Bridge Rehabilitation, Preservation, and Life-Cycle Design

20.1 — Engineering story

MDOT SHA — Chesapeake corridor deck preservation program

In 2012 the Maryland State Highway Administration inspected 187 concrete decks on the US-50 / US-301 Chesapeake corridor. Twenty-eight were already NBI 4 (Poor); another 96 were sliding through NBI 5 with chloride readings above the corrosion threshold. Full replacement of the 28 poor decks was budgeted at $46 M. The bridge preservation team argued the opposite: freeze the remaining 159 with cyclic actions — crack sealing, joint reset, and 2-inch latex-modified concrete (LMC) overlays — for $11 M, and stretch replacement dollars to only the decks truly beyond rehab. Ten years later, network-average deck NBI on the corridor rose from 5.6 to 6.3, and the replacement backlog shrank by 41%. That's the preservation thesis: keep good bridges good.

20.2 — Chapter objectives

What you will be able to do

Learning objectives

By the end of this chapter you will be able to:

1Distinguish preservation, rehabilitation, and replacement per the FHWA intervention hierarchy.
2Select an appropriate deck preservation or rehabilitation treatment (crack seal, methacrylate, thin polymer, LMC, HMA + membrane) from condition data.
3Size an FRP confinement wrap for an RC column using ACI 440.2R.
4Design an external post-tensioning retrofit that lifts a girder's flexural capacity by the target ΔM.
5Retrofit a fatigue-cracked steel detail with drill-stop + bolted cover splice and recover the AASHTO detail category.
6Perform a life-cycle cost comparison (NPV, real discount rate) among preservation, rehabilitation, and replacement alternatives.
7Re-run the LRFR rating (Ch. 19) after a retrofit and confirm the target rating factor is achieved.

20.3 — Engineering motivation

Preservation is a design activity, not a maintenance one

The Federal-Aid Highway System is ~60 years old on average. Replacing every deficient bridge with a new one is fiscally impossible — FHWA estimates the U.S. backlog at $125 B. Preservation and rehabilitation engineering is therefore the discipline where most working bridge engineers spend most of their careers. It is design in every sense: loads, resistance, detailing, constructability, and — added — the constraint that the bridge stays open during the work.

20.4 — Lecture

The FHWA intervention hierarchy

AASHTO LRFD FHWA Bridge Preservation Guide (HIF-11-042)— Preservation · Rehabilitation · Replacement

Fig. 20.1 — Intervention hierarchy (FHWA preservation model)

FHWA sorts every action into three buckets. Preservation is cyclic (deck washing, joint flushing, sealer) or condition-based (crack seal, spot painting, thin overlay) — cost per bridge is small, coverage is broad, and each dollar buys the most rating-year return. Rehabilitation restores a functionally or structurally deficient component — deck overlay + partial-depth patch, girder strengthening, bearing replacement. Replacement is reserved for bridges past rehabilitation or geometrically obsolete.

Fig. 20.2 — NBI rating vs. time — with and without preservation

The deterioration curve makes the economics visible: a bridge without preservation drops from NBI 9 to NBI 4 in ~35 years; with cyclic preservation the same bridge holds NBI ≥ 6 past year 60. Both bridges will eventually need major work, but the preserved one delivers 25 extra service-years for a fraction of the replacement cost.

20.5

Deck preservation and rehabilitation treatments

Fig. 20.3 — Rehabilitated deck cross-section (LMC overlay)

Deck treatments are selected against measured condition data — chloride content at the top mat, chain-drag sound survey, GPR delamination map, half-cell potentials, and cover depth. Rule-of-thumb triggers:

Crack seal (methacrylate/epoxy) — surface cracks, no delamination, chloride below threshold. Preserves.
Thin polymer overlay (¼–½ in) — early-life friction restoration, chloride barrier. Preserves.
HMA + waterproofing membrane (2 in) — northern climates, quick installation.
LMC / silica-fume overlay (1.5–2 in over hydro-demolition to sound concrete) — most common structural rehabilitation; restores rider quality and provides a chloride barrier for 25+ years.
Full-depth deck replacement — cover contamination through the bottom mat; often paired with girder repair.

20.6

Superstructure retrofits — FRP, external PT, section restoration

AASHTO LRFD ACI 440.2R · AASHTO Guide Specifications for LRFD Retrofit

Fig. 20.4 — FRP confinement of a circular RC column (ACI 440.2R)

FRP composites are the workhorse of concrete retrofit — thin (0.04 in/ply), light (no dead load penalty), and installed on live bridges with lane closures only. Design equations follow ACI 440.2R; the confined compressive strength is

f'_{cc} \;=\; f'_c \;+\; \psi_f \cdot 3.3\,\kappa_a \, f_l, \qquad f_l \;=\; \dfrac{2\,n\,t_f\,E_f\,\varepsilon_{fe}}{D}

with ψ_f = 0.95 (partial-safety), κ_a = 1.0 for a full-circular section, and ε_fe = 0.55·ε_fu capped at 0.004 for column confinement.

Fig. 20.5 — External post-tensioning retrofit of a simple-span girder

External post-tensioning shifts a fraction of the demand off the cross-section by applying a compressive axial force at eccentricity e. The relief moment is simply ΔM = P_eff · e. Anchor blocks and deviators must be designed for the tendon force plus the local shear introduced at each direction change.

Fig. 20.6 — Fatigue-crack retrofit — drill stop hole + bolted splice

Fatigue retrofits combine two moves. A drill stop-hole at the crack tip arrests propagation by turning the stress-concentration singularity into a finite-radius hole. A bolted cover splicerestores continuity and reclassifies the detail — a properly designed slip-critical splice recovers Category B or better in place of the Category E′ that caused the crack.

20.7

Substructure retrofit — jacketing and scour countermeasures

Fig. 20.7 — RC column seismic retrofit — steel shell jacket

Cover-loss and lightly-confined RC columns from the pre-1971 era are retrofit with steel-shell jackets or additional CFRP wraps to add confinement, shear capacity, and lateral drift capacity for seismic redesign (Ch. 15). The jacket is set proud of the column with a cementitious grout gap; a small vertical dowel array at the footing transfers shear. Scour countermeasures (Ch. 14) — riprap, articulating concrete block, collar plates — protect the foundation without replacing it.

20.8

Life-cycle cost analysis (LCC)

Fig. 20.8 — Life-cycle cost — three alternatives, NPV at r = 3%

Every alternative — do nothing, preserve, rehabilitate, replace — is compared on net present value of agency + user costs over a common analysis period (usually 40–75 years):

\text{NPV} \;=\; \sum_{t=0}^{T} \dfrac{C_t}{(1+r)^{t}}, \qquad r \approx 0.03\text{–}0.04 \; (\text{real})

Agency cost is the construction cost plus recurring maintenance. User cost is delay, detour, and crash — often larger than the agency component on high-ADT corridors. The alternative with the lowest total NPV wins, subject to any binding funding or performance constraint.

20.9 — Worked example

External post-tensioning retrofit of a deficient composite girder

Given

Simple span L = 80 ft interior composite steel girder. Post-inspection φR_n = 5,600 k-ft; demand from HL-93 gives M_u,DesignInv ≈ 6,050 k-ft, so RF_Inv = 5,600/6,050 ≈ 0.93 < 1.0. Target: bring RF_Inv ≥ 1.10 by external PT.

Available: 4 × 0.6-in Gr. 270 strand bundle (A_ps = 4·0.217 = 0.868 in²), jacked to 0.75 f_pu = 202.5 ksi; long-term losses 20% (Ch. 7). Eccentricity at midspan e = 22 in from the composite centroid.

Step 1 — Effective tendon force.

P_j \;=\; 0.75\,f_{pu}\,A_{ps} \;=\; 202.5 \times 0.868 \;=\; 175.8\ \text{kips}

P_{eff} \;=\; (1 - 0.20)\,P_j \;=\; 140.6\ \text{kips}

Step 2 — Relief moment at midspan.

\Delta M \;=\; P_{eff}\cdot e \;=\; 140.6 \times \dfrac{22}{12} \;=\; 257.7\ \text{k-ft}

Step 3 — Not enough. Increase to 2 bundles (A_ps = 1.736 in²).

P_{eff,2} \;=\; 281.2\ \text{kips}, \qquad \Delta M_2 \;=\; 515.5\ \text{k-ft}

Step 4 — Effective demand and new rating factor.

M'_{u} \;=\; 6{,}050 - 515.5 \;=\; 5{,}534.5\ \text{k-ft}

RF_{Inv,\text{new}} \;=\; \dfrac{5{,}600}{5{,}534.5} \;=\; 1.012 \;\;\text{(barely passing)}

Step 5 — Add a third bundle → ΔM = 773 k-ft, RF ≈ 1.06; add e = +2 in via lowered deviators → ΔM ≈ 848 k-ft.

RF_{Inv,\text{final}} \;=\; \dfrac{5{,}600}{6{,}050 - 848} \;=\; 1.077 \;\;\text{— still short of 1.10}

Conclusion: A 4-bundle tendon at e = 24 in gives RF_Inv ≈ 1.14. Anchor zones and deviators are then designed for P_j = 703 kips per bundle × 4 with the AASHTO §5.9.4 anchor-zone strut-and-tie model.

Fig. 20.5 — External post-tensioning retrofit of a simple-span girder

Fig. 20.5External tendon geometry

20.10 — Design challenge

Rehabilitate a 1972 3-span steel plate-girder overpass

Fig. 20.9 — Design challenge — 1972 3-span steel plate girder overpass

You inherit a 1972 3-span continuous welded plate-girder overpass — spans 90–120–90 ft, ADT 12,000 with 6 % trucks. NBI: Deck 5, Superstructure 5, Substructure 4. Field notes: 18 % of deck area delaminated, chloride ≈ 3× threshold at top mat; pack-rust at the north abutment bearings; column cover loss with rebar corrosion on both piers; a 3-in fatigue crack at the negative-moment field splice (Category E′).

Deliver (single PDF or PPT):

Preservation vs. rehabilitation vs. replacement decision, with justification tied to NBI, chloride, and traffic.
Deck treatment selection (crack seal / thin polymer / LMC / HMA+membrane / full replacement) with removal depth and traffic control plan.
Superstructure retrofit — external PT sizing OR bolted flange cover splice + drill stop-hole — with post-retrofit RF calculation.
Column jacket sizing (steel shell OR CFRP wraps) using ACI 440.2R equations.
Bearing replacement plan (jack, remove, replace) integrated with deck work.
LCC comparison of at least three alternatives on a 40-year horizon at r = 4 % real.

Submit your design challenge

Chapter 20 — Rehabilitation Design Challenge

You must be signed in to upload a submission.

20.11 — Summary

Take-aways

Preservation is a design activity — keep good bridges good with cyclic and condition-based actions.
Deck treatment selection follows measured condition (chloride, delamination, cover).
FRP wraps (ACI 440.2R) add confinement and ductility without dead load.
External PT relieves flexure by ΔM = P_eff · e; anchor zones govern detailing.
Fatigue crack retrofits combine drill stop-hole + bolted cover splice to recover a better detail category.
All alternatives are compared on NPV at a real discount rate over a 40–75 yr horizon.

20.12 — Graded quiz

PE-format assessment — 8 items, 70% to pass

Multi-step problems. Use AASHTO / ACI equations and the figures on the right. Each item has a hint that names the code section — reveal it only after an honest attempt.

Chapter 20 — Rehabilitation, Preservation, and Life-Cycle Design

8 questions · PE-exam format · 70% to pass · attempts save to your progress record when signed in.

Work each item to the requested precision. Use the Show clue button only after an honest attempt — hints reveal the AASHTO section and setup, not the answer.

Q1
The FHWA Bridge Preservation Guide (HIF-11-042) places preservation, rehabilitation, and replacement in a strict priority hierarchy. Which statement most accurately captures the intent of that hierarchy at the network level?
Every bridge should be replaced on a 50-year cycle regardless of condition.
Preservation is applied first and most broadly because each dollar buys the largest rating-year return; rehabilitation and replacement are reserved for bridges that can no longer be preserved.
Rehabilitation and replacement are always cheaper than preservation over a 40-year horizon because they extend service life by more years.
The hierarchy applies only to concrete bridges and is optional for steel superstructures.
Quiz reference — FHWA intervention pyramid
Q2
A concrete deck without preservation deteriorates linearly from NBI = 9 at year 0 to NBI = 4 at year 40. A cyclic preservation program applied at years 12, 24, 36 raises the current rating by +1.2 at each application. Assuming the same underlying decline rate between applications, at what age (years) does the preserved deck first drop below NBI = 5?
years
Quiz reference — deterioration timing
Q3
A 2-inch LMC overlay (γ = 145 pcf) is placed on an existing interior girder deck at spacing S = 8.0 ft. Compute the additional distributed dead load on the girder due to the overlay (kip/ft).
kip/ft
Quiz reference — LMC overlay
Q4
Using ACI 440.2R, size a CFRP wrap on the circular RC column shown. Compute the number of plies n required to develop a confining pressure f_l = 0.60 ksi. (Round up to the next whole ply.)
plies
Quiz reference — CFRP wrap sizing
Q5
An external post-tensioning tendon is installed on a simple-span girder. After all losses P_eff = 250 kips and the eccentricity at midspan is e = 18 in. Compute the midspan relief moment ΔM (k-ft) contributed by the tendon.
k-ft
Quiz reference — external tendon
Q6
Alternative B — preservation + rehab: spend $0.20 M at year 0, $0.60 M at year 15, then $2.00 Mreplacement at year 40. Compute the NPV of Alternative B at a real discount rate r = 4 % (report as $M, positive number, two decimals).
$ M
Quiz reference — LCC alternatives
Q7
For the retrofitted composite girder shown, compute the Design Inventory rating factor RF_Inv per AASHTO MBE §6A.4.2 using the values on the reference card. Report to two decimal places.
Quiz reference — post-retrofit LRFR
Q8
A Category E′ welded-attachment detail on a steel girder is cracked and is retrofitted with a drill stop-hole at the crack tipfollowed by a slip-critical bolted cover splice across the cracked flange (see figure). Per AASHTO §6.6.1.2 / FHWA fatigue retrofit guidance, which statement is most correct?
The retrofit does nothing — the original Category E′ still governs because a crack existed.
The drill stop-hole alone recovers Category B; the cover splice is only cosmetic.
The bolted cover splice controls the retrofitted stress range and the check reverts to Category B (slip-critical bolted, gross section) provided the splice develops the flange force.
The retrofitted detail must be re-classified as Category A (base metal) with no reduction for the drilled hole.
Quiz reference — retrofit detail category

Section 2

Fully Worked Examples

Complete AASHTO LRFD solutions with knowns, assumptions, step calculations, verification, and design commentary. Difficulty rises from basic to consulting-grade.

Worked Example 1

Concrete deck overlay thickness selection

Basic

Problem

Choose overlay type and thickness.

Step-by-Step

Mill down 1.5 in to sound, chloride-free substrate.

Latex-modified concrete (LMC): 1.5–2 in; Silica-fume: 1.25–1.5 in; Polymer (PPC): 3/8–3/4 in.

Design Verification

Verify total added DL ≤ 1.2 in equivalent (≈15 psf). PPC at 5/8 in ≈ 7 psf ✓

Discussion

LMC and silica-fume overlays are stronger but heavier — always check DL reserve before specifying. PPC is fastest to open to traffic (4–6 hr cure) but requires clean, dry substrate.

Worked Example 2

CFRP flexural strengthening of a deficient girder

Intermediate

Problem

Estimate required number of CFRP plies (bottom flange bonding).

Step-by-Step

f_{fe} = E_{f}\cdot \varepsilon _{fe} = 33{,}000\cdot 0.007

Result

f_{fe} = 231 ksi

F_{ply} = f_{fe}\cdot A_{ply} = 231\cdot (6\cdot 0.055)

Result

76.2 kip/ply

Design Verification

Two plies deliver ~430 k·ft — provides ~95% reserve. Rounding up to 2 plies is standard practice for construction tolerance.

Discussion

CFRP strengthening is capped at 40% of live-load increase over unstrengthened capacity — never rely on FRP to carry the entire deficit. Verify anchorage development length (≥6 in typical).

Worked Example 3

Cost-effectiveness index for deck replacement vs overlay

Intermediate

Problem

Compute equivalent uniform annual cost (EUAC) at i = 4% and recommend.

Step-by-Step

CRF_{8} = 0.04/(1-1.04^{-8}) = 0.1485; CRF_{40} = 0.04/(1-1.04^{-40}) = 0.0505

EUAC_{A} = 18\cdot 12{,}000\cdot 0.1485

Result

$32{,}076/yr

Design Verification

Pure capital EUAC favors overlays. Once user delay, MOT, and repeat-mobilization costs are included, deck replacement usually wins at ADT > 20,000.

Discussion

LCCA must include agency + user costs. Deferred replacement compounds risk of catastrophic failure — combine LCCA with condition-based decision matrix (FHWA-HIF-16-002).