore the potential factors of land use efficiency. To simplify the mode of transport, we keep the curb spaces for the compact car alone; this leaves us with 50,608 curb spaces. Next, we follow the partition established in Taipei Parking Supply and Demand Analysis and produce traffic-zone layer aligning with The World Geodetic System 1984
(WGS84). This map layer contains 684 traffic zone, leaving out four traffic zones which lack location information; it shows the unit of analysis in our study.
After producing the map layer, we combine the spatial information of curb spaces with traffic zones with overlay analysis in ArcGIS. About 5 percent of curb spaces were filtered out in this step because they are not included in any traffic zones. There are 48,016 curb spaces in our final data.
After producing the map layer, we combine the spatial information of curb spaces with traffic zones with overlay analysis in ArcGIS. About 5 percent of curb spaces were filtered out in this step because they are not included in any traffic zones. There are 48,016 curb spaces in our final data.