Bringing Freight Components into Statewide and Regional Travel Demand Forecasting

Transcription

1 Bringing Freight Components into Statewide and Regional Travel Demand Forecasting Center for Quality Growth and Regional Development Georgia Institute of Technology PI: David Jung-Hwi Lee Co-PI: Catherine L. Ross University Transportation Center (UTC) Conference for the Southeastern Region March 24, 4

2 Research Overview Need DOTs and MPOs need freight demand models that are reliable, accurate, and approachable. Purpose Leverage new data sources Benchmark freight modeling best practices Develop longterm guidelines for freight demand models Project Goals Study best practices and extent of usage of GPS data in freight modeling Build prototype tourbased truck models with GPS-based truck data Test model improvements compared with existing models

3 Problem Statement Lack of Urban Freight Demand Models Few practical freight forecasting models More significant in small and medium-sized MPOs Models missing freight component could overestimate capacity Incapability to provide adequate info to decision makers

4 DOT and MPO Survey Summary of Results Freight models are still relatively rare about half of DOTs and one quarter of MPOs Most models are vehicle-based 6% % 9% 4% 5% 9% 8% 35% 8% 4% 7% 3% 7% 3% 6% 6% 25% What What Have Percent modeling primary you of used MPOs method obstacles GPS and do data DOTs you do in you use? using a encounter freight in freight modeling model? toolsfreight? GPS data remains rare used in about one in five vehicle models Lack of data remains a large obstacle to freight modelers GPS data can help 5% % % 4% 4% 5% % 3% 3% % % % % % 5% % % % Freight Studies Freight Models Freight Unavailable Insufficient Insufficient Lack of Performance data funding staffing specialized Measures Yes Noknowledge DOTs MPOs DOTs MPOs

5 Tour-based Truck Model Conceptual Framework Tour Generation Tour Main Destination Choice Intermediate Stop Model Stop Location Model Time of Day Trip Accumulator Traffic Assignment

6 GPS Data Source Time Date Location Feb May Jul Oct

7 GPS Data Source Atlanta TRUCK RECORD: ATL_A_2. (,77,4 records) ATL_A_5. (,54,362 records) ATL_A_7. (,452,66 records) ATL_A_. (,349,4 records) ATL_B_2. (,57,29 records) ATL_B_5. (,973,48 records) ATL_B_7. (2,,84 records) ATL_B_. (2,32,84 records) Total 4,62,934 records ATRI provide 8 weeks of truck GPS data for 5, different trucks in (2 weeks in each season).

8 GPS Data Source Birmingham TRUCK RECORD: BMH_A_2. (497,762 records) BMH_A_5. (465,937 records) BMH_A_7. (387,992 records) BMH_A_. (4,87 records) BMH_B_2. (57,629 records) BMH_B_5. (688,292 records) BMH_B_7. (72,56 records) BMH_B_. (755,895 records) Total 4,488,84 records ATRI provide 8 weeks of truck GPS data for 5, different trucks in (2 weeks in each season).

9 GPS Data Truck Records Truckid: Parking_from: Readdate_from: TAZ from: To_readdate: To_TAZ_: To_Parking: This is a unique truck ID. This indicates if the vehicle is in a known truck stop at the first point: = at a truck stop, = not at a truck stop This is the first date/time stamp in a series This is the TAZ ID for the first position read in a series. This is the second time stamp in a series This is the second TAZ ID in a series This indicates if the vehicle is in a known truck stop at the second point: = at a truck stop, = not at a truck stop Distance traveled: This is distance traveled in miles from point A to point B. It uses the great circle distance equation (i.e. it is not snapped to a roadway).

10 GPS Data Processing Delete records on weekends and holidays. Remove records with improper geocoding Determination on Stopped; Starting to move; in motion; or coming to stop Converting TRUCK records to TRIPS Converting TRIPS records to TOURS Define TOUR All the movements from a Start location until the truck return to the same location From a Start location until midnight of that day Multi-day tours were NOT considered 2,7,995 TRUCK Records 73,36 TRIPS 2,752 TOURS Tours Stops Stops/Tour I/I, , I/X 25,75 39,99.55 X/I 5,845 69, X/X 32,732 48,82.49 Total 2, ,

11 Truck Tours Example TRUCK ID: DATE: Feb. 6, TOUR : Starting from zone 4 Taking stops at: 44, 39, 43, 57, 77, 43, 88 Ending at zone 44

12 Truck Tours Example TRUCK ID: DATE: Feb. 7, TOUR 2: Starting from zone 44 Taking stops at: 434, 678, 85, 89, 43, 39, 432 Ending at zone 4

13 Truck Tours Example TRUCK ID: DATE: Feb. 8, TOUR 3: Starting from zone 43 Taking stops at: 44, 344, 34, 33, 882, 44 Ending at zone 43

14 Truck Tours Example TRUCK ID: DATE: Feb. 6~8, TRUCK: 224 cleaned Truck Records TRIPS: 23 trips TOURS: 3 tours

15 Tour-based Truck Model Validation Observed (Count), 5,, 5, Observed vs. Estimated Link-Level Volume 5,, 5,, Estimated (Model)

16 Key Obstacles and Challenges GPS data is inconsistent Nothing is known about GPS sampling We have no description of truck or operator External station geocoding was not sufficiently accurate

17 Trip-based vs. Tour-based Model Atlanta Link Volume Comparison (54,56 Links) TOUR Model AM ARC Model 4 5 ARC Model 5 TOUR Model 3 PM 4 5

18 Trip-based vs. Tour-based Model Atlanta Link Volume Comparison (54,56 Links) TOUR Model 3 MD 4 5 ARC Model 4 5 ARC Model 5 TOUR Model NT 3 4 5

19 Conclusions and Future Research Conclusions GPS data can create robust tour-based freight models GPS data requires extensive processing to be useful Tour based structure reflects truck travel more accurately. Future steps will compare truck model results with existing freight models in Atlanta and Birmingham. The results are likely to provide new improvements and directions for future research. Future Research Develop methodology and GPS data source that distinguishes different types of trucks Work with modelers in practice to implement tour-based truck models with GPS data Examine usefulness for wideranging applications air quality models, traffic congestion forecasts, and investment decision making