TAM ETG Webinar Series Webinar 5: Optimization and Cross-Asset Allocation

Transcription

1 TAM ETG Webinar Series Webinar 5: Optimization and Cross-Asset Allocation Sponsored by the TAM Expert Task Group Submit questions and comments using the webinar s Q&A feature Webinar 6 January 20, 2016 (Moved from January 13 due to TRB)

2 Welcome The TAM ETG is pleased to sponsor this webinar on Optimization and Cross-Asset Allocation Since 2012, FHWA and AASHTO have been hosting these types of webinars to provide an opportunity to connect with the asset management community and to bring people together around a series of important topics Sharing knowledge is a critical component of advancing asset management practice All of the webinars in the series can be accessed online at: 2

3 Agenda FHWA Opening Remarks Steve Gaj, FHWA Creating a Vocabulary for Cross-Asset Allocation Gordon Proctor, Proctor and Associates Asset Management: Defending Program Funding Jack Smith, PE, North Dakota DOT Baby Steps on Tradeoffs in Wyoming Martin Kidner, Wyoming DOT Question and Answer Matt Hardy, AASTHO 3

4 TAM ETG Overview Established to serve as a forum to discuss changes in the way highway agencies are managing assets Activities include: Identifying strategies for advancing asset management practice and influencing change within state DOTs/MPOs Providing input to FHWA, AASHTO, and TRB Developing and implementing a plan for communicating the work of the TAM ETG to the transportation community. Meetings In-Person Meetings are open to the public November 17 and 18, St. Paul MN Membership Public sector representing federal, state and local agencies 4

5 TAM ETG Membership Members Tim Henkel (Chair), Minnesota DOT Brad Allen, New York State DOT Jennifer Brandenburg, North Carolina DOT Chris Champion, IPWEA (International) Chris Evilla, Waco MPO Moh Lali, Alberta, Canada (International) Laura Mester, Michigan DOT Randy Park, Utah DOT Omar Smadi, Iowa State University (Academia) Liaisons Steve Gaj, FHWA Office of Infrastructure DeLania Hardy, AMPO Matt Hardy, AASHTO Tom Palmerlee, TRB Neil Pedersen, TRB Nastaran Saadatmand, FHWA Office of Infrastructure Francine Shaw-Whitson, FHWA Office of Transportation Performance Management Egan Smith, FHWA Office of Planning 5

6 TAM ETG Webinars Overview 1. March 11: Target Setting and TAMPs 2. May 13: TAM 101 for MPOs 3. July 8: Network Life Cycle Cost Analysis Part 1 4. September 9: Network Life Cycle Cost Analysis Part 2 5. November 12: Optimization and Cross-Asset Allocation 6. January 20, 2016: Integrating TAM into Agency Processes 6

7 FHWA Opening Remarks from Steve Gaj, FHWA Office of Asset Management, Pavements, and Construction 7

8 Creating a Vocabulary for Cross- Asset Allocation Presenter: Gordon Proctor, Proctor and Associates

9 Asset Management: Defending Program Funding Presenter: Jack Smith, PE, North Dakota DOT 9

10 Baby steps on Tradeoffs in Wyoming Presenter: Martin Kidner, Wyoming DOT 10

11 Questions? Submit your questions using the webinar s Q&A feature 11

12 Creating a Vocabulary for Cross-Asset Allocation FHWA/AASHTO Asset Management Expert Task Group

13 Balancing Investment Options Agencies face difficult investment tradeoff decisions New assets vs. sustaining the old? Which assets to preserve? 11/13/2015 Asset Management ETG 2

14 ETG Perceived a Lack of Clarity The practice of tradeoffs and optimization is new resulting in: A lack of common definitions Inability to discuss trade-off options when the language is unclear Different agencies assigning different meanings to the same terms 11/13/2015 Asset Management ETG 3

15 ETG s Objective for White Paper Begin discussion to create: A dialog on tradeoffs and optimization Start a taxonomy Suggest some common definitions

16 Often Used, Seldom Defined Cross-Asset Tradeoff Cross-Asset Allocation Cross-Asset Optimization 11/13/2015 Asset Management ETG 5

17 Cross-Asset Tradeoffs Suggested definition is: The decision-making process by which resources from one asset class are transferred to another in order to maximize perceived utility. My bridges are good, my pavements are not: Let s move money from bridges to pavements Key words are perceived utility 11/13/2015 Asset Management ETG Webinar 6

18 Cross-Asset Allocation Suggested definition is: The decision-making process by which resources to multiple programs or asset classes are distributed based on the simultaneous quantification of utility. Utility is any perceived benefit Three types discussed Benefit/Cost Multi-Criteria Decision Analysis Risk/Reward-Base Allocations 11/13/2015 Asset Management ETG Webinar 7

19 Multi-Criteria Decision Analysis Frequently used in environmental documents and other decision contexts Can be simple, transparent However, is subjective Ignores risk Multi-Criteria Analysis Matrix Option 1 Option 2 Option 3 Option 4 Safety Criteria Safety Criteria Total Safety Value Safety Weight 4 Weighted Safety Score (weight X value) Pavement Criteria Pavement Criteria Total Pavement Value Pavement Weight 2 Weighted Pavement (weight X Score value) Bridge Criteria Bridge Criteria Total Bridge Value Bridge Weight 3 Weighted Bridge Score (weight X value) Mobility Criteria Mobility Criteia Total Mobility Value Mobility weight 1 Weighted Mobility Score Weighted Ranking of the 4 Options

20 Risk/Reward-Based Allocations ETG suggested definition:..is an investment strategy that seeks to apportion investment in different assets to maximize the overall utility while not exceeding acceptable levels of risk. Began on Wall Street in 1952 by Harry Markowitz His asset-allocation formula balanced: Risk Reward 11/13/2015 Asset Management ETG Webinar 9

21 Valuing Risk You have a 10% chance of losing $100. What is the value of the risk? $100 X 10% = $10 You face two investments: 90% likelihood of earning $1,000 80% likelihood of earning $1,200 Which is best? $1,000 X 90% = $900 $1,200 X 80% = $960

22 It Depends What is your level of risk tolerance?

23 Cross-Asset Optimization Most complex, least understood Defined as, The use of recursive mathematical computations to determine the maximum utility for a given set of investments constrained by defined parameters. Differentiated by: Computerized optimization routines to achieve an optimum overall utility 11/13/2015 Asset Management ETG Webinar 12

24 Strengths and Weaknesses Strengths Most sophisticated Focused on overall utility or benefit Examines many options Weaknesses Complex Black box Still relies on some subjective values 11/13/2015 Asset Management ETG Webinar 13

25 White Paper Conclusions It suggests definitions to prompt a maturation in the dialog It s hard to discuss something that is not defined Taxonomy of types of tradeoffs could lead to further research, refinement of each Paper hopes to stimulate thought, advancement in this promising field 11/13/2015 Asset Management ETG Webinar 14

26 Presented to TAM ETG Webinar on Optimization and Cross Asset Allocation November 12, 2015 Jack L. Smith, PE Assistant Planning/Asset Management Engineer North Dakota Department of Transportation Planning/Asset Management Division (701)

27 1. Asset Management Language and Principles 2. Why asset management? 3. Refining the Tie Between Planning, PM, TAM, & Programming (i.e. the STIP) 4. Implementation Plan 2

28 Tradeoff Analysis Like a slider moving $ s from one asset class to another. Shows estimated outcome of investment decisions. Optimization Analysis A logic-driven computation recommending the best investment regimen for or between asset classes. Typically, based on benefit-cost analysis.

29 Two ways to use TAM Strategic-level tool System-wide analysis E.g. the average condition is predicted to be Tactical-level tool Project and operational elements are recommended E.g. the recommended optimum series of projects to do is

30 TAM is a way of thinking not a piece of software

31 Fundamentally, TAM is a: Goal-Oriented, Data-Driven Decision-Making Process

32 TAM provides information to decision makers it does not make decisions

33 8

34 Federal Funding ESTIMATED FEDERAL HIGHWAY TRUST FUND OBLIGATIONS 45 Highway Highway Safety Transit Dollars in Billions * * Assumes Congress will fully fund MAP21 proposed expenditures in If no new revenues are found, federal highway obligations will fall by almost 100% in FY

35 National Service Demand 3.5 U.S. Vehicle Miles Travelled 3.0 VMT (Trillions of Miles) Year 10

36 State Biennium Road Budget 3 Uncertain Funding (Billions of $ s) Relatively Certain Funding

37 1950 = 17.1 Billion Pounds* 2010 = 89.4 Billion Pounds* 422% Increase 2013 = 81.8 Billion Pounds* 378% Increase * Approximate numbers 12

38 13

39 The Department loses autonomy in programing projects. Unless, our logical, data driven process for selecting projects can be shown in a transparent manner. Instead of agencies competing for limited funding, competition will move to individual program areas. 14

40 Pavement Management Projected Average Network Ride Quality Federal Funding Only IRI (in/mi)

41 We refine our Planning, PM, TAM, & Programming processes Transparency We present proposed budgets by showing impacts across the spectrum We show why our investment plan maximizes service provided Ultimately, we show economic impacts that investment plan has on North Dakota 16

42 We answer: How much funding is needed to maintain LOS? What LOS can be provided for a given funding level? Where is the best place to spend any given dollar? 17

43 18

44 19

45 Refining Tie Between Planning/PM/TAM/STIP PLANNING Find out broad-based customer goals/expectations PROGRAMMING STIP & bid projects PERFORMANCE Establish detailed indicators for customer expectations TAM Investment strategy (targets) & Candidate projects 20

46 Must Know What is Important to Our Customers 21

47 Refining Tie Between Planning/PM/TAM/STIP PLANNING Find out broad-based customer goals/expectations PROGRAMMING STIP & bid projects PERFORMANCE Establish detailed indicators for customer expectations TAM Investment strategy (targets) & Candidate projects 22

48 Biennial Report (Future Asset Trending Report Service Indicators) 23

49 Refining Tie Between Planning/PM/TAM/STIP PLANNING Find out broad-based customer goals/expectations PROGRAMMING STIP & bid projects PERFORMANCE Establish detailed indicators for customer expectations TAM Investment strategy (targets) & Candidate projects 24

50 Work Type Interstate Interregional State Corridor, District Corridor, & District Collector Goal 45% 20% 35% Preventative Maintenance 27% 26% 36% Minor Rehabilitation & 27% 42% 42% Structural Overlay Major Rehab, New or Reconstruction 38% 26% 16% Structures 3% 3% 3% Safety 5% 3% 3% NDDOT Moving to Strategy based on Investment Classes. How? 25

51 Asset-Class Funding Distribution (Investment Strategy): Rural Asset Class Preservation 1 Year Bond Mega BR Functional Payment Rural SubTotal (all funds) Grand Total (all asset Pavement Bridge Replacement 2 Capacity 3 Safety Coats NHS 4 classes) Seal Urban 26

52 27

53 Bridge Management Motor Vehicle Drivers License Rail Transit Operating Rdwy/Br. Maintenance Tradeoff Hub Safety Snow & Ice Control Pavement Management Functional Capacity Urban NHS Bike/ Pedestrian 28

54 29

55 Service Indicators, more commonly called Performance Measures need to be tied funding. Source of funding will need to be considered and programed into the trade off hub. Interdependencies between investment classes will need to be accounted for. Scope of each investment class must be clearly defined. 30

56 Different investment classes have different deterioration rates and service lives. The question we are trying to answer is the impact of an investment decision. The NDDOT is looking at a 20 year period of analysis. Remember that the service indicator chosen for each investment class need not represent every parameter used to manage the investment class. 31

57 Establish Service Indicators for Investment Categories November 17, Present Draft Hub results to Executive Staff April 15, Verify and refine (if needed) predicted budget results for final presentation to Executive Staff May 24,

58 This schedule is the result of years of extensive outreach with NDDOT internal stakeholders. The NDDOT is a smaller DOT, and many managers have broad experience throughout the Department. We are confident in our current process and are willing to be transparent. 33

59 The NDDOT keeps risk in mind as it relates to the freight network and highway performance classification system. We are awaiting the final TAM rule to determine if additional requirements are needed. 34

60 The process outlined in this presentation is the method we feel fits our organization best at this time. The NDDOT is open to any future optimization programs with the consideration of the cost of obtaining necessary data. 35

61 Goals: Establish process transparency. Refine internal budget allocation process at a program level. This is NOT about: Meeting federal TAM requirements TAM is about fulfilling a responsibility to our stakeholders. 36

62 Jack L. Smith, PE Assistant Planning/Asset Management Engineer North Dakota Department of Transportation Planning/Asset Management Division (701)

63 BABY STEPS ON TRADEOFFS IN WYOMING Martin Kidner WYDOT WYDOT 1

64 BASE PROGRAM REPORT CONCEPT Cumulative Benefit Cumulative Cost -- $ Budget Limit WYDOT 2

65 Pavements Interstate - 20% Systems Comparison 90% Percent Miles Good to Excellent 80% 70% 60% 50% 40% 30% 20% Interstate NHS NonNHS WYDOT 10% 0% Year

66 TOTAL BRIDGES IN WYOMING WYDOT 4

67 HSM DEFENSIBLE APPROACH The Highway Safety Manual (HSM) introduces a science-based technical approach that takes the guesswork out of safety analysis. The HSM provides tools to conduct quantitative safety analyses, allowing for safety to be quantitatively evaluated alongside other transportation performance measures such as traffic operations, environmental impacts, and construction costs. WYDOT 5

68 EMPIRICAL BAYES CORRECTED ( EXPECTED ) WYDOT 6

69 WYDOT 7

70 TREATMENT AND TARGET CONTRIBUTIONS WYDOT 8

71 TARGET SETTING HSIP funds = 18M$ 20:1 B/C ratio achievable 18M$ of treatments with 20:1 should result in an annual reduction in critical crashes of 8.05 HSIP funds $18,000,000 Benefit to cost average 20:1 Number of critical crashes per year that can be theoretically reduced 8.05 WYDOT An anticipated reduction of critical crashes by 1.5 per district per year seems to be an achievable target 9

72 CURRENT CONSTRUCTION SPENDING Program Bridge 10% Community Development 3% Environment 1% Maintenance 6% Mobility 8% Other 1% Pavement 54% Safety 15% Urban 2% WYDOT 10

73 EXAMPLE PROJECT WYDOT 11

74 WYDOT 12

75 CONCLUSION Goal areas for the state have different benefit definition Must decide method of allocation Projects sometimes have multiple impacts on goal areas HSM gives a national standard HSM gives predictive ability WYDOT 13

76 THANK YOU WYDOT 14