Work Share Session 3

Interesting work shared from multiple different projects.

This content was presented to Nelson\Nygaard Staff at a Lunch and Learn webinar on Thursday, September 16th, 2021, and is available as a recording here and embedded below.

Agenda

• Michael Carraher (15 min)
• Ulises Hernandez (15 min)
• Bryan Blanc (15 min)
• Amanda Wolfe (15 min)

Michael Carraher: Travel Flow Analysis & Transfer Analysis QA/QC

Description:

• Travel flow analysis: Convert a matrix of TAZ-to-TAZ trip flows into long form, consolidate into larger groups (to minimize number of lines), and output result to make spider lines in GIS.
• Transfer analysis QAQC: For SEPTA, transfers are valid within a 2 hour window. We have data that contains the ‘origin’ and ‘destination’ stops of a transfer. This script was a way for us to calculate the distance between endpoints of a transfer, play around with what we thought was a realistic transfer distance, and map the results to determine if we wanted to make a pretty map of it.

Relevant Links:

• Travel flow analysis: G:\Current\TARTA_COA_2021.0070\R\travel_flows.R
• Transfer analysis QAQC: G:\Current\SEPTA_Bus_Network_Redesign_2019_0467\R\transfers_qaqc.R

Ulises Hernandez: ACS Demographic Downloading and Mapping

Description: I will show how to pull several ACS demographic variables to develop an equity index for the city of Bellevue WA. This example uses relatively easy concepts that beginner users can apply to download, visualize and create different indexes quickly.

Relevant Links: G:\Current\BELLEVUE_Mobility_Implementation_Plan_2020_1022\R\pull_equity_layers_and_webmap.R

Bryan Blanc: HandyDART Ridership Modeling

Description: We developed a ridership model in November - December 2020 by Nelson\Nygaard for Translink for the purposes of estimating future HandyDART (paratransit) service demand and supply needs. The model incorporated multiple steps of Monte Carlo simulation for the different factors that could vary in ridership demand and capacity, and considered multiple scenarios of ridership recovery. We will be updating the model soon with more data from 2021.

Relevant Links:

• GitHub repository: https://github.com/PerkinsAndWill/HandyDART
• Final Report website: https://perkinsandwill.github.io/HandyDART/

Amanda Wolfe: Pulling text from Word into Shiny

Description: The Route Profile Shiny Dashboards help to display a route’s performance through a series of maps, charts, and tables for a transit system. Recently, we worked on a solution to incorporate more context into the dashboard through adding the text written by planners who are assessing the data for each route into the dashboard. This provides both the data visualizations and interpretation of the data in a consolidated location to create a more complete deliverable for a client.

A process description is provided below:

• 1. Create a tracking excel sheet for Route Evaluation Process to stay organized throughout and be able to track versioning and updates across firms
  • Sharepoint use works well for collaboration
• 2. Create a template in word that will be consistent with the dashboard and for each route
  • Styles and formatting are important
• 3. Copy the word template using R script (generate-ms-word-template-copies)- to create document for each route
• 4. Complete route profiles analysis (uploading docs to sharepoint)
• 5. Create R scripts to be able to load text in Shiny

  • save-ms-word-content-to-db: loop to organize and load each Word doc into a SQL db

  • global: Call the table object from the db

  • server:

    • Create a reactive to filter text for each route

    • Create renderUI objects for each text section

  • ui: Call the htmlOutput in order

• 6. Scan documents for style/formatting inconsistencies that will prevent upload

  • Extra paragraph “enters”

  • Different text styles

  • Default word autocorrects that are not in a style (ex- 3rd)|

  • Track changes and comments are actually ok- but better to resolve and save “clean” copies to avoid confusion

• 7. Copy documents so no other users have open (this prevents uploading)
• 8. Use R script (save-ms-word-content-to-db) to load text into database and mark versioning in tracking sheet.

Relevant Links:

• Code: https://github.com/PerkinsAndWill/capmetro_routeprofiles
• Package: https://davidgohel.github.io/officer/

This content was presented to Nelson\Nygaard Staff at a Lunch and Learn webinar on September 16th, 2021, and is available as a recording here and embedded at the top of the page.