Performance Measures via Cloud-based System Architecture Chicago - - PowerPoint PPT Presentation

▶
performance measures
SMART_READER_LITE
LIVE PREVIEW

Performance Measures via Cloud-based System Architecture Chicago - - PowerPoint PPT Presentation

Dec 20, 2022 196 likes •605 views

Automated Traffic Signal Performance Measures via Cloud-based System Architecture Chicago Regional Automated Traffic Signal Performance Measure Workshop 19 February 2019 Essential Performance Measures for Improved Arterial Roadway Management

slide-1
SLIDE 1

Automated Traffic Signal Performance Measures

via Cloud-based System Architecture

Chicago Regional Automated Traffic Signal Performance Measure Workshop 19 February 2019

slide-2
SLIDE 2

Essential Performance Measures for Improved Arterial Roadway Management

”Queue length processing is the only way to calculate the accurate wait time of vehicles at traffic lights and deliver ALL the data needed for generating accurate travel times on signalized arterial corridors…”

Liu, H., Ma, W., and Wu, X., (2009) Traffic Flow Monitoring for Intersections with Signal Control U.S. Patent No. 8,279,086 Awarded by USPTO October 2, 2012

2

slide-3
SLIDE 3

LTD’s Traffic Signal Performance Measurement System History

Henry Liu, Heng Hu developed real-time Arterial Performance Monitoring System at U of Minnesota Deployed by MnDOT & City of Pasadena, CA 3rd party validation by Alliant Eng / Iteris Live Traffic Data sub-licenses technology to further develop the system NY Headquarters created Development of ATSPM accelerates National rollout begins Traffic Signal Prediction Algorithm developed Cloud-based Signal Performance Measure software (SIGPAT™*) released Full agency deployments accelerate Signal Prediction Algorithm improved by AI models

*SIGPAT™ = Signal Performance Analysis Toolbox

2006-2011 2014 2015 2016 2017

3

SIGPAT integrated with BT / Wi-Fi Travel Time New dashboards launched

2018+

slide-4
SLIDE 4

Automakers

  • Data is collected via central software or a

device installed inside the traffic signal cabinet

  • Data is analyzed and provided to

traffic agencies through LTD’s performance measurement software (SIGPAT™)

Traffic & Signal Data Collection and Distribution

Real-Time, Predictive & Historical

CUSTOMER

Reliable • Scalable • Compatible Secure • Robust Traffic Signal Infrastructure Traffic Management Center LTD Server

4

slide-5
SLIDE 5

5

Traffic Volumes

Real Time & Historical Datasets Available For Analytics

Vehicle Delays Signal Phase and Timing Data Arterial Travel Times

Other available datasets:

▪ Queue Length ▪ Pre-emption ▪ Approach speed ▪ Vehicle Delay ▪ Vehicle Trajectory ▪ Saturation Flow Rate ▪ Etc

5

slide-6
SLIDE 6

6

Real Time & Predictive SPaT Applications

slide-7
SLIDE 7

Signal/Traffic Data Collection Methods - 1/4

  • Data Feed from Central Software

▪ Trafficware: ATMS.Now ▪ Transcore: TransSuite ▪ McCain: Transparity ▪ Intelight: Maxview ▪ Econolite: Centracs

▪ …

7

slide-8
SLIDE 8

Signal/Traffic Data Collection Methods - 2/4

✓ Econolite ✓ Siemens ✓ TrafficWare ✓ Peek ✓ Intelight

  • Based on Built-in Data Loggers of Signal Controllers

▪ Fetches logs via FTP in scheduled intervals (e.g. 15 mins) ▪ Only available in newer controllers & firmware

8

slide-9
SLIDE 9

Signal/Traffic Data Collection Methods - 3/4

  • Standards-based Communication Protocol

▪ NTCIP , AB3418, … ▪ Live Traffic Data’s Software Approach

9

slide-10
SLIDE 10

Signal/Traffic Data Collection Methods - 4/4

▪ Econolite ▪ Siemens ▪ McCain ▪ Peek ▪ Trafficware (Naztec) ▪ Intelight ▪ … ▪ Loop ▪ Video ▪ Radar ▪ Magnetometer ▪ Hybrid

NEMA, 170/2070, …

Supports All Controller Types Supports All Detector Types ▪ Collects signal/det data ▪ Fiber / Ethernet / serial & wireless communication environments ▪ Deployed since 2007 ▪ Collects data, cabinet health ▪ Fiber / Ethernet or cellular comm ▪ Wifi/BT for ARID Travel-time HARDWARE APPROACH

10

slide-11
SLIDE 11

Traffic Signal Data & Performance Measures from Intersection to Stakeholders

Agency Firewall

Distribute Traffic Signal Data

Local Server

Traffic Engineer

TCP/IP HTTPS

Collect Traffic Signal Data

11

Traffic Signal Data Server Performance Measure Webserver Application Server

Traffic Engineer

API HTTPS

Customer Backend Server

Four methods to access traffic signal data

slide-12
SLIDE 12

ATSPM Data Requirements

Data Requirement Single Intersection Metric Corridor Metric Signal Only Green Time, Split Monitor, Phase Duration Green Band Analysis Signal, Vehicle Detection Queue Length, Delay, Stops, Percent Arrival on Green, Purdue Coordination Diagram(PCD), Performance Index Time-Space Diagram, Arterial Congestion Map, Vehicle Trajectories & Travel-time (Virtual Probe) Vehicle Detection Only Volumes Communication, Real-time ATSPM Real-time alerts, SPaT Real-time Performance Map

12

slide-13
SLIDE 13

Real-time Map of Signal Performance

13

slide-14
SLIDE 14

Real-time Notification of Service Anomalies

  • Real-time alert through E-mail, via Text message
  • User-configurable performance measures

14

slide-15
SLIDE 15

Case Study: Olathe, KS -- 119th St @ Blackbob Rd

15

EB Traffic Volumes

slide-16
SLIDE 16

Olathe KS – 119th @ Blackbob

16

slide-17
SLIDE 17

Olathe, KS Case Study – 119th St Corridor

17

  • New signal timings implemented on May 7th 2018
slide-18
SLIDE 18

EB 119th @ Blackbob: PCD Comparison

18

AoG % 66% 72% 72% 55% Before: 82% 52% 75% 76% AoG % After:

slide-19
SLIDE 19

Weekly Performance Index (Apr 23 – Aug 3) at 119th & Blackbob

Control Change 7 May 2018

19

slide-20
SLIDE 20

Backup details for Perf Index (prev slide)

slide-21
SLIDE 21

EB Travel-time: 119th--Renner to Blackbob

21

Travel-Time (sec) 15% 154 158 50% 195 188 85% 243 201

slide-22
SLIDE 22

Green Band Analysis – 119th Corridor (7am to 8:10am)

22

Before (May 1, 2, 3) After (June 5, 6, 7)

slide-23
SLIDE 23

Arterial Congestion Heat Map

23

  • Ave. Delay / vehicle / 15-min

7 am 7 pm

▪ To better understand how congestion evolves ▪ Different cell colors indicate different congestion levels ▪ Easily identify the most and least congested time and location

Palo Alto, CA (Sandhill Rd)

slide-24
SLIDE 24

Arterial Performance - Time-Space Diagram

Field-data driven T-S Diagram

▪ Evaluate the actual performance ▪ Actual vehicle arrival and queuing base on detector inputs ▪ Fine-tune coordination

(TH 13, MN)

24

slide-25
SLIDE 25

Case Study: Palo Alto Traffic Doubles on Stanford Game Day

25

STANFORD UNIV VS. UCLA Kick-Off 7:35 PM

7 AM 12 AM GAME-TIME

400 600 250 250

30-min Vol Sep 16 vs. Sep 23

EB WB

slide-26
SLIDE 26

26

Travel-Time, O-D using Wi-Fi or BT via ARID (Anonymous Re-Identification)

Estimate travel time and origin-destination on signalized arterial roadway networks

slide-27
SLIDE 27

Regional Access to Performance Measures

27 Metro Server City Server State DOT Server County Server 40 TSI 3 TSI 8 TSI 5 TSI

Metro Workstn State DOT

Workstn City Workstn

County Workstn

56

LTD Arterial Performance Measure Cloud Server

56 TSI 56 TSI 56 TSI 56 TSI

slide-28
SLIDE 28

▪ Budget friendly ATSPM, including hardware, software, support ▪ Facilitate regional sharing of ATSPM for signal synchronization or other traffic engineering projects

▪ Enable smart use of existing signal infrastructure

  • Not required to upgrade controller equipment
  • Not required to revise detector configuration
  • I2V now
  • Enhance Smart City Mobility

▪ Automate signal performance monitoring and fine

tuning

  • Real-time alerts
  • Periodic reports

Benefit to Traffic Agencies

28

slide-29
SLIDE 29

Automated Traffic Signal Performance Measures (ATSPMs)

  • FHWA, (https://www.fhwa.dot.gov/innovation/everydaycounts/edc_4/factsheet/automated_traffic_signal.pdf)

29

slide-30
SLIDE 30

Questions?

THANK YOU FOR YOUR ATTENTION!

FOR MORE INFORMATION CONTACT : info@livetrafficdata.com 646.569.5785 Online Contact Form: www.livetrafficdata.com/agencies

slide-31
SLIDE 31
slide-32
SLIDE 32

Independent evaluations

LTD Data Accuracy Validation

slide-33
SLIDE 33

TH 55 (6 Intersections)

Conducted by Alliant Engineering, Inc. July 2009

LTD Data Accuracy Validation

Independent Evaluation: TH 55, Golden Valley, MN

33

slide-34
SLIDE 34

200 200 400 400 600 600 800 800 1000 1000 6:57:36 7:12: 2:00 00 7:26: 6:24 24 7:40:48 7:55:12 8:09: 9:36 36 8:24:00

July 22nd for TH55WB at Rhode Island nd Inters rsect ctio ion n (AM)

MaxQL MaxQL-Estimation MaxQL MaxQL-Ob Obse serv rvat atio ion

200 400 600 800 6:57:36 7:04:48 7:12:00 7:19:12 7:26:24 7:33:36 7:40:48 7:48:00 7:55:12 8:02:24 July 23rd for TH55WB at Rhode Island Intersection (AM)

MaxQL-Estimation MaxQL-Observation

LTD Data Accuracy Validation

Queue Length Comparison

34

slide-35
SLIDE 35

100 100 200 200 300 300 400 400 100 100 200 200 300 300 400 400 Observation (seconds) Estimation (seconds) Travel Time Estimation vs. Observation(July 22 & 23)

+10 10%

  • 10

10%

LTD Data Accuracy Validation

Travel Time Comparison

35

slide-36
SLIDE 36

Conducted by Iteris, Inc. October 2011

Orange Grove Blvd (6 Intersections)

LTD Data Accuracy Validation

Independent Evaluation: Orange Grove Blvd, Pasadena, CA

36

slide-37
SLIDE 37

LTD Data Accuracy Validation

Queue Length Result

37

slide-38
SLIDE 38

LTD Estimated results

  • vs. Measured results

LTD Data Accuracy Validation

Travel Time Results

38

slide-39
SLIDE 39

Regional Access to Performance Measures

Solution

Metro Server City Server State DOT Server County Server 40 TSI 3 TSI 8 TSI 5 TSI

Metro Workstn State DOT Workstn City Workstn County Workstn

56

LTD Arterial Performance Measure Cloud Server

56 TSI 56 TSI 56 TSI 56 TSI