Introduction to Cube: Cube Base, Scenario Analysis, Editing, Mapping, and Scripting

Transcription

1 Introduction to Cube: Cube Base, Scenario Analysis, Editing, Mapping, and Scripting

2 Objectives for Participants General Technical 2

3 Agenda/Training Contents Lesson 1: Lesson 2: Lesson 3: Lesson 4: Lesson 5: Lesson 6: Lesson 7: Lesson 8: Lesson 9: Lesson 10: 3

4 Typical Uses of Transportation Planning Models Systems planning Corridor and sub-area studies Strategic planning

5 Typical Transportation Planning Model Elements Inputs: Processes: Outputs:

6 Lesson 2: Four-Step Modeling Process 6

7 The Four-Step Modeling Process Trip Generation Trip Distribution Mode Choice Network Assignment

8 Typical Four-Step Travel Demand Model Highway Network Socioeconomic Data Transit Network Highway Path Building and Toll Estimation 1 Trip Generation Transit Path Building and Fare Estimation 2 Trip Distribution 3 Mode Split Transit Assignment 4 Highway Assignment Potential Feedback to Convergence 8

9 Zonal Data Data structure: Boundaries: Attributes:

10 Network data

11 Highway Networks Data structure: Centroids: Functional classification or facility types: Other attributes: Distances:

12 Intersection Data Only required if performing intersectionconstrained highway traffic assignment Calculate capacity for each turn at nodes Intersection types and models: Turning movements (flows and delays)

13 Public Transport (PT) Data Defined in relation to underlying transportation infrastructure network Public Transport lines: Non-transit legs: System: modes, operators, fares, users

14 Network Window Visualization Environments GIS Window

15 Answers the question: Step 1: Trip Generation Productions and attractions are a function of socio-economic attributes of zone, 150 HH 500 EMP 1300 trips Categories of trips: Typical model forms:

16 Cross-Classification Trip Chaining 16

17 Concepts underlying trip purpose I-I E-I E-E 17

18 Typical Trip Purposes Home-Based Work: Home-Based Non-Work: Non-Home Based: External Trips: 18

19 Trip Generation - Scripts FILEI ZDATI[x] Zonal variables are referenced in script as Zi.x.variable FILEI LOOKUPI[x] Lookup tables contain production and attraction rates FILEO PAO[x] Output file contains productions and attractions to be used in the distribution phase PARAMETERS zones=num ; defines the number of zones ILOOP Phase COMP P[num]= ; num refers to a particular user class COMP A[num]= ADJUST Phase COMP P[num][zone]= ; zone=0 produces the total of the array BALANCE.. 19

20 Example: 4-purpose Trip Generation ; E-I trip data file FILEI ZDATI[2] = "C:\TrainingModel\Model\EITRIPS.DBF" ; land use data file FILEI ZDATI[1] = "{Zones}" ; DBF: Z=zone field name ; TXT: Z=zone field location, var=field location, var=field location... ZONES = 25 IF (I<=16) ; calculate productions by purpose P[1] = 0.74*zi.1.hh1+1.67*zi.1.hh2+2.01*zi.1.hh3+2.58*zi.1.hh4 P[2] = 2.00*zi.1.hh1+4.10*zi.1.hh2+5.94*zi.1.hh3+7.89*zi.1.hh4 P[3] = 0.96*zi.1.hh1+1.82*zi.1.hh2+2.65*zi.1.hh3+3.13*zi.1.hh4 ; calculate attractions by purpose A[1] = 1.45*zi.1.total_emp A[2] = CmpNumRetNum(zi.1.areatype,'=',1,2.0,9.0)*zi.1.retail+1.7*zi.1.service+0.5*zi.1.other+0.9*zi.1.households A[3] = CmpNumRetNum(zi.1.areatype,'=',1,1.4,4.1)*zi.1.retail+1.2*zi.1.service+0.5*zi.1.other+0.5*zi.1.households ELSE P[4] = zi.2.eitrips ENDIF ; adjust zonal attractions so total attractions match total productions PHASE=ADJUST BALANCE, A2P=1,2, NHB=3 A[4]=(P[4][0]*(A[1]+A[2]+A[3])/(A[1][0]+A[2][0]+A[3][0])) ; output zonal productions and attractions FILEO PAO[1] = "{SCENARIO_DIR}\TRIPENDS.DBF", DBF=T, FORM=6.0, LIST=Z, P[1] P[2] P[3] P[4] A[1] A[2] A[3] A[4] 20

21 Answers the question Step 2: Trip Distribution D1 Function of activity concentrations and zone-tozone travel costs Process calibrated to match observed trip length distribution (e.g. from surveys) Typical forms: D4 D3 Origin D2

22 Trip Distribution I. Trip Generation Estimates II. Trip Distribution (Zones) Zone Production Attraction Zone 3 50 Attraction Zone Production 250 Attraction 20 Trips 5 Minutes Total Trips 15 Minutes 50 Trips remain as Intrazonal Trips Zone Attraction 22

23 Gravity Model 23

24 Trip Distribution Frequency NJRTM - 96 VALIDATION FREQUENCY DISTRIBUTION FOR HBW FREQUENCY (%) OBS. AVG. EST. AVG. (min.) (min.) TRAVEL TIME OBSERVED ESTIMATED Final model should replicate trips by time interval. This proves the model allocates trips properly 24

25 Desire Lines: Direction and Magnitude of Travel 25

26 Trip Distribution - Scripts GRAVITY Command: SETPA: 26

27 Example: 4-purpose Gravity Model LOOKUP, FILE = "C:\TrainingModel\MODEL\FFACTORS.DBF", INTERPOLATE=Y, NAME=FF, LOOKUP[1]=1,RESULT=2, LOOKUP[2]=1,RESULT=3, LOOKUP[3]=1,RESULT=4, LOOKUP[4]=1,RESULT=5 ; setup the working p's and a's SETPA P[1]=P1 A[1]=A1 SETPA P[2]=P2 A[2]=A2 SETPA P[3]=P3 A[3]=A3 SETPA P[4]=P4 A[4]=A4 ; get the los matrix into work matrix 10 MW[10] = MI.1.1 ; do 3 gravity models GRAVITY PURPOSE=1, LOS=MW[10], FFACTORS=FF GRAVITY PURPOSE=2, LOS=MW[10], FFACTORS=FF GRAVITY PURPOSE=3, LOS=MW[10], FFACTORS=FF GRAVITY PURPOSE=4, LOS=MW[10], FFACTORS=FF Source: NCHRP

28 Step 3: Mode Split (aka Mode Choice) Answers the question, Possible modes: The probability of selecting a given mode is a function of the relationship between the cost of competing modes Typical forms:

29 Nested Logit Model 29

30 Nested Logit Model Logsum Function 30

31 Typical Nested Logit Structure Total Person Trips Auto Mode Share Transit Mode Share Drive-Alone Mode Shared-Alone Mode Walk Access Auto Access HOV-2 Mode HOV-3 Mode HOV-4+ Mode 31

32 Mode Split Choice Modeling in Cube XCHOICE Command Choice Models that can be developed in Cube include: ) 32

33 Example: Absolute logit model for 3 modes

34 Time-of-Day Factoring (Intermediate Step) Answers the question, Trip Generation, Distribution, and Mode Split create in production-attraction format To translate from P/A into origin-destination format Time Segment 1 Time Segment S Time D Segment 1 D i D N D z 1 D i D O 1 T D z 1,1 1 T D 1,i i T D 1,z z O 1 T 1,1 T 1,i T O O 1,z i 1 T T i,1 1,1 T T i,i 1,i T T O i,z 1,z i T i,1 T i,i T i,z O O z i T T z,1 i,1 T T z,i i,i T T O z,z i,z z T z,1 T z,i T O z,z z T z,1 T z,i T z,z Time-of-day factors are simultaneously applied,

35 Time of Day Trip Allocation Figure -- Time-of-Day of Travel 6.00% 5.00% Starting Time of Trips 4.00% 3.00% 2.00% 1.00% 0.00% 00:00-00:29 01:30-01:59 03:00-03:29 04:30-04:59 06:00-06:29 07:30-07:59 09:00-09:29 10:30-10:59 12:00-12:29 13:30-13:59 15:00-15:29 16:30-16:59 18:00-18:29 19:30-19:59 21:00-21:29 22:30-22:59 Percent of Total Trips HBW HBNW NHB TOTAL 35

36 Time of Day Trip Distribution - Example 36

37 Example: Time of Day and Vehicle Trip Factoring

38 Step 4: Highway and Transit Assignment Answers the question Function of interaction between travel demand and transportation supply including congestion Equilibrium: Typical forms:

39 Highway Module Structure Phases multiple iterative loops Methods convex combinations 39

40 Highway Module Commands and Keywords PARAMETERS: LINKREAD functions: ILOOP functions: 40

41 Example: Single Matrix Equilibrium 41

42 Highway Select Link Analysis Select Link Map 42

43 Highway Turning Movement Estimation Highway Select Link Analysis 43

44 Lesson 3: Overview of Citilabs & Cube 44

45 CITILABS THE COMPANY Develops software for the modeling of transportation systems Offices Used in 2,500 cities on 6 continents in more than 80 countries

46 CITILABS THE COMPANY (cont d) Longstanding ESRI business partner Owners: Professional services/consulting

47 CUBE: PROFESSIONAL TRANSPORTATION MODELING SUITE System Interface Cube Base Demand Modeling Cube Voyager: Cube Land : Cube Cargo: Simulation Cube Avenue: Cube Dynasim: Cube Base Specialized Cube Cluster: Cube Analyst & Analyst Drive: Cube Cloud Application and sharing framework for transportation planning Cube Cloud Services

48 Other Citilabs Products Sugar Network Editor Accession Beta versions Turnkey Models and Tools GIS Web Apps

49 The Flow-Chart: Easy Model Development Famous for its flow-charting environment for designing and building transportation models. Modules are accessed through pull-down menus Dropped into a flow chart Data inputs and outputs linked by drag-and-drop.

50 Built for Scenario Testing Easy to use environment: Menus prompt user for inputs and parameters to test Integrated report and charting generators to assist in analysis

51 Flexibility The only system equipped with its own comprehensive scripting language for transportation modeling Create customized models without difficult programming languages Access many scripts through simple menu clicks Move custom scripts and add as point and click functions within Cube

52 Transportation GIS Built on ESRI Only modeling system that comes with a complete transportation GIS built on ESRI s market leading GIS technology. Store all data directly in ESRI s geodatabase format. No need to convert data back and forth between the GIS Department and the modeling team. ArcGIS Extension for network editing (Sugar)

53 Data Manager Toolbar buttons Add Data Create Geodatabase Build Network From Shape Import/Export Data Workspace

54 Cube GIS Window Node/Point Chart Graphics

55 New Features in Cube 6 Cube User Interface Components Cube GIS Welcome Screen New Text Editor Cube PT New Features New Functions and Commands as well Displayed by default after launch

56 Cube File Options Tools available to use now include:

57 Cube File Options (Cont d)

58 New Ribbon Interface Ribbon is designed to help quickly find commands that users need Commands organized in logical groups, collected together under tabs. Each tab is related to a type of activity such as: Some tabs are shown only when needed

59 Ribbon Customization Right You may drag commands onto and off ribbon, the Quick Access toolbar, etc

60 Docking Windows The side windows from Scenario Manager (scenario, data, application, keys..) are now fully:

61 Tabbed Window Interface Easy to select a needed window Tabs may also be dragged left and right to reorder Tabbed interface can be enabled in Options Can add special commands to Quick Access Toolbar

62 Application Manager Click for process templates New graphics with rounded corners, and color schemes. Application stays organized when changing resolution

63 Quick Access Bar Customization Click dropdown arrow to right of toolbar. Select to launch same customization interface used with Ribbon Alternately, right-click the Quick Access Toolbar

64 1 Tab Mode Toggle 1. Click More Buttons 2. Choose Tab Mode Toggle 3. Click Tab Mode Toggle 4. Choose a tab option 2 4 3

65 Network Window Toolbar Toolbar can float over interface Easy to customize

66 Cube - Bookmark this view Allows user to save network views And restore saved views

67 Cube Build Shortest Path User can specify additional criteria to find shortest path and Either display or save the paths to a text or database file

68 Cube Data Manager Data Manager allows between and within geo-databases

69 Cube GIS GIS Editor now allows post links and post nodes options

70 Cube GIS (cont d) GIS Windows are now fully dockable, collapsible with auto-hide functionality

71 Multi-bandwidth now available Cube GIS (cont d)

72 New Text Editor Smart autocomplete for commands and keys with new code assistant Column mode editing Search and replace with bookmark support Line numbers Use of markers by clicking to the right of line number Collapsible comment groups Zooming functionality Incorporation of tab Color themes

73 Updates to Public Transport (PT) Module PT Matrix Estimation is now supported with full features. PT Period-Based Keywords. Headway[p] used to be the only variable indexed by the HDWAYPERIOD parameter now you can also index:

74 Updates to PT Module (Cont d) PT Fares during path-building (in Enumeration phase) PT Drive Access generation enhancements tons of new options Some highlights:

75 New PT Keywords Enhanced Transit Route Enumeration Enhanced Drive-Access Generation Enhanced Fares along with BestPathOnly Enhanced Transit Line Keywords

76 Enhanced PT Drive-Access Generation Enhanced Drive-Access Generation Drive-Access Generation in 5.x Drive-Access Generation Examples in 5.x Drive-Access Generation in 6.x Enhanced Drive-Access Generation Example

77 Examples of PT Drive-Access Coding Aerial photograph for parking lot & transit stops

78 Examples of PT Drive-Access Coding Schematic diagram

79 PT Drive-Access Generation in 6.x Cube Voyager 6.0 has built upon features of existing Generate command to now consider:

80 PT Drive-Access Generation in 6.x (cont d)

81 Enhanced PT Fare Process in 6.x User may specify a simple fare structure while enumerating routes to enhance path building process. Cube Voyager 6.0 includes additional fare keywords along with BestPathOnly:

82 Enhanced PT Transit Line Keywords in 6.x

83 CUBE 6.1

84 PRODUCT COMPATIBILITY COMPATIBILITY UPDATES Cube is now compatible with ESRI ArcGIS 10.1/SP 1 and uses this as the default ArcGIS Engine Cube is also now compatible with Windows 8 & Server bit version of Cube is in Beta testing

85 USABILITY UPDATES USABILITY UPDATES

86 MODELING IMPROVEMENTS IMPROVEMENTS TO HIGHWAY MODELING Cube Base, Cube Voyager and Cube Avenue now support 1,000 volume sets! Bi-Conjugate Frank-Wolfe Assignment has a new SMOOTH parameter to reduce oscillation patterns at the tail of the curve. Update: currently in internal beta:

87 Cube 6 - The only system offering Cloud Computing

88 CUBE CLOUD CUBE CLOUD: BIG DATA At Cube Cloud store: Now available:

89 CUBE CLOUD CUBE CLOUD STORE: APPS Free apps provide access to commonly used analytic tools. Comes standard with Cube Cloud! Paid apps allow users to sell their models on the Cloud

90 CUBE CLOUD CUBE CLOUD STORE: RESOURCES

91 Cube 6: Moving Your Model to Cube Cloud TRAVEL MODEL Amazon s EC2 Cloud Computing Environment Model Developed with Cube P u b l i s h Model Run with Cube Cloud TRAVEL MODEL

92 Run Scenarios with a Simple Web-Interface

93 Make it Easy to Map Results

94 Make it Easy to Get Charts and Tables

95 Cube Cluster Distributed processing to increase operational speed Cube Cluster for your model can be managed in two ways:

96 Hours Nashville (Minimum # of Cores =8 ) Run Times (Hours) Cube Cluster Number of Cores % reduction vs. single core 7.3

97 MiniQuiz 1 The following programs are used to build a passenger travel demand model in Cube The following programs are used to built a meso-scopic simulation in Cube