Performance Measure Summary - Louisville-Jefferson County KY-IN. Performance Measures and Definition of Terms

Similar documents
Performance Measure Summary - Large Area Sum. Performance Measures and Definition of Terms

Performance Measure Summary - Medium Area Sum. Performance Measures and Definition of Terms

Performance Measures and Definition of Terms

Performance Measure Summary - Austin TX. Performance Measures and Definition of Terms

Performance Measure Summary - Pittsburgh PA. Performance Measures and Definition of Terms

Performance Measure Summary - New Orleans LA. Performance Measures and Definition of Terms

Performance Measure Summary - Portland OR-WA. Performance Measures and Definition of Terms

Performance Measure Summary - Oklahoma City OK. Performance Measures and Definition of Terms

Performance Measure Summary - Seattle WA. Performance Measures and Definition of Terms

Performance Measure Summary - Buffalo NY. Performance Measures and Definition of Terms

Performance Measure Summary - Fresno CA. Performance Measures and Definition of Terms

Performance Measure Summary - Hartford CT. Performance Measures and Definition of Terms

Performance Measure Summary - Boise ID. Performance Measures and Definition of Terms

Performance Measure Summary - Tucson AZ. Performance Measures and Definition of Terms

Performance Measure Summary - Wichita KS. Performance Measures and Definition of Terms

Performance Measure Summary - Spokane WA. Performance Measures and Definition of Terms

Performance Measure Summary - Grand Rapids MI. Performance Measures and Definition of Terms

Performance Measure Summary - Washington DC-VA-MD. Performance Measures and Definition of Terms

Performance Measure Summary - Charlotte NC-SC. Performance Measures and Definition of Terms

Performance Measure Summary - Toledo OH-MI. Performance Measures and Definition of Terms

Performance Measure Summary - Pensacola FL-AL. Performance Measures and Definition of Terms

Performance Measure Summary - Omaha NE-IA. Performance Measures and Definition of Terms

Performance Measure Summary - Allentown PA-NJ. Performance Measures and Definition of Terms

Performance Measure Summary - Nashville-Davidson TN. Performance Measures and Definition of Terms

Performance Measure Summary - Corpus Christi TX. Performance Measures and Definition of Terms

Performance Measure Summary - Boston MA-NH-RI. Performance Measures and Definition of Terms

Performance Measure Summary - El Paso TX-NM. Performance Measures and Definition of Terms

Performance Measure Summary - Minneapolis-St. Paul MN-WI. Performance Measures and Definition of Terms

Performance Measure Summary - New York-Newark NY-NJ-CT. Performance Measures and Definition of Terms

Technical Memorandum Analysis Procedures and Mobility Performance Measures 100 Most Congested Texas Road Sections What s New for 2015

Metropolitan Freeway System 2013 Congestion Report

2016 Congestion Report

Signal System Timing and Phasing Program SAMPLE. Figure 1: General Location Map. Second St.

5. OPPORTUNITIES AND NEXT STEPS

Appendix SAN San Diego, California 2003 Annual Report on Freeway Mobility and Reliability

Alpine Highway to North County Boulevard Connector Study

Jeff s House. Downtown Charlottesville. PEC Office

Executive Summary. Draft Environmental Impact Statement/Environmental Impact Report ES-1

MPO Transit Study. Transit Concept for 2050 November 5, Transit Technologies

New Jersey Pilot Study Testing Potential MAP-21 System Performance Measures

VILLASPORT ATHLETIC CLUB AND SPA Draft Environmental Impact Report. Volume I

Reducing Energy Consumption and Emissions Through Congestion Management

Metropolitan Freeway System 2007 Congestion Report

2017 Mobility Report

Interstate Operations Study: Fargo-Moorhead Metropolitan Area Simulation Output

Operate a Reliable and Convenient. Becky Allmeroth, State Maintenance Engineer. Tracker. Measures of Departmental Performance

Act 229 Evaluation Report

Facts and Figures. October 2006 List Release Special Edition BWC National Benefits and Related Facts October, 2006 (Previous Versions Obsolete)

4 COSTS AND OPERATIONS

Interstate Operations Study: Fargo-Moorhead Metropolitan Area Simulation Results

Valley Metro Overview. ITE/IMSA Spring Conference March 6, 2014

Michigan/Grand River Avenue Transportation Study TECHNICAL MEMORANDUM #18 PROJECTED CARBON DIOXIDE (CO 2 ) EMISSIONS

Trip Generation & Parking Occupancy Data Collection: Grocery Stores Student Chapter of Institute of Transportation Engineers at UCLA Spring 2014

Vehicle Miles Traveled in Massachusetts: Who is driving and where are they going?

Caltrain Downtown Extension Study Ridership Forecast Summary

RTCSNV CRASH ANALYSIS REPORT

TRANSPORTATION STUDY FOR THE 8899 BEVERLY BOULEVARD PROJECT

Spatial and Temporal Analysis of Real-World Empirical Fuel Use and Emissions

NEW YORK SUBURBAN RAIL SUMMARY (COMMUTER RAIL, REGIONAL RAIL)

QUALITY OF LIFE EXECUTIVE SUMMARY REPORT I O N S TAT I O N

CLRP. Performance Analysis of The Draft 2014 CLRP. Long-Range Transportation Plan For the National Capital Region

2030 Multimodal Transportation Study

Click to edit Master title style

KANSAS Occupant Protection Observational Survey Supplementary Analyses Summer Study

APPENDIX I: [FIXED-GUIDEWAY TRANSIT FEASIBILITY]

Congestion Management. SFMTA Board Annual Workshop January 29, 2019

Comprehensive Regional Goods Movement Plan and Implementation Strategy Goods Movement in the 2012 RTP/SCS

Subarea Study. Manning Avenue (CSAH 15) Corridor Management and Safety Improvement Project. Final Version 1. Washington County.

ConnectGreaterWashington: Can the Region Grow Differently?

Final Report. Investigating the Effect of Freeway Congestion Thresholds on Decision-Making Inputs. Tongbin Qu

EXECUTIVE SUMMARY. The following is an outline of the traffic analysis performed by Hales Engineering for the traffic conditions of this project.

1 On Time Performance

TEXAS CITY PARK & RIDE RIDERSHIP ANALYSIS

Abstract. Executive Summary. Emily Rogers Jean Wang ORF 467 Final Report-Middlesex County

Craigieburn Employment Precinct North and English Street

RTID Travel Demand Modeling: Assumptions and Method of Analysis

Hillsborough County MPO Transit Study. Transit Concept for 2050 October 17, 2007

Evaluation of Renton Ramp Meters on I-405

Operating & Maintenance Cost Results Report

state, and federal levels, complete reconstruction and expansion of I35 in the near future is not likely.

2012 Air Emissions Inventory

REPORT CARD FOR CALIFORNIA S INFRASTRUCTURE WHAT YOU SHOULD KNOW ABOUT CALIFORNIA S TRANSIT FACILITIES

Online Appendix for Subways, Strikes, and Slowdowns: The Impacts of Public Transit on Traffic Congestion

HOT Lanes: Congestion Relief and Better Transit

TPB CLRP Aspirations Scenario

Proposed location of Camp Parkway Commerce Center. Vicinity map of Camp Parkway Commerce Center Southampton County, VA

Sample Validation of Vehicle Probe Data Using Bluetooth Traffic Monitoring Technology

Summary of Potential Target Formula Scenarios

Project Title: Using Truck GPS Data for Freight Performance Analysis in the Twin Cities Metro Area Prepared by: Chen-Fu Liao (PI) Task Due: 7/31/2013

Annex 10: Equations used for Direct Cost Calculation

PHILADELPHIA SUBURBAN RAIL SUMMARY (COMMUTER RAIL, REGIONAL RAIL)

Dulles Corridor Air Rights Study Investigation

Table of Contents INTRODUCTION... 3 PROJECT STUDY AREA Figure 1 Vicinity Map Study Area... 4 EXISTING CONDITIONS... 5 TRAFFIC OPERATIONS...

The Boston South Station HSIPR Expansion Project Cost-Benefit Analysis. High Speed Intercity Passenger Rail Technical Appendix

Summary FEBRUARY 2019

THE WILSHIRE CORRIDOR: RAIL AND ITS ALTERNATIVES. Prepared By: Jacki Murdock Transportation and Environmental Planner

The Case for. Business. investment. in Public Transportation

March 2, 2017 Integrating Transportation Planning, Project Development, and Project Programming

AUGUST 2015 URBAN MOBILITY. Scorecard

Disruptive Technology and Mobility Change

Transcription:

Performance Measure Summary - Louisville-Jefferson County KY-IN There are several inventory and performance measures listed in the pages of this Urban Area Report for the years from 1982 to 2014. There is no single performance measure that experts agree says it all. A few key points should be recognized by users of the Urban Mobility Scorecard data. Use the trends The multi-year performance measures are better indicators, in most cases, than any single year. Examining a few measures over many years reduces the chance that data variations or the estimating procedures may have caused a "spike" in any single year. (5 years is 5 times better than 1 year.) Use several measures Each performance measure illustrates a different element of congestion. (The view is more interesting from atop several measures.) Compare to similar regions Congestion analyses that compare areas with similar characteristics (for example, population, growth rate, road and public transportation system design) are usually more insightful than comparisons of different regions. (Los Angeles is not Peoria.) Compare ranking changes and performance measure values In some performance measures a small change in the value may cause a significant change in rank from one year to the next. This is the case when there are several regions with nearly the same value. (15 hours is only 1 hour more than 14 hours.) Consider the scope of improvement options Any improvement project in a corridor within most of the regions will only have a modest effect on the regional congestion level. (To have an effect on areawide congestion, there must be significant change in the system or service.) Performance Measures and Definition of Terms Travel Time Index A measure of congestion that focuses on each trip and each mile of travel. It is calculated as the ratio of travel time in the peak period to travel time in free-flow. A value of 1.30 indicates that a 20-minute free-flow trip takes 26 minutes in the peak. Planning Time Index A travel time reliability measure that represents the total travel time that should be planned for a trip. Computed with the 95th percentile travel time it represents the amount of time that should be planned for a commute trip to be late for only 1 day a month. If it is computed with the 80th percentile travel time it represents the amount of time that should be planned for a trip to be late for only 1 day a week. A PTI of 2.00 means that for a 20-minute trip in light traffic, 40 minutes should be planned. Peak Commuters Number of travelers who begin a trip during the morning or evening peak travel periods (6 to 10 a.m. and 3 to 7 p.m.). "Commuters" are private vehicle users unless specifically noted. per Commuter A yearly sum of all the per-trip delays for those persons who travel in the peak period (6 to 10 a.m. and 3 to 7 p.m.). This measure illustrates the effect of traffic slowdowns as well as the length of each trip. Total Delay The overall size of the congestion problem. Measured by the total travel time above that needed to complete a trip at free-flow speeds. The ranking of total delay usually follows the population ranking (larger regions usually have more delay). Free-Flow Speeds These values are derived from overnight speeds in the INRIX speed database. They are used as the national comparison thresholds. Other speed thresholds may be appropriate for urban project evaluations or sub-region studies. Excess Fuel Consumed Increased fuel consumption due to travel in congested conditions rather than freeflow conditions. Congestion Cost Value of travel delay for 2014 (estimated at $17.67 per hour of person travel and $94.04 per hour of truck time) and excess fuel consumption estimated using state average cost per gallon. Urban Area The developed area (population density more than 1,000 persons per square mile) within a metropolitan region. The urban area boundaries change frequently (every year for most growing areas), so increases include both new growth and development that was previously in areas designated as rural. Number of Rush Hours Time when the road system might have congestion.

Inventory Measures 2014 2013 2012 2011 2010 Population (1000s) 1,110 1,070 1,030 1,015 1,000 Rank 41 43 44 44 45 Commuters (1000s) 585 565 544 535 525 Freeway 12,020 11,877 12,310 12,250 12,160 Arterial Streets 9,178 9,069 9,430 9,384 9,315 Value of Time ($/hour) 17.67 17.39 17.14 16.79 16.30 Commercial Cost ($/hour) 94.04 89.60 89.56 86.81 88.12 Gasoline ($/gallon) 3.21 3.45 3.48 3.29 2.64 Diesel ($/gallon) 3.60 3.88 3.84 3.61 2.90 System Performance 2014 2013 2012 2011 2010 Congested Travel (% of peak VMT) 30 -- -- -- -- Congested System (% of lane-miles) 26 -- -- -- -- Congested Time (number of "Rush Hours") 3.60 -- -- -- -- Total Fuel (1000 gallons) 17,841 16,999 16,637 16,486 16,226 Rank 43 43 43 43 43 Fuel per Peak Auto Commuter (gallons) 22 21 20 20 20 Rank 23 27 36 30 27 Total Delay (1000s of person-hours) 35,622 33,942 33,219 32,916 32,398 Rank 45 45 45 45 45 Delay per Peak Auto Commuter (pers-hrs) 43 43 44 44 44 Rank 35 34 30 28 29 Travel Time Index 1.20 1.20 1.20 1.20 1.20 Rank 37 36 38 36 36 Commuter Stress Index 1.23 1.23 1.23 1.23 1.23 Rank 43 42 40 40 40 Freeway Planning Time Index (95th Pctile) 2.42 -- -- -- -- Rank 32 -- -- -- -- Total Cost ($ millions) 860 833 827 836 849 Rank 43 44 45 44 45 Cost per Peak Auto Commuter ($) 1,048 1,015 1,008 1,019 1,035 Rank 32 37 37 35 36

Inventory Measures 2009 2008 2007 2006 2005 Population (1000s) 990 980 970 960 950 Rank 44 44 45 45 46 Commuters (1000s) 518 511 502 494 485 Freeway 12,050 12,000 12,085 11,925 11,800 Arterial Streets 8,900 8,780 8,730 8,700 8,645 Value of Time ($/hour) 16.01 16.10 15.47 15.06 14.58 Commercial Cost ($/hour) 89.75 81.52 82.56 80.43 78.05 Gasoline ($/gallon) 2.18 3.43 2.92 2.56 2.24 Diesel ($/gallon) 2.48 4.05 3.21 2.73 2.38 System Performance 2009 2008 2007 2006 2005 Total Fuel (1000 gallons) 15,328 16,304 15,434 15,242 15,050 Rank 44 44 44 43 43 Fuel per Peak Auto Commuter (gallons) 19 20 19 19 18 Rank 35 30 41 33 39 Total Delay (1000s of person-hours) 30,605 32,553 30,817 30,434 30,049 Rank 45 45 47 44 45 Delay per Peak Auto Commuter (pers-hrs) 42 45 43 44 44 Rank 32 30 36 31 29 Travel Time Index 1.20 1.21 1.20 1.20 1.20 Rank 35 37 38 37 36 Commuter Stress Index 1.22 1.24 1.23 1.23 1.23 Rank 44 42 43 43 41 Total Cost ($ millions) 815 864 850 862 879 Rank 45 45 45 43 44 Cost per Peak Auto Commuter ($) 994 1,053 1,036 1,051 1,072 Rank 40 37 44 39 37

Inventory Measures 2004 2003 2002 2001 2000 Population (1000s) 940 910 875 860 850 Rank 46 47 46 46 45 Commuters (1000s) 477 460 435 421 410 Freeway 11,730 11,500 11,000 10,400 10,200 Arterial Streets 8,605 8,495 8,140 8,095 8,445 Value of Time ($/hour) 14.10 13.73 13.43 13.22 12.85 Commercial Cost ($/hour) 74.17 72.23 70.86 71.38 70.47 Gasoline ($/gallon) 1.83 1.44 1.33 1.40 1.48 Diesel ($/gallon) 1.84 1.42 1.29 1.45 1.42 System Performance 2004 2003 2002 2001 2000 Total Fuel (1000 gallons) 14,659 14,187 13,559 12,910 12,263 Rank 44 42 42 41 42 Fuel per Peak Auto Commuter (gallons) 18 17 17 16 15 Rank 33 36 31 31 33 Total Delay (1000s of person-hours) 29,268 28,327 27,072 25,778 24,485 Rank 44 44 44 44 43 Delay per Peak Auto Commuter (pers-hrs) 43 43 43 43 41 Rank 28 28 28 26 26 Travel Time Index 1.20 1.20 1.20 1.20 1.19 Rank 35 35 35 32 34 Commuter Stress Index 1.23 1.23 1.23 1.23 1.22 Rank 41 41 39 39 39 Total Cost ($ millions) 885 880 860 832 813 Rank 44 44 44 44 43 Cost per Peak Auto Commuter ($) 1,079 1,072 1,048 1,014 990 Rank 34 34 36 38 36

Inventory Measures 1999 1998 1997 1996 1995 Population (1000s) 840 830 830 825 825 Rank 45 45 44 43 43 Commuters (1000s) 398 387 381 372 366 Freeway 10,035 9,900 9,600 9,210 8,665 Arterial Streets 8,350 8,210 7,985 7,855 7,690 Value of Time ($/hour) 12.43 12.17 11.98 11.71 11.37 Commercial Cost ($/hour) 66.76 65.76 66.83 66.20 64.27 Gasoline ($/gallon) 1.09 1.06 1.12 1.25 1.14 Diesel ($/gallon) 1.04 1.10 1.19 1.34 1.22 System Performance 1999 1998 1997 1996 1995 Total Fuel (1000 gallons) 11,616 10,971 10,479 9,957 9,512 Rank 41 40 40 40 40 Fuel per Peak Auto Commuter (gallons) 14 13 13 12 12 Rank 34 34 27 31 20 Total Delay (1000s of person-hours) 23,194 21,906 20,923 19,880 18,993 Rank 43 42 42 42 40 Delay per Peak Auto Commuter (pers-hrs) 40 39 37 36 35 Rank 26 23 22 21 20 Travel Time Index 1.19 1.18 1.17 1.17 1.16 Rank 30 29 32 30 32 Commuter Stress Index 1.21 1.21 1.20 1.19 1.19 Rank 39 39 39 41 40 Total Cost ($ millions) 796 768 745 724 712 Rank 43 41 41 41 39 Cost per Peak Auto Commuter ($) 970 936 908 883 868 Rank 35 34 32 30 30

Inventory Measures 1994 1993 1992 1991 1990 Population (1000s) 825 820 815 810 810 Rank 42 42 42 41 41 Commuters (1000s) 361 353 345 337 332 Freeway 8,175 7,850 7,110 6,560 6,265 Arterial Streets 7,770 7,495 7,215 6,810 6,305 Value of Time ($/hour) 11.06 10.78 10.47 10.17 9.75 Commercial Cost ($/hour) 62.23 60.84 59.01 57.31 55.03 Gasoline ($/gallon) 1.01 1.05 1.03 1.04 1.06 Diesel ($/gallon) 1.08 1.13 1.12 1.15 1.06 System Performance 1994 1993 1992 1991 1990 Total Fuel (1000 gallons) 9,031 8,535 8,212 7,397 6,536 Rank 40 40 39 38 38 Fuel per Peak Auto Commuter (gallons) 11 10 10 9 8 Rank 20 24 16 22 25 Total Delay (1000s of person-hours) 18,033 17,041 16,398 14,769 13,051 Rank 40 40 39 39 39 Delay per Peak Auto Commuter (pers-hrs) 33 32 31 29 26 Rank 23 22 23 24 27 Travel Time Index 1.16 1.15 1.15 1.13 1.12 Rank 30 30 26 33 34 Commuter Stress Index 1.18 1.18 1.17 1.16 1.15 Rank 42 40 41 41 42 Total Cost ($ millions) 695 674 668 620 571 Rank 39 39 38 37 38 Cost per Peak Auto Commuter ($) 848 821 814 755 696 Rank 29 24 23 26 29

Inventory Measures 1989 1988 1987 1986 1985 Population (1000s) 805 805 790 785 785 Rank 40 39 40 40 38 Commuters (1000s) 328 325 316 312 309 Freeway 6,130 5,980 5,530 4,930 4,430 Arterial Streets 6,000 5,970 6,200 6,105 5,950 Value of Time ($/hour) 9.25 8.83 8.48 8.18 8.03 Commercial Cost ($/hour) 52.81 50.04 48.53 46.57 47.83 Gasoline ($/gallon) 1.10 1.02 1.02 0.99 1.30 Diesel ($/gallon) 0.99 0.92 0.92 0.90 1.17 System Performance 1989 1988 1987 1986 1985 Total Fuel (1000 gallons) 6,085 5,689 5,500 5,054 4,667 Rank 37 37 34 33 31 Fuel per Peak Auto Commuter (gallons) 7 7 7 6 6 Rank 29 21 19 23 17 Total Delay (1000s of person-hours) 12,149 11,360 10,983 10,092 9,319 Rank 39 38 36 33 35 Delay per Peak Auto Commuter (pers-hrs) 24 23 23 21 19 Rank 28 28 25 26 30 Travel Time Index 1.11 1.11 1.10 1.10 1.09 Rank 36 32 36 32 35 Commuter Stress Index 1.14 1.13 1.13 1.12 1.12 Rank 42 44 40 40 39 Total Cost ($ millions) 560 549 553 526 495 Rank 38 37 35 34 34 Cost per Peak Auto Commuter ($) 683 669 673 641 603 Rank 27 26 24 24 23

Inventory Measures 1984 1983 1982 Population (1000s) 780 780 770 Rank 38 38 39 Commuters (1000s) 305 303 296 Freeway 4,330 4,415 4,045 Arterial Streets 5,670 5,610 5,845 Value of Time ($/hour) 7.75 7.43 7.20 Commercial Cost ($/hour) 46.47 44.23 43.08 Gasoline ($/gallon) 1.31 1.34 1.41 Diesel ($/gallon) 1.19 1.21 1.27 System Performance 1984 1983 1982 Congested Travel (% of peak VMT) -- -- -- Congested System (% of lane-miles) -- -- -- Congested Time (number of "Rush Hours") -- -- -- Total Fuel (1000 gallons) 4,333 4,074 3,936 Rank 32 31 29 Fuel per Peak Auto Commuter (gallons) 5 5 5 Rank 22 19 16 Total Delay (1000s of person-hours) 8,653 8,135 7,859 Rank 33 32 31 Delay per Peak Auto Commuter (pers-hrs) 18 17 17 Rank 30 29 24 Travel Time Index 1.09 1.08 1.08 Rank 29 33 28 Commuter Stress Index 1.11 1.11 1.10 Rank 39 38 37 Freeway Planning Time Index (95th Pctile) -- -- -- Rank -- -- -- Total Cost ($ millions) 476 467 465 Rank 32 32 30 Cost per Peak Auto Commuter ($) 580 569 567 Rank 23 21 19