Merge branch 'dev'

EmmeProject.py

@@ -1,3 +1,17 @@
+#Copyright [2014] [Puget Sound Regional Council]
+
+#Licensed under the Apache License, Version 2.0 (the "License");
+#you may not use this file except in compliance with the License.
+#You may obtain a copy of the License at
+
+#    http://www.apache.org/licenses/LICENSE-2.0
+
+#Unless required by applicable law or agreed to in writing, software
+#distributed under the License is distributed on an "AS IS" BASIS,
+#WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
+#See the License for the specific language governing permissions and
+#limitations under the License.
+
 import inro.emme.desktop.app as app
 import inro.modeller as _m
 import inro.emme.matrix as ematrix
@@ -12,6 +26,7 @@
 from multiprocessing import Pool, pool
 sys.path.append(os.path.join(os.getcwd(),"inputs"))
 from input_configuration import *
+from EmmeProject import *
 
 
 class EmmeProject:
@@ -101,20 +116,32 @@ def process_shape(self, linkshape_file):
             scenario = self.current_scenario)
     def change_scenario(self):
         self.current_scenario = list(self.bank.scenarios())[0]
+
     def delete_matrix(self, matrix):
         NAMESPACE = "inro.emme.data.matrix.delete_matrix"
         process = self.m.tool(NAMESPACE)
         process(matrix, self.bank)
 
+    def delete_matrices(self, matrix_type):
+        NAMESPACE = "inro.emme.data.matrix.delete_matrix"
+        process = self.m.tool(NAMESPACE)
+        for matrix in self.bank.matrices():
+            if matrix_type == "ALL":
+                process(matrix, self.bank)
+            elif matrix.type == matrix_type:
+                process(matrix, self.bank)
+
     def create_matrix (self, matrix_name, matrix_description, matrix_type):
         NAMESPACE = "inro.emme.data.matrix.create_matrix"
         process = self.m.tool(NAMESPACE)
+        print self.current_scenario
         process (matrix_id= self.bank.available_matrix_identifier(matrix_type),
                           matrix_name= matrix_name,
                           matrix_description= matrix_description,
                           default_value=0,
                           overwrite=True,
                           scenario=self.current_scenario)
+
     def matrix_calculator(self, **kwargs):
         spec = json_to_dictionary('matrix_calc_spec')
         for name, value in kwargs.items():
@@ -179,7 +206,7 @@ def network_calculator(self, type, **kwargs):
                 spec[name] = value
         NAMESPACE = "inro.emme.network_calculation.network_calculator"
         network_calc = self.m.tool(NAMESPACE)
-        self.link_calc_result = network_calc(spec)
+        self.network_calc_result = network_calc(spec)
 
     def process_function_file(self, file_name):
         NAMESPACE=("inro.emme.data.function.function_transaction" )
@@ -203,6 +230,32 @@ def matrix_balancing(self, **kwargs):
         compute_matrix = self.m.tool(NAMESPACE)
         report = compute_matrix(spec) 
 
+    def import_matrices(self, matrix_name):
+        NAMESPACE = "inro.emme.data.matrix.matrix_transaction"
+        process = self.m.tool(NAMESPACE)
+        process(transaction_file = matrix_name,
+            throw_on_error = False,
+            scenario = self.current_scenario)
+
+    def transit_line_calculator(self, **kwargs):
+        spec = json_to_dictionary("transit_line_calculation")
+        for name, value in kwargs.items():
+            spec[name] = value
+
+        NAMESPACE = "inro.emme.network_calculation.network_calculator"
+        network_calc = self.m.tool(NAMESPACE)
+        self.transit_line_calc_result = network_calc(spec)
+
+    def transit_segment_calculator(self, **kwargs):
+        spec = json_to_dictionary("transit_segment_calculation")
+        for name, value in kwargs.items():
+            spec[name] = value
+
+        NAMESPACE = "inro.emme.network_calculation.network_calculator"
+        network_calc = self.m.tool(NAMESPACE)
+        self.transit_segment_calc_result = network_calc(spec)
+
+
 def json_to_dictionary(dict_name):
 
     #Determine the Path to the input files and load them

README.md

@@ -1,8 +1,229 @@
-soundcat
-========
+#SoundCast: PSRC's Travel Skims to Activity Based Model#
 
-It's like soundcast, but less ess
+##Background##
+This code is intended to control the exchange of data between the PSRC's network model in Inro's Emme 4.0 software to our new Activity Based Model (ABM) system that is being developed in the DaySim software. DaySim requires several measures of accessibility from our network model in the form of matrices of travel times, costs, and distances. This Python code is intended to:
 
-soundcat is the repository for the code before it has been fully vetted by the team into an official version.
+	+ Import estimates of personal travel by vehicle class and time of day from DaySim
+	+ Estimate Truck Trips by time of day for:
+		- Light Trucks
+		- Medium Trucks
+		- Heavy Trucks
+	+ Estimate any special generators of travel not captured in DaySim for items like
+		- External Stations
+		- Sport's Stadiums
+		- Airports
+		- Convention Centers
+	+ Run Highway, Transit and Non-Motorized Assignments for various vehicle classes and times of day
+	+ Generate matrices of travel information related to Time, Cost and Distance for all od-pairs in our 4000 zone travel model system.
+
+The vehicle skims include 756 matrices (12 time periods x 63 skims). Previous data exchanges between our model systems relied upon transfer via comma-separated formats.  We have implemented code that utilizes an HDF5 database as a storage container for all Emme model output.  The intent is to use hdf5 during runtime for all model processes as there are a variety of api's in various languages to access the data. For now we are assuming an HDF5 database for each time period but this might change as we move forward with implementation.
+
+##Emme Data Structure##
+To effectively utilize multiple cores on a pc for model runs, we need to have a separate Emme databank and corresponding Emme project for every time period that we wish to run in parallel.  So in order to run 12 highway assignments concurrently, we need to have 12 distinct project files with only one databank in each.
+
+The current folder structure is:
+
+Root Directory (for example, C:\ABM)
+ -> Banks
+    -> Bank1
+    -> Bank2
+    -> Bank3
+    -> Bank4
+    etc.
+ -> Projects
+    -> Project1
+    -> Project2
+    -> Project3
+    -> Project4
+    etc.
+
+During code testing, we are relying on hardcoding all paths to the project files in the code.  Once testing of the code is complete, we plan to implement a refined approach to selecting the projects either through the use of a control file or possibly a tkinter based dialog selection using tkFileDialog.
+
+##Input Files##
+As of now, there are a variety of input files that exist in various ascii formats.  These files currently reside in the "Inputs" folder under each bank.  The inputs inlcude:
+
+	1. user_classes.txt (dictionary)
+		This file contains relevant data about the vehicle classes used in the skimming process. It is used to create all relevant 				matrices, link attributes and assignment and skim parameters used in the course of the run.
+	2. vdfs.txt (input)
+		This file contains the specification of the volume delay functions need for assignments in Emme
+	3. tolls.txt (input)
+		This file contains the specification of the link level tolls for the network
+	4. link_calculation.txt (Emme Tool Specification)
+		This file contains the specification for the link calculator tool from Emme Modeller
+	5. node_calculation.txt (Emme Tool Specification)
+		This file contains the specification for the node calculator tool from Emme Modeller
+	6. general_attribute_based_skim.txt (Emme Tool Specification)
+		This file contains the specification for Path Based Skimming of a network attribute from Emme Modeller
+	7. general_generalized_cost_skim.txt (Emme Tool Specification)
+		This file contains the specification for Path Based Skimming of an od object from Emme Modeller
+	8. general_path_based_assignment.txt (Emme Tool Specification)
+		This file contains the specification for Path Based Assignments from Emme Modeller
+	9. general_path_based_volume.txt (Emme Tool Specification)
+		This file contains the specification for Path Based Class Specific volumes from Emme Modeller
+
+We are working on a solution to generalize the creation of as many of these inputs as possible.  The vdfs and tolls file could reside in our hdf5 datastore.  The Emme specifications might be auto-generated based on the inputs in the user_class dictionary.  For now, these specification files work for networks with 21 user classes.
+
+##Time Periods##
+DaySim calculates travel for all hours of the day.  In order to provide meaningful accessibility data to DaySim and still maintain reasonable model run times, the PSRC network model will be various time periods per day for various modal purposes.  The difference by mode reflects the availability of network related data b ymode and time of day.  
+
+###Time Periods for Highway Assignments###
+The time periods for Highway Assignments are are defined as:
+
+	1. Early AM		5:00 am - 6:00 am
+	2. AM Peak Hour 1	6:00 am - 7:00 am
+	3. AM Peak Hour 2	7:00 am - 8:00 am
+	4. AM Peak Hour 3	8:00 am - 9:00 am
+	5. AM Peak Hour 4	9:00 am - 10:00 am
+	6. Midday		10:00 am - 2:00 pm
+	7. PM Peak Hour 1	2:00 pm - 3:00 pm
+	8. PM Peak Hour 2	3:00 pm - 4:00 pm
+	9. PM Peak Hour 3	4:00 pm - 5:00 pm
+	10. PM Peak Hour 4	5:00 pm - 6:00 pm
+	11. Evening		6:00 pm � 8:00 pm
+	12. Overnight		8:00 pm - 5:00 am
+
+###Time Periods for Transit Assignments###
+The time periods for Transit Assignments are are defined as:
+
+	1. AM			6:00 am - 9:00 am
+	2. Midday		9:00 am - 3:00 pm
+	3. PM			3:00 pm - 6:00 pm
+	4. Evening		6:00 pm - 8:00 pm
+	5. Night		8:00 pm - 6:00 am
+
+###Vehicle Classification for Highway Assignments###
+Assignment Specifications are currently set to work for a 21 class assignments as needed by DaySim.  The 21 classes are:
+
+	1. SOV Toll Income Level 1
+	2. SOV Toll Income Level 2
+	3. SOV Toll Income Level 3
+	4. SOV No Toll Income Level 1
+	5. SOV No Toll Income Level 2
+	6. SOV No Toll Income Level 3
+	7. HOV 2 Toll Income Level 1
+	8. HOV 2 Toll Income Level 2
+	9. HOV 2 Toll Income Level 3
+	10. HOV 2 No Toll Income Level 1
+	11. HOV 2 No Toll Income Level 2
+	12. HOV 2 No Toll Income Level 3
+	13. HOV 3 Toll Income Level 1
+	14. HOV 3 Toll Income Level 2
+	15. HOV 3 Toll Income Level 3
+	16. HOV 3 No Toll Income Level 1
+	17. HOV 3 No Toll Income Level 2
+	18. HOV 3 No Toll Income Level 3
+	19. Light Trucks
+	20. Medium Trucks
+	21. Heavy Trucks
+
+###Value of Time###
+The values of time are currently being refined.  Originally, the value of time categories used the assignment are coming from DaySim and are:
+
+	Class		$ per Hour		
+			Cat #1	Cat #2	Cat #3
+	SOV		$2.00	$8.00	$20.00
+	HOV 2		$4.00	$16.00	$40.00
+	HOV 3+		$6.00	$24.00	$60.00
+	Trucks		$40.00	$45.00	$50.00		
+
+	Class		minutes per cent		
+			Cat #1	Cat #2	Cat #3
+	SOV		0.3000	0.0750	0.0300
+	HOV 2		0.1500	0.0375	0.0150
+	HOV 3+		0.1000	0.0250	0.0100
+	Trucks		0.0150	0.0133	0.0120
+
+
+###Matrix Definition###
+The code uses the Emme Tool for creating matrices to create the 84 total demand and skim matrices needed for the model run.  The code overwrites any existing matrices as DaySim will be feeding new demand each time it access Emme for skimming and the skims should change due to the new demand.  There are 21 demand trip tables and 63 total skim tables being created for auto modes (time, cost and distance).  
+
+Naming convention is: class (2 characters), toll/notoll(2 characters), income (1 number), type (1 character)
+
+	1. svtl1v - SOV Toll Income Level 1 Demand
+	2. svtl2v - SOV Toll Income Level 1 Demand
+	3. svtl3v - SOV Toll Income Level 1 Demand
+	4. svnt1v - SOV No Toll Income Level 1 Demand
+	5. svnt2v - SOV No Toll Income Level 1 Demand
+	6. svnt3v - SOV No Toll Income Level 1 Demand
+	7. h2tl1v - HOV 2 Toll Income Level 1 Demand
+	8. h2tl2v - HOV 2 Toll Income Level 1 Demand
+	9. h2tl3v - HOV 2 Toll Income Level 1 Demand
+	10. h2nt1v - HOV 2 No Toll Income Level 1 Demand
+	11. h2nt2v - HOV 2 No Toll Income Level 1 Demand
+	12. h2nt3v - HOV 2 No Toll Income Level 1 Demand
+	13. h3tl1v - HOV 3+ Toll Income Level 1 Demand
+	14. h3tl2v - HOV 3+ Toll Income Level 1 Demand
+	15. h3tl3v - HOV 3+ Toll Income Level 1 Demand
+	16. h3nt1v - HOV 3+ No Toll Income Level 1 Demand
+	17. h3nt2v - HOV 3+ No Toll Income Level 1 Demand
+	18. h3nt3v - HOV 3+ No Toll Income 
+	19. lttrkv - Light Truck Demand
+	20. mdtrkv - Medium Truck Demand
+	21. hvtrkv - Heavy Truck Demand
+	22. svtl1t - SOV Toll Income Level 1 Time
+	23. svtl2t - SOV Toll Income Level 1 Time
+	24. svtl3t - SOV Toll Income Level 1 Time
+	25. svnt1t - SOV No Toll Income Level 1 Time
+	26. svnt2t - SOV No Toll Income Level 1 Time
+	27. svnt3t - SOV No Toll Income Level 1 Time
+	28. h2tl1t - HOV 2 Toll Income Level 1 Time
+	29. h2tl2t - HOV 2 Toll Income Level 1 Time
+	30. h2tl3t - HOV 2 Toll Income Level 1 Time
+	31. h2nt1t - HOV 2 No Toll Income Level 1 Time
+	32. h2nt2t - HOV 2 No Toll Income Level 1 Time
+	33. h2nt3t - HOV 2 No Toll Income Level 1 Time
+	34. h3tl1t - HOV 3+ Toll Income Level 1 Time
+	35. h3tl2t - HOV 3+ Toll Income Level 1 Time
+	36. h3tl3t - HOV 3+ Toll Income Level 1 Time
+	37. h3nt1t - HOV 3+ No Toll Income Level 1 Time
+	38. h3nt2t - HOV 3+ No Toll Income Level 1 Time
+	39. h3nt3t - HOV 3+ No Toll Income 
+	40. lttrkt - Light Truck Time
+	41. mdtrkt - Medium Truck Time
+	42. hvtrkt - Heavy Truck Time
+	43. svtl1c - SOV Toll Income Level 1 Cost
+	44. svtl2c - SOV Toll Income Level 1 Cost
+	45. svtl3c - SOV Toll Income Level 1 Cost
+	46. svnt1c - SOV No Toll Income Level 1 Cost
+	47. svnt2c - SOV No Toll Income Level 1 Cost
+	48. svnt3c - SOV No Toll Income Level 1 Cost
+	49. h2cl1c - HOV 2 Toll Income Level 1 Cost
+	50. h2cl2c - HOV 2 Toll Income Level 1 Cost
+	51. h2cl3c - HOV 2 Toll Income Level 1 Cost
+	52. h2nt1c - HOV 2 No Toll Income Level 1 Cost
+	53. h2nt2c - HOV 2 No Toll Income Level 1 Cost
+	54. h2nt3c - HOV 2 No Toll Income Level 1 Cost
+	55. h3cl1c - HOV 3+ Toll Income Level 1 Cost
+	56. h3cl2c - HOV 3+ Toll Income Level 1 Cost
+	57. h3cl3c - HOV 3+ Toll Income Level 1 Cost
+	58. h3nt1c - HOV 3+ No Toll Income Level 1 Cost
+	59. h3nt2c - HOV 3+ No Toll Income Level 1 Cost
+	60. h3nt3c - HOV 3+ No Toll Income 
+	61. lttrkc - Light Truck Cost
+	62. mdtrkc - Medium Truck Cost
+	63. hvtrkc - Heavy Truck Cost
+	64. svtl1d - SOV Toll Income Level 1 Distance
+	65. svtl2d - SOV Toll Income Level 1 Distance
+	66. svtl3d - SOV Toll Income Level 1 Distance
+	67. svnt1d - SOV No Toll Income Level 1 Distance
+	68. svnt2d - SOV No Toll Income Level 1 Distance
+	69. svnt3d - SOV No Toll Income Level 1 Distance
+	70. h2dl1d - HOV 2 Toll Income Level 1 Distance
+	71. h2dl2d - HOV 2 Toll Income Level 1 Distance
+	72. h2dl3d - HOV 2 Toll Income Level 1 Distance
+	73. h2nt1d - HOV 2 No Toll Income Level 1 Distance
+	74. h2nt2d - HOV 2 No Toll Income Level 1 Distance
+	75. h2nt3d - HOV 2 No Toll Income Level 1 Distance
+	76. h3dl1d - HOV 3+ Toll Income Level 1 Distance
+	77. h3dl2d - HOV 3+ Toll Income Level 1 Distance
+	78. h3dl3d - HOV 3+ Toll Income Level 1 Distance
+	79. h3nt1d - HOV 3+ No Toll Income Level 1 Distance
+	80. h3nt2d - HOV 3+ No Toll Income Level 1 Distance
+	81. h3nt3d - HOV 3+ No Toll Income 
+	82. lttrkd - Light Truck Distance
+	83. mdtrkd - Medium Truck Distance
+	84. hvtrkd - Heavy Truck Distance
+
+###Vehicle Matrix Calculations###
+The code creates three sets of skims for use by DaySim - travel time, generalized cost and distance.  All three skim procedures utilize the standard path based assignment analysis toolkits from Emme Modeller.  The travel time skims are created by skimming auto time (timau) across all paths, distance skims are based on link length and the generalized cost skims use the conversion of toll costs to time via values of time as noted above.
 
-When it's done, we put it in SoundCast

configuration_template.properties

@@ -107,12 +107,12 @@ IxxiFirstLineIsHeader = false
 
 # zone list, district lookup
 ImportZones = true 
-RawZonePath = TAZIndex_5_28_14.txt 
+RawZonePath = TAZIndex.txt 
 RawZoneDelimiter = 9 
 
 # park and ride nodes
 ImportParkAndRideNodes = true 
-RawParkAndRideNodePath = p_rNodes_040814.csv 
+RawParkAndRideNodePath = p_r_nodes.csv 
 RawParkAndRideNodeDelimiter = 44 
 
 # daysim output text files
@@ -259,9 +259,9 @@ PathImpedance_TransitFerryPathConstant = 0.0
 PathImpedance_TransitUsePathTypeSpecificTime = true 
 
 PathImpedance_TransitPremiumBusTimeAdditiveWeight = 0.00 
-PathImpedance_TransitLightRailTimeAdditiveWeight = -0.10 
-PathImpedance_TransitCommuterRailTimeAdditiveWeight = -0.35 
-PathImpedance_TransitFerryTimeAdditiveWeight = -0.60 
+PathImpedance_TransitLightRailTimeAdditiveWeight = -0.10
+PathImpedance_TransitCommuterRailTimeAdditiveWeight = -0.50
+PathImpedance_TransitFerryTimeAdditiveWeight = -0.50 
 PathImpedance_BikeUseTypeSpecificDistanceFractions = false 
 PathImpedance_BikeType1DistanceFractionAdditiveWeight = 0.0 
 PathImpedance_BikeType2DistanceFractionAdditiveWeight = 0.0