Use of Mobile Data in Indonesia UNSD conference data.pdf

This presentation use mobile phone data to produce tourism and commuting data statistics in Indoneisa

1. China, 09 September 2021
The Use of Mobile Phone Data for
Tourism and Commuting Statistics
Titi Kanti Lestari
2
nd
International Seminar on Big Data
STATISTICS INDONESIA

2. Complementforimmigrationdata(administrativedata)
Inboundtourism,whenthereisnoimmigrationcheckpoint,
destinationanalysis
Outboundtourism,toobtaincountryofdestinationandlengthof
stayineachcountry
2
Complement for other data sources
Replacement for other data sources
Domestictourism,replacementofhouseholdsurvey
Commuting,replacementofhouseholdsurvey
Inboundtourism,replacementofshuttletrade(crossborder)
survey
Eventanalysis,replacementofsurveyorticketsales
Mobile Phone Data can be used as

3. Use Cases in Tourism and Commuting Statistics
InboundTourists(numberoftourists,lengthofstay,
placevisited)
OutboundTourists(numberoftourists,lengthofstay,
countryvisited)
DomesticTourists(numberoftourists,lengthofstay,
placevisited,O-Dmatrix)
EventAnalysis(numberofvisitors,venuevisited)
Commuting(numberofcommutes,O-Dmatrix)
3

4. 4
Tourism Statistics Manuals

5. MP
D
IMMIGRATION
How do we implement the tourism concept to mobile phone data

6. Capturemoredata(verybig,especiallyfordomestictourism)
Goodfortourismstatisticsandcommuting
Addnoises(statisticalandnonstatistical)
6
Signalling (probe)
Call Detail Record (CDR)
Lessdata
Possibleundercoverage,especiallyforinboundandoutbound
What data is used

7. 7
Signalling vsCDRNatuna island
Malaka on the Timor
Leste border
Anambas islands
Kupang on the Timor
Leste border
Talaud islands
Sangihe islands
Boven Digoel near Papua
1
10
100
1000
1 10 100 1 000 10 000 100 000
N
u
m
b
e
r
o
f
t
r
ip
s
f
r
o
m
s
ig
n
a
llin
g
d
a
t
a
(
m
u
lt
ip
lie
r
r
a
t
io
)
Number of trips from CDR data
Signalling contributes most in hard-to-reach areas
More trips
Even more trips from signaling data
1.Islands
2.Border to less
developed countries

8. Fastfliers
Seamen
AccidentalRoamers
Othertransit
8
Statistical and Non Statistical Noises
Methodology is important
Filteringmethod

9. 9
MPD for Official Tourism Statistics
Press Release
https://www.bps.go.id/pressrelease.html

10. 10
MPD for Official Tourism Statistics
Publications

11. 11
Quality Assurance
1
Sound Methodology
2
Privacy-Preserving Processing
3
For Official Statistics
•In line with UN-QAF, Big Data QAF, NSO’s QAF
•Various methodologies
•Choose that reflect reality
•Privacy protected
•Aggregate data

12. 12
Quality Assurance
In-linewithBPSQAFHandbook(forCensus,SurveyandAdministrativedata)
In-linewithUNQAFandUneceQAFforBigData
Qualitycheck(Input,Throughput,Output)
Datagaps
Missingdata
Incorrecttimestamps
Duplicaterecord
Errorsindataprocessing
Overwrites
Quality Assurance
Input Quality Checking
(First gate)
Throughput Quality
Checking (Second gate)
Anomalieschecking
Coherencewithotherdata
Newphenomenacanbeexplained
PassedCalibration/Comparisonwithotherdata
Output Quality Checking
(Third gate)

13. 13
Some Quality Assurance Results
Natural hourly rhythm
Steady data flow Logical daily present
Logical daily present

14. 14
Usual Environment
Outsideusualenvironment,
tourist
Home-work,commute
Changinghome,overayear,
internalmigration

15. Thesubscribersismaskedwithhash,whendata
scientistsprocessed
Thedataproduceisaggregatedata(tables)
15
Privacy Protected through Pseudonymization
and k-Anonymity

16. 16
Data Architecture of Sandboxing
ForMobilePositioningData

17. 17
MPD vsConventional Survey Result (at Jakarta Greater Area)
MPD
(2019)
Commuter Survey
(2019)

18. 18
Commuting Statistics

19. +
Central City
Threshold5 percent
Threshold1,5 percent
Presidential Regulation No. 4 of
2018
MPD for MA Delineation
on Cekungan Bandung
Scenario1 Scenario2 Scenario3

20. Contribution of Tourism in Indonesia, 2016 -2019
7,10
4,63 4,65
7,08
4,67 4,67
7,00
4,90 4,91
7,15
4,97 4,97
0,00
1,00
2,00
3,00
4,00
5,00
6,00
7,00
8,00
GVATI TDGVA TDGDP
2016 2017 2018 2019

21. 21
MPD for Tourism Satellite Account
SDGGoal8(Indicator8.9.1)
MPDgivebettercoveragethanhousehold
survey,bettermatchwithsupplyside

22. THANK
YOU
Credit to: