Flexibility in Infrastructure Design

Transcription

1 Flexibility in Infrastructure Design A Future Proof Approach MIT Professor Richard de Neufville Engineering Systems + Civil and Environmental Engineering Author: Flexibility in Engineering Design MIT Press, 2011 with Prof. Stefan Scholtes, University of Cambridge UK

2 Goedemorgen iedereen! Het is een genoegen en een grote eer om met U!

3 This is the book MIT Press, 2011, 2013 Available online from Amazon Germany pocketbook and ebook (about 19)

4 Theme for Today Flexibility in Design is new approach that is Desirable it recognizes uncertainty Practical not difficult to implement Valuable can increase expected value significantly -- at lower cost! Economical way to future proof system I am happy to have the opportunity to collaborate on this with Rijkswaterstaat

5 Outline 1. Flexibility in Design a definition what it is, and why it is new approach 2. Its overall characteristics Desirable it is realistic about issues Practical examples demonstrate Valuable 20 to 30% increases in expected value at lower cost! 3. Economical way to future proof system

6 What is Flexibility in Engineering Design? Flexibility is the capability easily to change the capacity or function of a system. Flexibility exists in a system because of prior planning and design decisions. Flexibility can be either in or on system in as part of technical design (usb ports on lap top to permit other devices) on from management arrangements (mothballing rather than closing a factory) Also known as real options

7 Example of Flexibility in Design Vertical Expansion of Health Care Service Corporation Building, Chicago. Phase 1 (left) and Phase 2 (right, opened 15 years later) in center of image. Source: Goettsch Partners, 2008 and Pearson and Wittels, 2008.

8 Why Flexibility in Design Desirable? The simple answer is because of uncertainty about what we need. The fact is that we cannot predict future exactly over 50-year life of infrastructure. Too many trend-breakers and surprises We need capability to provide for real needs not what we mistakenly imagined years earlier, based on incomplete information This is contrary to established practice!

9 Traditional Design Avoids Uncertainty User Needs Requirements Definition Product System Conceptual Design Marketing or Political Process Systems Engineering System System Validation System Testing Final Assembly Fielding/ Launch Subsystem Preliminary Design Subsystem Development Subsystem Subsystem Integration Adapted from O. de Weck Components Detailed Design Component Design Components Component Testing System Operation

10 Flexibility in Design is Practical Suspension bridges, strength for 2 nd deck George Washington Bridge, New York 25 de Avril bridge across Tagus, Lisboa Skyscrapers for vertical extension HCSC Building, Chicago (image shown) Tufts Dental School, Boston Flood Control Works See image, following

11 Floodwall in Sacramento, California Source: Building of Morrison Creek floodwall, Sacramento, CA, flickr.com via Kim Smet A base strong enough for higher wall, enables building lower wall (less cost) with option to add height if needed

12 Value of Flexibility in Design Flexibility in design can deliver measurably great value. Case after case show 20 to 30% increased expected value. What s the intuition? Mitigates, lessens risk (downside) a win Opens opportunities (upside) more win Lowers first capital expenditures (capex) More value, lower cost, great benefit/cost! A win-win approach

13 Example: Deep Water Oil Platform

14 Details of Case Reservoirs in deep water, off Country *** Design team took traditional approach optimize for best estimates Preliminary design: Single large facility Note Uncertainties: Price of Crude Oil highly volatile Estimates of economically recoverable quantity of oil+gas differ by factor of 2

15 Cummulative Probability Flex Design: small modules + subsea 1 Value-at-Risk-Gain (VARG) Curve (with reservoir uncertainty) X-axis: Value Y-axis: Probability 0.8 Strategies 3&4 0.6 Strategies 1&2 0.4 Strategy 7 Strategy 8 Results compared to base case: Loss: risk eliminated Strategy Strategy Net Present Value [ % of ENPV of strategy 1] Source: Lin et al, 2009 Gain: maximum value much higher Overall: Expected value plus 78%

16 Looking Back: Flexibility in infrastructure is Desirable Realistic view of uncertainty Possible Plenty of examples Valuable many demonstrations Now: Economical Way to Future Proof our systems

17 An Essential Issue: Uncertainty Our infrastructure lasts a long time One generation for sure Maybe two or three We do not, indeed cannot, know what our needs will be so far in advance Experience of surprises, trend-breakers (climate change, fracking, nuclear )

18 Forecast is always wrong What actually occurs is generally differs greatly from what was forecast long ago Even when overall numbers are within 10%, big differences in needs, services (think of aviation traffic through Schiphol) The point is, we really do not know what future our infrastructure will serve

19 How do we future proof infrastructure? What is your reflex? How do we guarantee future performance? Make infrastructure strong enough (raise dikes by x meters, for example) to meet all conceivable needs. By definition this would guarantee future provided we could really conceive of all possible futures

20 Over-building is expensive and wasteful Overbuilding is clearly expensive (not only in money, also in space, time, and effort) Overbuilding is wasteful because focus on extreme means strong possibility that much of system will be unnecessary Building later is cheaper (deferring interest payments) Better to build later, with less ignorance, more knowledge of future.

21 Flexible Design Recommended It is the more efficient, economical way to prepare for unknown future. Properly done, flexible design: Enables appropriate response to several possible futures Defers some investment (saves interest) Avoids some investment (more savings) Lower cost per project permit more projects, more service to more people.

22 Overall Suggestion Less effort on trying to predict long-term future in any case unknowable Less effort on building infrastructure now that will satisfy guessed at future Emphasis on implementation of facilities that could by extension or adaptation meet several futures Thus reducing immediate project costs, and facilitating more projects and service

23 This is the book MIT Press, 2011, 2013 Available online from Amazon Germany pocketbook and ebook (about 19)