I know researchers in Europe who have gotten very rich data allowing tracking of individual cellphones throughout the day without any disclosure of the actual owner. Each time the cellphone is turned on it is assigned an ID number which is not the cellphone number. This location can be tracked because they must know which cell tower it is connected with at all times. The location is reported back to the company whenever a call is made or received or a text message is sent or received. If it is on, but not being used, it reports every two hours to keep the cell tower location active. By seeing where the phone is for many hours overnight the home location can be assumed. Then, when it moves in the morning and then is steady at one location, that is assumed to be the work location. However, there is no guarantee this is work, it may be shopping, school, or other.
Of course, when using a "smartphone" many of the apps report back using the GPS (or the cell tower location if the GPS is turned off) to supply the location aware information. For example, when you use the map application, this reports your location and speed back to report traffic conditions. This data can be further mined to provide OD data. INRIX, TomTom, and others use this GPS data for traffic data analysis and routing recommendations, but it can also be used for OD. WSDOT and UW researchers have used truck GPS data sources to determine freight bottlenecks and truck OD and there are reports available for download.
Because these are all samples, and the purpose of the trip needs to be assumed, there are still some further data needs, but this helps fill more pieces of the puzzle. My experience is the cellphone data, while usually good enough for most TAZ to TAZ movements, is not location specific enough for a detailed TAZ system. We still don't know the actual origin and destination due to the resolution of the cellphone data (based upon triangulation with the towers) or we could tie that in with GIS data and perform some very detailed analysis. For an example of the spatial resolution, when looking at using this data for a detailed study, we "watched" some historical data from one cellphone that was probably being used for calls from an office that moved randomly (and sometimes what appeared to be very quickly) over a 4 block area. This location data may have better resolution in Europe as they use only GSM cellphones whereas in North America we have both CDMA and GSM. I hear the location issues for those are somewhat different but I don't know more than that.
I recommend if you want more information on cellphone data to contact Airsage or for GPS data contact INRIX or TomTom or others. I hope this is helpful.
Bob
Robert Shull, PE
President
Eco Resource Management Systems Inc.
PO Box 1850
Vashon, WA 98070
206.414.8751
rshull@transportmodeler.com
On Fri, Feb 17, 2012 at 1:38 PM,
<Elaine.Murakami@dot.gov> wrote:
The people from Stats Netherlands said they got the data from the telecommunications provider, it is NOT a sample, for “all” data for a 2 week period, but they did not discuss HOW they negotiated the use of the dataset.
Elaine
Thanks for sharing this. One question - How did the Stats Netherlands people get access to the data from the cellphone providers? Maybe there are some lessons there about structuring the request and alleviating public fears that can be useful to data collection efforts here. Or did they select their sample and then approached the providers with explicit permission from this sample of the cellphone customers. In general, my impression is that overseas it is a bit easier for agencies to obtain data than here.
Krishnan
On Fri, Feb 17, 2012 at 2:41 PM, <Elaine.Murakami@dot.gov> wrote:
I thought I would share my notes from this conference. FCSM is held every 2 years, and focuses on the many different surveys conducted by federal agencies.
I have rarely attended, but this time, I gave a presentation about the Transportation Secure Data Center
http://www.nrel.gov/vehiclesandfuels/secure_transportation_data.html , of which FHWA is the major sponsor. My attached notes do not cover my (and Evan Burton’s) presentation.
It is always interesting to see what people are trying, given that we have so many of the same problems: cell-phone only households, incentives, non-response, quality of response, disclosure avoidance, etc. So, even if this post is not specifically about CTPP, there are some tidbits related to ACS and census research at new university centers.
Elaine Murakami
FHWA Office of Planning
206-220-4460 (in Seattle)
_______________________________________________
ctpp-news mailing list
ctpp-news@ryoko.chrispy.net
http://ryoko.chrispy.net/mailman/listinfo/ctpp-news
--
Krishnan Viswanathan
1101 High Meadow Dr
Tallahassee FL 32311
_______________________________________________
ctpp-news mailing list
ctpp-news@ryoko.chrispy.net
http://ryoko.chrispy.net/mailman/listinfo/ctpp-news