Transformative Planning Partnerships and Big Data AMPO 2017 Annual - - PowerPoint PPT Presentation

▶
transformative planning partnerships and big data
SMART_READER_LITE
LIVE PREVIEW

Transformative Planning Partnerships and Big Data AMPO 2017 Annual - - PowerPoint PPT Presentation

Feb 29, 2024 111 likes •412 views

Transformative Planning Partnerships and Big Data AMPO 2017 Annual Conference Presentation Agenda Introduction/Presentation Summary Defining Big Data CAT and MPO Partnership Data Purchasing Process Benefits (sample size,

slide-1
SLIDE 1

Transformative Planning Partnerships and Big Data

AMPO

2017 Annual Conference

slide-2
SLIDE 2

Presentation Agenda  Introduction/Presentation Summary  Defining Big Data  CAT and MPO Partnership

‒ Data Purchasing Process ‒ Benefits (sample size, etc.)

 How each agency used the data  Lessons Learned  Next Steps / Applying the Results  Questions / Discussion

AMPO – 2017 Fall Conference 2

slide-3
SLIDE 3

Introduction

AMPO – 2017 Fall Conference 3

 Transformative advantages of Partnerships

and Big Data

‒ Recommendations to enhance the operational efficiency of the transit system ‒ Analysis of travel patterns needed for congestion management process update.

slide-4
SLIDE 4

Regional Overview

 Regional Population 276,406; 50% in Savannah  Areas included: All of Chatham County, Richmond Hill (Bryan County) and

portions of Effingham County

 Chatham Area Transit(CAT) operates 16 core routes within Savannah and

portions of Chatham County; including shuttle and Belles Ferry

 The largest single container terminal & the fastest growing port in the US  Significant infill and decentralized suburban growth pressures  Historic Preservation: Nations largest Historic Landmark District  Tourism: 13.7 Million Annual visitors  Hunter Army Airfield & Fort Stewart

AMPO – 2017 Fall Conference 4

slide-5
SLIDE 5

big da·ta

noun COMPUTING

 extremely large data sets that may be analyzed computationally to

reveal patterns, trends, and associations, especially relating to human behavior and interactions.

Source: dictionary.com

Defining Big Data

NPMRDS

slide-6
SLIDE 6

Defining Big Data

slide-7
SLIDE 7

Data Procurement

Study area expanded to include external trip data.

‒ Screven ‒ Effingham ‒ Bulloch ‒ Bryan ‒ Liberty/Long ‒ McIntosh ‒ Jasper/Beaufort

*Counties selected based on census commute data.

slide-8
SLIDE 8

Data Procurement

 AirSage data purchase driven by the following criteria:

Option 1: Base Data Set

  • 1 month sample
  • Average weekdays (T‐Th)
  • AM Peak, Mid‐day, PM Peak, 24 hour total
  • 3‐class trip purposes: HBW, HBO, OBO
  • 2 Resident Classes: Visitors and Residents

390 Zones Option 2: Premium Data Set

  • 1 month sample
  • Average weekdays (T‐Th)
  • AM Peak, Mid‐day, PM Peak, 24 hour total
  • 3‐class trip purposes: HBW, HBO, OBO
  • 6 Resident Classes: Res Worker, Home

Worker, In‐Commuter, Out‐Commuter, Short‐Term Vis, Lg‐Term Vis)

230 Zones

slide-9
SLIDE 9

Data Purchase Partnership

 Premium Data Set

  • 2 months sample
  • Average weekdays (T-T)
  • AM Peak, Mid-day, PM Peak,

24 hour total

  • 3-class trip purposes: HBW,

HBO, OBO

  • 6 Resident Classes: Res

Worker, Home Worker, In- Commuter, Out-Commuter, Short-Term Vis, Lg-Term Vis)

 242 Zones

slide-10
SLIDE 10

DATA APPLIED

slide-11
SLIDE 11

Applying the Data

14

Travel Patterns

AirSage travel patterns Census household / employment data CAT ridership data – stop & route level CORE Data Source 1 CORE Data Source 2

CORE Congestion Management Process CORE Congestion Management Process CAT Origin Destination Analysis CAT Origin Destination Analysis

NPMRDS

slide-12
SLIDE 12

Analysis Process

15

  • AirSage data contains

various details, including trip purpose, time of day, and subscriber types (resident, visitor, etc.)

  • Analysis consisted of

73 possible data combinations that were each mapped / assessed.

  • Travel behaviors were

layered with transit and census data.

Category (April 2016) Aggregate Percentage Category (October 2015) Aggregate Percentage 24‐Hour Counts* 1,445,268 N/A 24‐Hour Counts* 1,452,606 N/A Day Period Counts** 1,159,165 100% Day Period Counts** 1,562,488 100% AM Peak ‐ All Trips/Resident Classes 271,771 23.45% AM Peak ‐ All Trips/Resident Classes 367,454 23.52% Mid‐Day Peak ‐ All Trips/Resident Classes 560,973 48.39% Mid‐Day Peak ‐ All Trips/Resident Classes 744,792 47.67% PM Peak ‐ All Trips/Resident Classes 326,420 28.16% PM Peak ‐ All Trips/Resident Classes 450,241 28.82% Home Worker ‐ All Trips 149,169 12.87% Home Worker ‐ All Trips 161,587 10.34% Home Worker ‐ AM Peak/All Trips 31,573 2.72% Home Worker ‐ AM Peak/All Trips 32,651 2.09% Home Worker ‐ Mid‐Day Peak/All Trips 78,387 6.76% Home Worker ‐ Mid‐Day Peak/All Trips 84,542 5.41% Home Worker ‐ PM Peak/All Trips 39,209 3.38% Home Worker ‐ PM Peak/All Trips 44,394 2.84% Home Worker ‐ AM Peak/HBO 29,637 2.56% Home Worker ‐ AM Peak/HBO 30,425 1.95% Home Worker ‐ Mid‐Day Peak/HBO 66,671 5.75% Home Worker ‐ Mid‐Day Peak/HBO 71,471 4.57% Home Worker ‐ PM Peak/HBO 33,381 2.88% Home Worker ‐ PM Peak/HBO 37,498 2.40% Home Worker ‐ AM Peak/NHB 1,936 0.17% Home Worker ‐ AM Peak/NHB 2,226 0.14% Home Worker ‐ Mid‐Day Peak/NHB 11,716 1.01% Home Worker ‐ Mid‐Day Peak/NHB 13,072 0.84% Home Worker ‐ PM Peak/NHB 5,828 0.50% Home Worker ‐ PM Peak/NHB 6,895 0.44% Resident Worker ‐ All Trips 388,418 33.51% Resident Worker ‐ All Trips 385,055 24.64% Resident Worker ‐ AM Peak/All Trips 92,669 7.99% Resident Worker ‐ AM Peak/All Trips 91,100 5.83% Resident Worker ‐ Mid‐Day Peak/All Trips 177,336 15.30% Resident Worker ‐ Mid‐Day Peak/All Trips 173,407 11.10% Resident Worker ‐ PM Peak/All Trips 118,413 10.22% Resident Worker ‐ PM Peak/All Trips 120,548 7.72% Resident Worker ‐ AM Peak/HBW 31,077 2.68% Resident Worker ‐ AM Peak/HBW 33,827 2.16% Resident Worker ‐ Mid‐Day Peak/HBW 30,203 2.61% Resident Worker ‐ Mid‐Day Peak/HBW 34,994 2.24% Resident Worker ‐ PM Peak/HBW 23,769 2.05% Resident Worker ‐ PM Peak/HBW 25,523 1.63% Resident Worker ‐ AM Peak/HBO 26,220 2.26% Resident Worker ‐ AM Peak/HBO 25,125 1.61% Resident Worker ‐ Mid‐Day Peak/HBO 42,950 3.71% Resident Worker ‐ Mid‐Day Peak/HBO 42,561 2.72% Resident Worker ‐ PM Peak/HBO 35,222 3.04% Resident Worker ‐ PM Peak/HBO 36,887 2.36% Resident Worker ‐ AM Peak/NHB 35,372 3.05% Resident Worker ‐ AM Peak/NHB 32,148 2.06% Resident Worker ‐ Mid‐Day Peak/NHB 104,183 8.99% Resident Worker ‐ Mid‐Day Peak/NHB 95,851 6.13% Resident Worker ‐ PM Peak/NHB 59,422 5.13% Resident Worker ‐ PM Peak/NHB 58,138 3.72% Outbound Commuter ‐ All Trips 27,832 2.40% Outbound Commuter ‐ All Trips 34,851 2.23% Outbound Commuter ‐ AM Peak/All Trips 7,581 0.65% Outbound Commuter ‐ AM Peak/All Trips 8,864 0.57% Outbound Commuter ‐ Mid‐Day Peak/All Trips 11,414 0.98% Outbound Commuter ‐ Mid‐Day Peak/All Trips 15,136 0.97% Outbound Commuter ‐ PM Peak/All Trips 8,836 0.76% Outbound Commuter ‐ PM Peak/All Trips 10,851 0.69% Outbound Commuter ‐ AM Peak/HBW 2,438 0.21% Outbound Commuter ‐ AM Peak/HBW 2,630 0.17% Outbound Commuter ‐ Mid‐Day Peak/HBW 2,657 0.23% Outbound Commuter ‐ Mid‐Day Peak/HBW 2,898 0.19% Outbound Commuter ‐ PM Peak/HBW 1,857 0.16% Outbound Commuter ‐ PM Peak/HBW 1,936 0.12% Outbound Commuter ‐ AM Peak/HBO 2,129 0.18% Outbound Commuter ‐ AM Peak/HBO 2,340 0.15% Outbound Commuter ‐ Mid‐Day Peak/HBO 2,284 0.20% Outbound Commuter ‐ Mid‐Day Peak/HBO 3,162 0.20% Outbound Commuter ‐ PM Peak/HBO 2,408 0.21% Outbound Commuter ‐ PM Peak/HBO 2,963 0.19% Outbound Commuter ‐ AM Peak/NHB 3,014 0.26% Outbound Commuter ‐ AM Peak/NHB 3,895 0.25% Outbound Commuter ‐ Mid‐Day Peak/NHB 6,473 0.56% Outbound Commuter ‐ Mid‐Day Peak/NHB 9,076 0.58% Outbound Commuter ‐ PM Peak/NHB 4,572 0.39% Outbound Commuter ‐ PM Peak/NHB 5,951 0.38%

slide-13
SLIDE 13

Origins

16

Travel Time and Type of Traveler:

  • Morning (7:00 AM – 10:00 AM)
  • “Home” Based “Work”
  • “Resident Worker”

Origins and Desire Lines

Dataset: April 2016

slide-14
SLIDE 14

Destinations

17

Travel Time and Type of Traveler

  • Morning (7:00 AM – 10:00

AM)

  • “Home” Based “Work”
  • “Resident Worker”

Destinations and Desire Lines

Dataset: April 2016

slide-15
SLIDE 15

Destinations – Incoming Commuters

18

Travel Time and Type of Traveler

  • Morning (7:00 AM – 10:00 AM)
  • “Home” Based “Work”
  • “Inbound Commuter”

Destinations and Desire Lines

Dataset: April 2016

slide-16
SLIDE 16

Sample: Applied Analysis

Boardings

‒ Ridership: > 50,000 monthly riders ‒ Primary boarding locations

  • Transit Center: 500+ daily
  • Savannah Mall & Oglethorpe

Mall: ~140 daily

Key Areas Served / Major Destinations

‒ Oglethorpe Mall area ‒ St. Joseph’s Hospital ‒ Armstrong State University ‒ Walmart on Fulton Rd. ‒ Savannah Mall

High propensity

Primary Transfers

‒ To Route 14 From: 3, 6, 25 ‒ From Route 14 To: 3, 6, 25, 27, 31 ‒ Transfer Location: Transit Center

Origins and Destinations

‒ Route activity is dominated by downtown trips.

slide-17
SLIDE 17

Sample: Recommendations

Recommendations:

Enhance frequency between Oglethorpe Mall and Downtown Transfer Center.

Phase I

‒ Split Route 14 at Oglethorpe Mall ‒ Utilize 4 peak buses on North leg of route to improve frequency to downtown. ‒ Maintain service on South leg of route using remaining 2 vehicles.

Phase II

‒ Add frequency on North leg of Abercorn using new electric buses.

Route Analysis Synopsis:

Significant demand between downtown and Oglethorpe Mall area, which is the primary

  • pportunity for trips coming from West side of town

to get to the desired mall area zone most quickly/directly.

Only approx. 20% of transit trips originating at the Downtown Station continue beyond the Oglethorpe Mall.

Roughly 60% of trips originating south of Oglethorpe Mall continue into the downtown area.

Recommendation Results:

Service Hours Increase: Maintain current service hours using existing buses.

Frequency Impacts: Extended 15 minute headways

  • n North alignment
slide-18
SLIDE 18

General Findings

21

 AirSage data provided insights into areas where transit service

could be considered and service streamlining.

 AirSage data provided confirmation where routes are

underperforming due to travel behavior discrepancies when compared to service lines.

‒ i.e. ridership performance is not projected to improve with reasonable route modification.

 Data highlighted critical destinations for local travel.  Much travel occurs within the region and

jobs/shopping/services located on the outer perimeter of the study area.

 Travel patterns are primarily North and South oriented inside

the urban core.

slide-19
SLIDE 19

Congestion Management Process

slide-20
SLIDE 20

CMP Overview

Objectives

 Develop congestion management measures.  Reduce non‐recurring congestion duration.  Evaluate travel time reliability to 95th percentile.  Consider the full range of congestion management strategies.  Acceptable Level of Service (LOS).

slide-21
SLIDE 21

Sample: AM Commute

AMPO – 2017 Fall Conference 25

slide-22
SLIDE 22

External Trip Makers

slide-23
SLIDE 23

Functional Class by Zone

slide-24
SLIDE 24

Strategies

AMPO – 2017 Fall Conference 28

slide-25
SLIDE 25

LESSONS LEARNED

slide-26
SLIDE 26

Lessons Learned

AMPO – 2017 Fall Conference 30

 We have BIG DATA, now what?

‒ Data quality control ‒ Readiness ‒ Storage ‒ Training ‒ Local data sets to pair it with ‒ Staff, timing, resources

  • Consultant support
slide-27
SLIDE 27

NEXT STEPS

slide-28
SLIDE 28

Next Steps

AMPO – 2017 Fall Conference 32

 We have BIG DATA, now what?

‒ Resource to inform the decision makers ‒ CAT applying the results to the Transit Development Plan Update ‒ MPO applying it to the Long Range Plan and future CMP updates

slide-29
SLIDE 29

Questions?

33