Industries · Road & bridge authority

Road & bridge authority

A road authority has a pavement survey it trusts, a bridge inventory it half-trusts, and a capital programme that has been built the same way for nine years: worst-first, by inspection.

Who this is for

Three people, one budget argument

👷

Pavement engineer

Owns the plan and answers for it. Everything on screen is built to make that week easier.

💷

Bridge asset engineer

Holds the budget line and carries the valuation. Cares about defensibility before features.

🏛️

Director, network assets

Signs, and is the one asked the awkward question in public. Needs one page, not a tour.

Asset classes

What AIPP scores in this industry

Roads Bridges Culverts Footpaths Retaining Walls Traffic Signals
📐

Configured for this industry out of the box: condition on the ASTM_D6433_PCI scale, the TRANSPORT weighting profile, budget shown as 10-year capital programme, and the instruments it answers to (ASTM_D6433, AASHTO_MBEI, AUSTROADS_AGAM, FHWA_NBIS) named on every score. All of it is yours to change — the weights are sliders, not our opinion.

The honest bit

6 of 36 assets will not score

That is the demo register for this industry, and the gaps are deliberate. AIPP refuses to rank an asset the evidence does not support — it will not apply a curve, a zero or a middle grade to a thing nobody has inspected. 17% of this register comes back unranked, with the reason named.

6 of 36 assets carry NO condition evidence and NO age. That is deliberate. A demo register where everything scores teaches a prospect nothing about what happens to the two-thirds of their own register that will not — and it is exactly where most tools quietly invent a zero. Open the demo on these.

How it runs

From export to defensible programme

01

PCI comes in on its own 0–100 scale and converts on the published D6433 bands, not a straight line. Bridges load on NBI 0–9 or MBEI condition states.

02

Pavement risk runs on PCI, rutting and cracking; bridges on structural health, scour and load rating.

03

Weighting balances network criticality against safety — the two things a road authority is actually trading.

04

Ten-year programme runs against the envelope, with treatment timing so a reseal is not spent on a pavement that needs reconstruction.

05

The pack shows the deferred bridges by name, with the detour length each failure imposes.

What you will see

The screens that matter here

The scale trapYour PCI is 100-is-good. Our grade is 5-is-bad. Load that with a linear conversion and your best pavements rank first — I have seen it happen. This uses the published D6433 bands, and states that the crosswalk is ours.
Worst-first is not risk-firstHere is your worst-first list, and here is the risk list. The difference is the arterial with 26,000 vehicles a day sitting above a local road in worse condition. That difference is the entire argument for doing this.
The bridge nobody costedThis bridge is grade 4. Its detour is 30 kilometres. That is not in your condition score, and it is why it outranks three bridges in worse condition.
Treatment timingA reseal on a pavement past intervention is money burnt. The optimiser will not schedule one — it moves to reconstruction and tells you the cost difference.
What is not fundedThis is the honest half of the programme. These do not fit. If council adds this much, this is what comes back in — and in this order.

Fair questions

What this industry actually asks us

We use dTIMS / HDM-4 for pavements.

Then keep it. HDM-4 is listed in our registry as NOT built — AIPP does not model pavement deterioration to that depth. Feed its outputs in as condition and treatment cost, and use AIPP for the cross-class prioritisation and the constrained programme it does not do.

Bridges are element-level for us, not asset-level.

MBEI condition states load as a pack, but AIPP converts the governing state only. If you need element quantities driving renewal cost, that is a real limitation and it is written into the pack.

Our council will not accept a black-box score.

Nor should they. Every score opens to its inputs, its coefficient and whether each input was measured or defaulted. The coefficients start as ours and are meant to be replaced with yours.

Run it on your own worst class

Bring one export — the class with the missing condition data and the costs from 2019. If AIPP tells you the truth about it in twenty minutes, that is worth a conversation. If it scores everything and looks clean, tell us so.