Big data Europe the transport pilot in Thessaloniki - Josep Maria Salanova

Dr. Josep Maria Salanova GrauCenter for Research and Technology Hellas - Hellenic Institute of TransportHead of “Data collection and processing, algorithm design, and use of specialized transport software packages” laboratoryEmail: [email protected]: www.hit.certh.gr

Big Data Europe. The transport pilot in Thessaloniki.

Mobile sensors in Thessaloniki• Stationary sensors network: Point to point tracking of MAC

ids along the network through 43 Bluetooth device detectors.▫ Travel time estimation▫ Route choice model calibration▫ Origin – Destination matrix estimation / Mobility patterns

estimation▫ Traffic flow extrapolation

• Dynamic sensors fleet: Floating Car Data provided in real time by a professional fleets composed of 1.200 taxis and 600 buses▫ Traffic status estimation (average speed)▫ Origin – Destination matrix estimation / Mobility patterns

estimation▫ Taxi/bus performance indicators

• Social media (geolocated tweets & Facebook check-in events)▫ Activity patterns estimation▫ Events / incidents detection▫ Attraction models estimation

Mobile sensors in Thessaloniki• http://160.40.63.110/desktop/

Mobile sensors in Thessaloniki• www.trafficthess.imet.gr/

Mobile sensors in Thessaloniki• www.trafficpaths.imet.gr

Mobile sensors in Thessaloniki• http://opendata.imet.gr/

Datatank (Back office + APIs)CKAN (front end)

BDE pilot in Thessaloniki

•Probe data that is used▫Floating Car Data (500-2.500 locations per

minute)▫Bluetooth detections (millions of daily

detections in 43 locations)•Services that are being implemented

▫Map matching▫Mobility patterns recognition and forecasting

(ARIMA)

BDE pilot in Thessaloniki – pilot 1

GPS DataMap Matching Traffic Classification and

Prediction

Classification and Prediction Data

Bluetooth Data

Results

•Probe data that is used▫Floating Car Data (500-2.500 locations per

minute)▫Bluetooth detections (millions of daily

detections in 43 locations)•Services that are being implemented

▫Improved topology-based map matching▫Mobility patterns recognition and forecasting

(ARIMAX + NN)

GPS DataMap Matching Traffic Classification and

Prediction

Classification and Prediction Data

Bluetooth Data

Results

BDE pilot in Thessaloniki – pilot 2&3

Map Matching Algorithm

BDE pilot in Thessaloniki – pilot 2&3

rownum recorded_timestamp transfer osmids

263005 08:43:18 1 213910068

263171 08:43:27 1 16457977

263350 08:43:37 1 213910068

263657 08:43:55 1 213910068

263919 08:44:10 1 22564731

264174 08:44:25 1 222137984

264701 08:44:56 1 222137984

264813 08:45:04 1 222137984

265071 08:45:19 0 22564731

265363 08:45:35 0 213910068

265565 08:45:48 0 213910068

265737 08:45:57 0 16457977

265906 08:46:07 0 213910068

Map Matching Algorithm

BDE pilot in Thessaloniki – pilot 2&3

Start

Historical Link Traffic State (FCD)

Historical Link Traffic State Classification

Historical Link Traffic State (Loop Detectors)

Historical Link Traffic State BT

Compare Traffic States(ML, NN)

Define Current Link Traffic State

Predict (ARIMAX | NN)

Store in Historical States

Validate Prediction

Traffic Classification and Prediction

BDE pilot in Thessaloniki – pilot 2&3

Dr. Josep Maria Salanova [email protected] +30 2310 498 433

Thank you!