Model Data Analysis¶

Python libraries and the EMME python¶

In general the RTM uses the default python libraries installed with the EMME python. In some cases this means foregoing features that have been recently introduced or having code that is not compatible with python installations outside EMME. However, it ensures that the RTM will run without further configuration providing the user has installed the correct version of EMME.

Python + util¶

The RTM scripting employs a utility toolbox that creates functions for repeated operations. It is worth familiarizing yourself with the functions here as they are often employed outside the RTM run itself to post-process data.

For example, util can be used to create a pandas dataframe for full matrices and append RTM matrices to it for further analysis or plotting in python.

    import inro.modeller as _m

    util = _m.Modeller().tool("translink.util")

    eb = _m.Modeller().emmebank

    df = util.get_pd_ij_df(eb) # create a longform pandas dataframe for full matrices

    # add columns for light and heavy truck OD tables to the data frame
    df['amLGVDemand'] = util.get_matrix_numpy(eb, "mfLGVAM").flatten() 
    df['amHGVDemand'] = util.get_matrix_numpy(eb, "mfHGVAM").flatten()

Which yields the following results:

In the above example matrix names or numbers can be used to access the data in util.get_matrix_numpy().

Data extraction using variables of interest tool¶

In this example we use a spreadsheet to build a sqlite database. This tools allows the user to create tables and fields within a table using EMME matrix calculations. The Variables_of_Interest.xlsx spreadsheet is located in the RTM/Scripts/Phase3Analytics folder.

Setting up the spreadsheet¶

The Datamap tab describes what each field does in the Sheet1 tab. The Sheet1 tab is used to build the database.

A simple matrix calculation can be used to extract data. Entering the following in Sheet1 will create a table in the output sqlite database with AM demand for light and heavy goods vehicles equivalent to the pandas data frame example above:

In addition to simple data extraction, new data can be generated based on RTM outputs using EMME matrix calculator expressions. For example, we can create a matrix of the minimum time on transit between all zones considering both the bus and rail modes. Note that we can reference EMME matrices by name or number.

Running the Modeller Tool¶

Once you have specified the data you wish to output in in Sheet1, the modeller tool Variables of Interest Extraction located in the Phase3Analytics Toolbox can be run to produce the outputs.

To create the database with your specified tables and fields simply run the tool with the appropriate sheet name in the field. If you have not changed the name from Sheet1 there is nothing to update and you can simply run the tool.

Viewing results¶

The requested results will be output to a sqlite database named Variables_of_Interest_Results.db. The output can be viewed in the same manner as the trip_summaries.db and rtm.db discussed here

Post-processing tools¶

RTM provides several standard tools to help users quickly analyze the transportation results after the model run. The common-use pro-processing tools are select link analysis for traffic and select transit analysis for public transit. These two tools are also located in Phase3Analytics Toolbox.

Auto Select Link tool¶

The auto select link tool is for analyzing the auto traffic on a select link. The tool provides a simple capability to let user perform a quick select assignment. Set @selectlink=1 to select the link that to be evaluated.

Transit Select link/line tool¶

The transit select tool provides larger capability to perform more complicated post analysis. It executes the transit assignment based on the select link/ line. Upon the request, the outputs can be:

• transit segment volumes, boardings, alightings for select transit link/line
• link auxiliary transit volume for select walk link
• matrices by transit sub-mode

Before using the tool

There are some procedures needs to be done prior to applying the select transit tool: - Ensure the databank dimension has enough room for additional extra attributes. The recommended number is 2,500,000 - Strategies generated by extended transit assignments. (Finish full model run or transit assignment) - Recommend copy AM, MD, and PM scenario to prevent from overwriting the original scenario
- Create the extra attribute(s) and set the appropriate value in all three scenarios (following the below instruction)

After using the tool

The tool performs post-analyses on paths extracted from strategies generated by extended transit assignments, and it creates: 1. new matrices and save the O-D results:

ID	Name	Descritption
mf110	AMBUSsline	AM Bus Select transit O-D
mf111	AMRALsline	AM Rail Select transit O-D
mf112	AMWCEsline	AM WCE Select transit O-D
mf113	AMTOTsline	AM TOT Select transit O-D
mf130	MDBUSsline	MD Bus Select transit O-D
mf131	MDRALsline	MD Rail Select transit O-D
mf132	MDWCEsline	MD WCE Select transit O-D
mf133	MDTOTsline	MD TOT Select transit O-D
mf150	PMBUSsline	PM Bus Select transit O-D
mf151	PMRALsline	PM Rail Select transit O-D
mf152	PMWCEsline	PM WCE Select transit O-D
mf153	PMTOTsline	PM TOT Select transit O-D

2.new extra attributes and save network results:

Name	Descritption	Name	Descritption
@voltr_bussline	Transit Volume BUS	@board_bussline	Boardings BUS
@voltr_ralsline	Transit Volume RAL	@board_ralsline	Boardings RAL
@voltr_wcesline	Transit Volume WCE	@board_wcesline	Boardings WCE
@voltr_totsline	Transit Volume TOT	@board_totsline	Boardings TOT
@volax_bussline	Aux Transit Volume BUS	@alight_bussline	Alightings BUS
@volax_ralsline	Aux Transit Volume RAL	@alight_ralsline	Alightings RAL
@volax_wcesline	Aux Transit Volume WCE	@alight_wcesline	Alightings WCE
@volax_totsline	Aux Transit Volume TOT	@alight_totsline	Alightings TOT

3.Aggregate O-D result matrices and export to CSV: \rtm\RTM\\Outputs\transit_sline_gy.csv)

transit link/line selection

The transit link/line can be selected by setting up the trip components. There are multiple ways to select the trip component. One or more trip components can be taken into account:

• For in-vehicle trip component, the selections can be link (@selectsegment), transit line (@selectline), or segment (@selectlink).
• Check the box if the trip component is an auxiliary transit link
• For boarding or alighting trip component, the selections is node attribute(@selectinboanode,@selectfialinode @selecttrboanode @selecttralinode ). Check one box from the node component upon the purpose.

Examples

These are some examples related to trip component selection, and path selection threshold.

Case 1: The people who are in-vehicle in segment A Case 2: The people who in-vehicle in segment A OR walk on link B Case 3: The people who in-vehicle in segment A AND walk on link B Case 4: The people who in-vehicle in segment A AND walk on link B AND transfer boarding at node C Case 5: The people who take any segment D of a transit line Case 6: The people who initially board at node C AND finally alight at node E