USING BIG DATA OF AUTOMATED FARE COLLECTION …engineering.nyu.edu/citysmart/trbpaper/15-5262.pdf1...
Transcript of USING BIG DATA OF AUTOMATED FARE COLLECTION …engineering.nyu.edu/citysmart/trbpaper/15-5262.pdf1...
USING BIG DATA OF AUTOMATED FARE COLLECTION SYSTEM FOR ANALYSIS 1
AND IMPROVEMENT OF BRT- BUS RAPID TRANSIT LINE IN ISTANBUL 2
3
4
Ilgin Gokasar, Assistant Professor 5
Bogazici University 6
34342 Bebek/Istanbul Turkiye 7
Tel: +90(212) 359-7278 Fax: +90(212) 287-2457; Email: [email protected] 8
9
Kevser Simsek, Corresponding Author 10
Istanbul Metropolitan Municipality Belbim Inc., Bogazici University 11
Hava Limani Karsisi Istanbul Dunya Ticaret Merkezi A3 Blok Kat:2-3 34149 12
Yesilkoy/ISTANBUL 13
Tel: +90(212) 468-0000 Fax: +90(212) 465-5193; Email: [email protected] 14
15
Kaan Ozbay, Professor 16
New York University 17
1 MetroTech Center, 19th Floor, Brooklyn, NY 11201 18
Tel: +1(646) 997-0552; Email: [email protected] 19
20
21
22
Word count: 4,452 words text + 12 tables/figures x 250 words (each) = 7,452 words 23
24
25
26
27
28
29
Submission Date: 2014-07-3130
TRB 2015 Annual Meeting Paper revised from original submittal.
Gokasar, Simsek, Ozbay 2
ABSTRACT 1
2
Istanbul´s smart card fare collection system generates large amounts of operational data from the 3
BRT-Bus Rapid Transit line. In addition to ridership, it captures system-wide transaction data and 4
provides comprehensive data records on usage. Processing and analysis of these data open new 5
opportunities in transportation and travel behavior research. This paper presents a qualitative 6
analysis of smart card (Istanbulkart) activity for the BRT-Bus Rapid Transit and investigates its 7
potential for understanding complexities of the system and characterizing travel behavior. In this 8
paper, an assessment of spatial and temporal travel behavior of commuters including mode 9
choice, travel, and waiting times, is performed. As a result of this qualitative analysis an 10
evaluation of the automated fare collection system with some recommendations for improving 11
the planning and management of the BRT-Bus Rapid Transit line are also provided. 12
13
14
15
16
Keywords: Smart card (Istanbulkart); Automated fare collection; Transportation planning; BRT-17
Bus Rapid Transit 18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
TRB 2015 Annual Meeting Paper revised from original submittal.
Gokasar, Simsek, Ozbay 3
I INTRODUCTION 1
2
Automated Fare Collection (AFC) - Smart card (Istanbulkart) Technology 3
Since their invention in the 1969 with the improvement of smart card technology, automated fare 4
collection systems have become the most common collection method used by public transit 5
authorities (1). 6
7
Istanbul first used the Akbil (ancient anonymous ticket ) for fare collection in 1994, and then, 8
with the improvement of contactless transport systems, which respond to the requirements of 9
operators and end-users alike, the smart card, appropriately named the Istanbulkart, was put into 10
service in Istanbul’s public transportation system in 2004. The operators of Istanbulkart 11
technology reduced the system’s fraud and maintenance costs by replacing the inefficient paper-12
based fare collection system, and the data collection and reporting capacity has been improved. 13
Today, the Istanbulkart can be used for payment for all modes of public transport. 14
15
Istanbulkart is similar in look and size to a credit card, and each has a unique serial number. A 16
card can be assigned to a specific individual or it can be anonymous, and each unique card ID 17
represents one single person. This enables analysis of individual itineraries and opens new ways 18
for understanding people’s travel behavior on short as well as long term scales. 19
20
In Istanbul, the fare charge for each customer is based on travel distance, transport mode, and 21
certain demographic attributes, such as prioritized rates for elderly people, children, students, 22
senior citizens, government employees, etc. Currently, for example, 37 different media types and 23
69 different fare types are provided by the Istanbulkart (Table 1). 24
25
Fares are collected by an automated reader (Validator) next to the driver or at a turnstile before 26
one boards a transport vehicle. The Validator is a smart device that does validity checks, collects 27
fares in accordance with specified tariffs, and records the result of all transactions. Data collected 28
from the stations are transferred from the stations’ data transfer computers to the automated fare 29
collection server located in the data center by means of an established external network. 30
31
TABLE 1 Istanbul Public Transportation System Ticket Types 32
33
34
35
36
37
38
39
40
41
42
43
44
45
Percentages of fare collection methods for the BRT line are Istanbulkart 95%, Limited Usage 46
Card 3%, and Akbil 2%. (2) The BRT-Bus Rapid Transit pricing system depends on the trip 47
Ticket Types Definition
Fro
m I
stan
bul
AF
C
Akbil Ancient anonymous ticket
Limited Usage Tickets
birGec 1 Journey Ticket
ikiGec 2 Journeys Ticket
besGec 5 Journeys Ticket
onGec 10 Journeys Ticket
Stored Value Tickets
Istanbulkart
Anonymous Anonymous Ticket
Discounted Reduced Fare Tickets
Free Free Entry Tickets
Function Function Cards
TRB 2015 Annual Meeting Paper revised from original submittal.
Gokasar, Simsek, Ozbay 4
distance. When a traveler first uses his or her card, the full price is charged; then, at the 1
destination station, he or she uses the card again, either at a refund machine or on another transit 2
vehicle. There are refund machines at the exits from BRT- Bus stations. These machines 3
recognize the cards of travelers who have used only a portion of the line and credits them with 4
refunds up to 46% of the full fare. First and last usage dates (by mining more than 6,000,000 5
different passenger data per day) were recorded to obtain mean travel time and origin-destination 6
station data. Hence, the trip time of each traveler can be calculated from information taken from 7
the refund machines (Figure 1). The graphics are obtained for approximately 46% of all travelers 8
because only some travelers receive refunds to their Istanbulkarts (3).Thus, besides of the 9
information on boarding time and location, the data collected from Istanbulkart cards contain 10
detailed records of alighting times and destination location for Bus Rapid Transit (BRT) stations. 11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
FIGURE 1 Flow chart to determine BRT- Destination station 32
33
Typically, the date and time of the transaction, status of payment (transfer, acceptance, or 34
refusal), card ID, fare type (student, adult, senior), route ID, and other related data are stored in 35
the Validator and the central server (4). Because these data allow for more detailed assessment of 36
travel behavior and mobility patterns they are invaluable for transit planning, long term planning, 37
and daily management of the transportation system. Massive amount of data are collected and 38
stored, complicating the process of analyzing it. 39
40
In brief, by using a smart card system, transportation authorities have access to: 41
1. personal travel data of millions of people, 42
2. information about each card and/or traveler, 43
3. continuous trip data including refund information, 44
4. identity of user and frequency of use. 45
46
TRB 2015 Annual Meeting Paper revised from original submittal.
Gokasar, Simsek, Ozbay 5
Despite the accumulation of so much information, it is difficult to improve the Istanbulkart’s 1
usability and accessibility for the following reasons: 2
1. Its prevalent purpose is not to monitor performance of the transportation system; hence, 3
additional passenger trip information such as destinations and delays cannot be directly 4
retrieved. 5
2. Each passive data collection method has its disadvantages, and processing usually requires 6
additional knowledge. Interoperating and mining heterogeneous datasets would enhance 7
both the depth and reliability of transportation studies. 8
3. The amount of data obtained is increasing tremendously (approximately 6 million distinct 9
Istanbulkart data every day) and traditional data processing methods might not be equal to 10
the task. 11
12
Such data barriers make the development of a large-scale transportation performance monitoring 13
system cumbersome and slow (5). 14
15
Istanbul BRT –Bus Rapid Transit 16
The Bus Rapid Transit system (BRT) of Istanbul began service in 2007. Istanbul-BRT is a bus-17
based mass transit system, and aims to combine the capacity and speed of light rail with the 18
flexibility, lower cost and simplicity of a bus system. BRT buses operate their journey within a 19
fully dedicated right of way to avoid traffic congestion. In addition, it has an alignment in the 20
centre of the road to avoid typical curb-side delays, and stations with off-board fare collection to 21
reduce boarding and alighting delay related to paying the driver. 22
23
The line connects the Asian and European sides of Istanbul. IT is the only intercontinental BRT 24
system in the world. Thanks to this connection, the duration of the 52 kilometer, 44 stations 25
journey from Beylikduzu on the European side to Sogutlucesme on the Asian side has been 26
reduced to 98 minutes (Figure 2). 27
28
The salient features of BRT-Bus Rapid Transit are 29
1. fast transportation (30 seconds service frequency and average speed of 34.8km/h); 30
2. appropriate for metropolitan area with high population; 31
3. environmentally friendly; 32
4. comfortable alternative transportation. 33
34
35
36
37
38
TRB 2015 Annual Meeting Paper revised from original submittal.
Gokasar, Simsek, Ozbay 6
1
2 3
FIGURE 2 BRT-Bus Rapid Transit Line and Route Map (6). 4
5
The yearly and monthly increase of ridership on the BRT-Bus Rapid Transit is illustrated in 6
Figure 3. The increase is due mainly to improvement of the transfer points and the addition of 7
new lines and stations. 8
9
10 FIGURE 3 Number of BRT Passengers per month. 11
TRB 2015 Annual Meeting Paper revised from original submittal.
Gokasar, Simsek, Ozbay 7
2.LITERATURE REVIEW 1
2
There are several BRT applications similar to Istanbul BRT line in other cities of the world, such 3
as Guangzhou, China, Bogotá, Colombia, Rio de Janeiro, Brazil, Lima, and Peru. (7) There is an 4
increasing number of studies and thus published articles, and papers on smart card data analysis 5
of BRT systems. A comprehensive survey on this topic is offered by CALSTART (2005) reflects 6
what the transit operators are saying regarding the effect of BRT vehicles on ridership. Specific 7
issues addressed in these studies include 1) whether the vehicles are branded and/or styled 8
differently than the communities’ regular buses, 2) whether the BRT vehicles themselves were 9
responsible for changes in ridership levels and 3) the effect of the vehicles on community 10
acceptance of the BRT system. The survey results indicate that ridership levels increased after 11
BRT system implementation, and, in some cases, up to one third of the new riders came were 12
new to transit and an additional third were riding more often. (8) Moreover previous work with 13
focus on Case Studies in Bus Rapid Transit identifies the potential range of bus rapid transit 14
(BRT) applications, and provides planning and implementation guidelines for BRT was released 15
in January 2004 (9). Chu et al. (2010) presented a methodology for characterization of trips 16
based on socio-demographic characteristics, multiday travel patterns and association of travel 17
with specific locations. (10) 18
19
Main motivation of this paper is the analysis of smart card (Istanbulkart) data generated by the 20
BRT-Bus Rapid Transit in Istanbul and the investigation of its potential for understanding 21
complexities of the system and characterizing travel behavior. An assessment of spatial and 22
temporal travel behavior, including mode choice, travel, and waiting times based on smart card is 23
another goal of this paper. 24
25
3. PRESENTATION AND ANALYSIS OF BRT-BUS RAPID TRANSIT DATA 26
27
In this paper, we adopt the definitions used by Istanbul Metropolitan Municipality Department of 28
Transportation, and define a journey as “one-way travel from one activity to another”. Each 29
journey consists of one or more consecutive journey stages or trips on the BRT line or a different 30
transportation mode. To deal with the problems of analyzing this big data, we perform 31
visualization and analysis of transportation performance measures by using Rstudio and 32
Microsoft Excel to highlight connections among heterogeneous transportation data sets, 33
including Istanbulkart data (11). Functions in Rstudio provide a data-rich visualization platform 34
to monitor transit network performance for purposes of planning and operations. 35
36
Single trips of each customer were aggregated to journeys according to fare rules and stations 37
(2014). For further processing of the data, we used MySQL open-source database in combination 38
with MS Excel.csv file format and Rstudio software to extract passenger origin information from 39
the Istanbulkart data. All BRT stations are geocoded using information provided by Department 40
of Transportation. The original aggregated one-day data set contains ca. 800,000 journey records 41
and ca. 460,000 unique card IDs. In the process of preparation and reviewing of these data, about 42
4,000 journeys (1.1%) were removed from the data set because of errors, illogical journey routes, 43
and missing values. 44
45
A variety of methods can be used to characterize and analyze public transport usage on BRT 46
lines. Description in spatial and temporal dimensions are often used by transportation planning 47
TRB 2015 Annual Meeting Paper revised from original submittal.
Gokasar, Simsek, Ozbay 8
authorities as indicators of performance and quality of service. The objective of this section is 1
not to provide comprehensive statistical description of the BRT line system but to show how 2
Istanbulkart data can be specifically used as indicators of region- and culture-specific travel 3
behavior and external factors that influence people’s travel decisions and waiting times. 4
5
In the analysis of times, the time at which people decide to start their travel can provide valuable 6
data. Figure 4 shows arrivals and departures. Red lines show direction to the Asian side, blue 7
lines show direction to the European side. 8
9
10 11
FIGURE 4 Arrivals (gray)/ departures (dark gray) /Asian (red) and European (blue) side 12
per stations (2013). 13
14
In the European side, Zincirlikuyu (Station number 37) is the main transit station that traveler 15
ends its journey or continues towards to the Asian side or European side. When we look at the 16
direction data of this station we observe peak values from Istanbulkart data in 8:00 and 18:00 as 17
travelers are traveling from / to the European and Asian sides. In the night, travelers are mostly 18
passing towards to the European side. There is a sharp increase between 17:00 and 20:00 19
followed with a decrease after 20:00 as shown in Figure 5. 20
21
TRB 2015 Annual Meeting Paper revised from original submittal.
Gokasar, Simsek, Ozbay 9
1 2
FIGURE 5 BRT journeys towards Asian side (red) and European side (blue) (2013) 3
(Zincirlikuyu Main Transit Station). 4
5
Figure 6 shows the temporal distribution of boarding times at the first stop or station of the 6
journey for all BRT line journeys during 24 hours. The sharp peaks in the morning and evening 7
hours are noteworthy. The evening peak shows a high number of journey starts concentrated 8
within a 25-45 minutes period. The fact that many people start their journeys almost 9
synchronously during the morning as well as the evening peak hours suggests that there is not 10
much flexibility in the travel time choice for commuting, probably as a result of commonly 11
inflexible working hours in Istanbul. Additionally, it is worth noting that the small peak around 12
1:00 p.m. in the afternoon might be the result of lunch time travel, part-time workers going to or 13
leaving work, and class hours in educational institutions. 14
15
The representation of trip distribution highlights the main advantages of analysis based on AFC 16
data as opposed to the commonly used self-report surveys. High temporal resolution and 17
reliability of records allow aggregation on a minute by minute basis and thus the detection of 18
sharp peaks, which are clearly visible in Figure 6 and similar to the Figure 5 continental transit 19
values. 20
21
6.604
22.37533.883
20.284 20.77928.367
38.447
70.774
110.919
74.750
40.268
27.11317.364
15.728
54.852
34.59829.167
37.135
49.111
59.305
92.530
135.937
99.223
65.797
47.918
30.029
0
50000
100000
150000
200000
250000
300000
0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23
EuropeanSide
Asian Side
To
tal
of
BR
T T
ran
sit
Tra
vel
ers
Per
Ho
ur
Transit Traveler value per hour
TRB 2015 Annual Meeting Paper revised from original submittal.
Gokasar, Simsek, Ozbay 10
1 2
FIGURE 6 Percent of Average trip start distribution during 24h (2013) 3
4
Identification of Waiting Times 5
Another important temporal measure, which has implications on travel behavior and route 6
choice, is waiting time, i.e., the amount of time between a passenger’s arrival at a station or stop 7
and the boarding of a vehicle. No information on the waiting time of bus passengers can be 8
obtained from Istanbulkart data, since the smart card readers, on which passengers tap their 9
Istanbulkarts as they board a bus, are located inside the bus. However, Istanbulkarts are tapped 10
when a passenger enters and leaves a BRT station, not on board. This means that waiting time is 11
a part of total recorded travel time and can be extracted from Istanbulkart data for BRT. 12
13
In this paper we use the term “waiting time” to describe the amount of time a passenger spends 14
inside a station, which includes time spent walking in the station and actual waiting time. In order 15
to extract this time for each station, all BRT trips from any one station to other stations on the same 16
BRT line are considered. Hence only direct, uninterrupted trips without interchanges are used for 17
the calculation of waiting times. Travel times of passengers traveling between each pair of stations 18
in each direction are extracted. It is assumed that the fastest passenger arrived at the platform just 19
in time and did not have to wait for the bus at all. That passenger’s waiting time is considered to 20
be zero and is used as a benchmark. By subtracting this benchmark time from the travel times of 21
all other customers travelling between the same two stations, waiting times of these passengers are 22
obtained (Figure 7). 23
24
TRB 2015 Annual Meeting Paper revised from original submittal.
Gokasar, Simsek, Ozbay 11
1
2
3
4
FIGURE 7 Average waiting times (in-station times) per stations for West-East BRT Line 5
during 24h (2013) 6
7
Dependent on the focus of the analysis of waiting times at BRT stations, information on 8
significant service frequency changes during the day as well as potentially overcrowded buses 9
can be gained. However, it is important to note that because of the method for defining and 10
identifying waiting time, the absolute time values do not precisely indicate the service frequency 11
at a particular station. Large hub stations with several access points tend to have larger variations 12
and higher averages of waiting times, which reflects different lengths of access routes from 13
different entry points. In Figure 7 the average waiting times per station for Istanbul’s West-East 14
BRT line are shown. As expected, large hub stations with long underground access routes such as 15
Saadetdere, BRT Station 9, show longer waiting times. Notably the longest waiting time is found 16
for Uzuncayir, BRT station 42. This exceptionally high value in comparison with other stations 17
can be explained by high passenger volumes at the first station on the Asian side. Next to the 18
common waiting time distribution around 7 minutes, a second peak of waiting times ranging 19
from 8 to 9 minutes is clearly visible from the Figure. This peak results from passengers who 20
take BRT in the opposite direction towards the last stop at Zincirlikuyu and then stay on board to 21
secure a seat during the morning peak hour. This observation leads to an interesting conclusion: 22
the value of having a seat is exceptionally high and for some people it is worth about 6 minutes 23
of additional travel time. This finding is also an indication of high passenger volumes and long 24
BRT journeys. In the case of journeys from the Zincirlikuyu station, the average distance to the 25
destination station for passengers taking the detour over the Edirnekapi Station is 7 stations 26
longer compared to passengers taking the direct service. 27
28
29
Station Numbers (1-44)
Av
era
ge
wait
ing
tim
e (m
) 9.0
7.0
5.0
3.0
TRB 2015 Annual Meeting Paper revised from original submittal.
Gokasar, Simsek, Ozbay 12
1 2
FIGURE 8 Waiting times at the Uzuncayir BRT Station for journeys starting between 3
7am-10am (Station number 42) 4
5
Analysis of activities for BRT travelling 6
People usually travel from one place to another because they want to conduct activities, such as 7
work, education, leisure, or social intercourse, that cannot be performed in the desired way at 8
their current location. Hence, in order to understand people’s travel behavior in terms of mobility 9
patterns and traffic volumes, it is necessary to look at their daily activities and locations. In the 10
absence of comprehensive data sources such as census data, one of the main challenges in 11
modeling travel behavior results from a lack of verified data of high spatial resolution on 12
people’s home and particularly work locations (12). However such information is important not 13
only for transport planning and implementation of agent-based transport models, but is also 14
invaluable in the areas of urban development and land-use planning. Travel patterns observed in 15
smart card data from public transport can provide important information on people’s primary 16
activity locations and can help to verify and refine existing models and assumptions. 17
18
To describe the activities of a particular person, the recorded daily trip chain of that person must 19
be consistent. Consistency in the context of AFC smart card data means that the person who 20
arrived at an activity location by public transport has to leave at the end of the activity by public 21
transport; otherwise the duration of the activity cannot be extracted. The assumption of 22
consistency based only on AFC data is hard to verify since the use of any means of transport 23
other than public transport, e.g., walking, cannot be detected. However, obvious cases of 24
inconsistency can be detected by analyzing the distances between the alighting location of the 25
last journey and the boarding location of the following journey (Figures 9a and 9b). 26
27
TRB 2015 Annual Meeting Paper revised from original submittal.
Gokasar, Simsek, Ozbay 13
1 2
FIGURE 9a Total Transit BRT Trips 3
4
5 6
FIGURE 9b BRT most common boarding stations. 7
5,46%
63,17%
17,32%
0,70%
13,37%
0175.000350.000525.000700.000875.000
1.050.0001.225.0001.400.0001.575.0001.750.0001.925.0002.100.0002.275.0002.450.0002.625.0002.800.0002.975.0003.150.0003.325.0003.500.000
Bus BRT
Public Buses Private Buses
Land Transport Marine Transport Rail Transport
Transportation types and percentages of BRT transit travelers
Tota
l Tra
nsi
ts o
f B
RT
tra
vele
rs
0
50.000
100.000
150.000
200.000
250.000
BRT Public Buses
Private Buses
MarineTransport
Land Transport Rail Transport
01.Tuyap Congress Center
05.Beylikduzu
12. Avcılar
20. Yenibosna
21.Sirinevler
24. Zeytinburnu
26.Cevizlibag
29. Edirnekapı
36.Mecidiyekoy
37. Zincirlikuyu
38.Bosphorus Bridge
41. Acıbadem
BRT- Most common transit boarding stations
Tota
ltra
vel
ers
per
sta
tio
n
TRB 2015 Annual Meeting Paper revised from original submittal.
Gokasar, Simsek, Ozbay 14
The analysis of distances between Istanbulkart journeys is shown in the form of a cumulative 1
relative frequency graph in Figure 10. Only persons with more than one journey recorded in the 2
one-day Istanbulkart record were evaluated. The graph shows that 90.2% of journeys following a 3
previous journey start less than 2 km away from the previous alighting location. This indicates 4
that the majority of public transport users do not switch to other transport modes between public 5
transport journeys. 6
7
A slightly different picture is obtained by looking at the distances between the first boarding and 8
the last alighting station of the day; if they are nearby they are likely to be in the vicinity of 9
passengers’ home locations. As shown in the cumulative relative frequency graph of these 10
distances for all persons in the Istanbulkart data, including those with one journey (Figure 10), 11
only 73% return to a station within a radius of 1 km from their first departure station of the day. 12
This observation is largely due to the fact that 27% of BRT journeys are one-way journeys. To 13
better understand mode choice and identify transportation modes used in combination with 14
public transport in Istanbul, in the following section we refer to previous annual records of the 15
Istanbul Metropolitan Municipality Land Transportation Authority (IETT) (2010-2014). 16
17
18 19
FIGURE 10 Cumulative and relative frequency graph of distances between journey stages 20
TRB 2015 Annual Meeting Paper revised from original submittal.
Gokasar, Simsek, Ozbay 15
Looking at the distances between starting point and final destination of the day, we observe that 1
more than 92% of all persons return to their starting location at the end of the day. Analysis of 2
the alternative modes of transport used by BRT users reveals that a total of 64.7% of reported 3
journeys are conducted completely on public buses, 7% are on ships, and 29.3% are on rail. We 4
cannot detect other transit modes such as taxi, private car, cycle, foot, or some unknown mode in 5
instances when a new journey starts more than 2 km away from a previous alighting location or 6
more than 2 hours later. Almost all BRT journeys are made as the first or last journey of the day, 7
which indicates the high usage of private buses for travel to and from work or academic locations 8
(Figures 9 and 10). 9
10
Another important indicator of activity type is the regularity and frequency of trips to and from a 11
particular activity area. In order to be able to conduct such analysis, a longer term Istanbulkart 12
record of at least several consecutive days is obtained. This enables the analysis of multi-day 13
travel behavior and identification of principal activity spaces. The regularity of activities 14
performed there would give a strong indication of type. Additionally, analysis of activities can be 15
based on demographic factors and differentiation of work activities for adults and educational 16
activities for students (13). In this way, the accuracy and reliability of estimated home and work 17
locations and their densities can be much improved. 18
19
20
4. CONCLUSION 21
22
Improving the data-mining process while reducing the time needed for data processing has 23
become an important problem for transportation decision makers who now have access to 24
unprecedented amounts of operational data. In this paper, Istanbul´s automated fare collection 25
system and pricing policies, operational sources of big data, have been introduced, and 26
approaches for the use of public transport smart card fare payment data for characterizing and 27
analyzing travel behavior have been presented. Even though examination of this data source 28
alone is not enough to measure the full potential of Istanbulkart data for planning purposes, the 29
processed data used in this study reveal the potential for solving short and long term problems 30
the users of the system. .In this study, we used the Rstudio to analyze this large amount of usage 31
data. Some of the immediate research questions that can lead to the improvement of BRT-Bus 32
Rapid Transit line planning and operations that we were able to identify as a result of the current 33
study can be summarized as follows: 34
1. Deployment of efficient data mining techniques to analyze large data sets for the purpose 35
of discovering patterns of persistent problems; 36
2. Development of fast and easy to implement visualization tools for the analysis of this type 37
of big data; 38
3. Use of the Istanbulkart data to devise data-driven algorithms to generate products for 39
operational and planning purposes including time- dependent OD tables and travel time 40
estimates. 41
42
In this paper, the link between public transport usage and specific activities as a reason for travel 43
was also established. Methods for identifying primary activity locations were described. This is a 44
first step towards an activity location model based on the public journeys record extracted from 45
Istanbul BRT-line data. The disaggregated, multi-day data record, which is becoming accessible 46
for research, will allow expansion and implementation of methods presented in this work. 47
TRB 2015 Annual Meeting Paper revised from original submittal.
Gokasar, Simsek, Ozbay 16
1
Furthermore, models of work locations based on Istanbulkart data can be compared with existing 2
models used by transportation planners. Multi-day data open the door for further analysis of 3
travel patterns and behavior, for the identification of demand profiles for bus routes, stations, and 4
interchanges; and for development of public transport route-choice model. Such future research 5
possibilities add additional value to Istanbulkart data records and contribute to the development 6
of methods for processing and analyzing smart card data records for the purpose of transportation 7
planning. 8
9
In the second stage of this study, an integrated data fusion procedure that models the travel 10
patterns and regularities of transit riders along the BRT-Bus Rapid Transit line will be 11
developed. This procedure will incorporate transit riders' trip chains based on their temporal and 12
spatial characteristics and effectively capture their historical travel patterns. We will examine big 13
data in the rate of streaming data; and “veracity,” the relative certainty of data. (14) Then, 14
through examination of travel patterns and transfer data, rider-level destinations can be estimated 15
from multi-day observations and “agent-based micro-simulations”. (15) In these simulations, 16
travelers and vehicles are modeled through agents that interact with the public transport system 17
according to their individual goals. 18
19
In the future, we will also use numerical simulation models to test pricing strategies . We believe 20
that an improved version of our visualization and analysis methods and use of agent-based 21
micro-simulations can improve the design of demand management systems, including origin-22
destination based schemes, in the realm of public transport. 23
24
5. ACKNOWLEDGEMENT 25 26
The authors thank Istanbul Metropolitan Municipality – Belbim Inc. for supporting this research 27
and providing the data. 28
29
REFERENCES 30
31
1. AmericanHistory,http://sipwebsrch02.si.edu/search?site=americanhistory&clie32
nt=americanhistory&proxystylesheet=americanhistory&output=xml_no_dtd&fi33
lter=0&q=roland+moreno&submit.x=13&submit.y=8&s=SS 34
2. IETT www.iett.gov.tr 35
3. IBB www.ibb.gov.tr 36
4. Blythe, P., 1998. Integrated Ticketing-Smart Cards in Transport. IEE 37
Colloquium: Using ITS in Public Transport and Emergency Services, Paper No. 38
4, 20 p 39
5. Catherine Morency, Martin Trepanier, Bruno Agarda Measuring transit use 40
variability with smart-card data, December 2010 41
6. http://www.metrobusharitasi.com/ 42
7. Data Mining Solutions: Methods and Tools for Solving Real-World Problems 43
(27 July 1998) by Christopher Westphal, Teresa Blaxton 44
8. Global BRT Data http://www.brtdata.org/#/location 45
9. http://www.nbrti.org/docs/pdf/weststart_brt_ridership_analysis_final.pdf 46
10. Recognizing Demand Patterns from Smart Card Data for Agent-Based Micro-47
TRB 2015 Annual Meeting Paper revised from original submittal.
Gokasar, Simsek, Ozbay 17
simulation of Public Transport, Paul Bouman, Milan Lovric, Ting Li, Evelien 1
van der Hurk, Leo Kroon, Peter Vervest, ATT 2012 2
11. Chu, K.K.A. and Chapleau, R. (2010). Augmenting Transit Trip 3
Characterization and Travel Behavior Comprehension. Journal of 4
Transportation Research Board 2183, pp. 29-40. 5
12. Open source and enterprise-ready professional software-http://www.rstudio.com/ 6
13. Use Of Public Transport Smart Card Fare Payment Data For Travel Behaviour 7
Analysis In Singapore: A. Chakirov , A. Erath, 2012 8
14. Bagchi, M. and White, P.R. (2005) The potential of public smart card 9
data.Transport Policy 12, pp.464-474. 10
15. Jang, W. (2010) Travel Time and Transfer Analysis Using Transit Smart Card 11
Data. Journal of Transportation Research Board 2144, pp. 142-149. 12
13
TRB 2015 Annual Meeting Paper revised from original submittal.