A bridge life cycle cost analysis adds up what an asset costs to build, run, maintain and eventually dispose of over a defined period, and discounts those amounts to a present value. The method is well established and set out in the international standard on life-cycle costing for constructed assets, and it is not where projects go wrong.
Where they go wrong is the assumptions. Four of them decide the answer almost entirely, and if each bidder chooses their own, the resulting models are three works of fiction that cannot be compared. Fixing those four in the tender document is the single most useful thing a client can do.
The four assumptions that decide the outcome
| Assumption | Effect if set high |
|---|---|
| Period of analysis | Favours durable assets |
| Discount rate | Favours cheap capital |
| Maintenance intervals | Favours low-maintenance options |
| Residual value | Favours recoverable materials |
Each of these is a policy choice belonging to the client, not a technical input belonging to the supplier. That is why they should be published with the tender rather than requested from bidders.
The discount rate does more than everything else combined
Discounting converts a future cost into what it is worth today, and the rate chosen decides how much the future matters. At a high rate, a major repaint in year twenty-five is worth almost nothing in present value, and the model will always prefer the cheapest structure to build. At a low rate, the same repaint carries real weight and a durable asset with no coating cycle wins.
Public sector clients usually have a mandated rate in national guidance, and using it is not optional. What is optional, and worth doing, is running the comparison at a second rate to see whether the ranking changes. A result that holds at both rates is a robust conclusion; a result that flips is telling you the two options are genuinely close and the decision should be made on other grounds.
Period of analysis is not the same as design life
A thirty-year appraisal of an asset designed for eighty years is a perfectly legitimate exercise, provided the analysis credits what remains at the end. Ignore the residual value and the model charges the full capital cost of a long-lived structure against a third of its life, which systematically favours something cheap and temporary.
The residual line has two components. There is the remaining service value of a structure that still has decades in it, and there is the material value at eventual disposal. For aluminium the second of these is a genuine positive figure, since structural sections have a real scrap value and the metal is recycled at a fraction of the energy needed to make it new. For a coated steel structure the disposal line is scrap value less the cost of dealing with the coating, and for composites it is a disposal cost with nothing on the other side.
Maintenance intervals are where suppliers should differ
Having fixed the first three assumptions, the client should require each bidder to supply the fourth input themselves: the schedule of maintenance activities their structure needs, with intervals, durations and the access method each one assumes.
This is the part of the model that reflects the actual product on offer, and it is the part that should vary between bids. It is also the part that needs evidence rather than assertion. A supplier claiming a coating system will last twenty-five years should be able to say on what basis, in what environment classification, and what the manufacturer’s own warranty period is.
The comparison this produces is usually clearer than clients expect, and the underlying logic is set out in our analysis of how the two common materials compare over a full life.
The cost most models leave out
Maintaining a bridge usually means interfering with whatever is underneath it. Repainting a span over a road requires containment, scaffolding or access equipment, traffic management and often lane closures. Over a railway it requires possessions. The direct cost of the work is frequently smaller than the cost of getting to it.
Beyond that sits user cost: the delay imposed on everyone using the road or the route while the work happens. Public sector appraisal frameworks usually have a method for valuing this, and it is routinely omitted from bridge models because it is awkward to calculate. Omitting it is a choice that favours structures needing frequent access, which is precisely the choice the analysis was supposed to test.
A practical middle course is to include the access cost, which is a real invoice, and to state user disruption separately in days of closure rather than in money. Days of closure is a figure decision-makers understand and it is hard to argue with.
What the tender should actually say
Publish the period of analysis, the discount rate and the residual value method. Require each bidder to submit a maintenance schedule with intervals, durations, access methods and the evidence behind the intervals. Require the total days of closure over the period. Require the model itself as a spreadsheet rather than as a summary figure, so the assumptions can be inspected.
That last requirement is the one that changes behaviour. A model that has to be handed over is a model that gets built carefully, and a supplier unwilling to hand one over has told you something useful.
What to do with the result
A life-cycle figure is an input to a decision, not the decision. It sits alongside capital affordability, which is a separate constraint for a public client whose capital and revenue budgets are not interchangeable, and alongside risk, since a structure needing no intervention for thirty years also needs no budget approvals for thirty years.
That second point is worth stating explicitly in a business case. Avoided maintenance is not only a cost saving; it is the removal of a recurring administrative and political exercise that a municipality has to win every time.
Before the models come back
Three life-cycle analyses built on three sets of assumptions tell you about the assumptions. Three built on one set tell you about the structures, which is what the exercise was for.
Our engineers will supply the maintenance schedule, intervals and access assumptions behind any structure we quote, and the gangbroer i aluminium we deliver come with a maintenance plan that states what the life-cycle figure in the offer was based on.