Forecasting what you will have to replace, and when

A capital renewal forecast says what you will have to replace over the next ten or twenty years and roughly what it will cost. A first version takes an afternoon: age, expected life, replacement cost, three columns, and it is already more than most organisations have.

The afternoon version

For each major system in each building, record four things.

• •What it is: roof, boiler, chiller, lift, electrical distribution, windows
• •When it went in, or your best estimate, marked as an estimate
• •Expected life for that system type
• •What replacing it would cost today

In service year plus expected life gives the renewal year. Group the costs by year and you have a forecast. It will be wrong in the detail and right in the shape, and the shape is what the conversation needs.

Then correct it with condition

Age-based forecasting will tell you a 20-year boiler must be replaced this year when it is running perfectly, and will miss the 12-year one that has failed four times. Overlay condition assessments wherever you have them and let condition win over age every time.

Expect spikes, and do not flatten them dishonestly

Buildings are built at once, so their systems reach the end of life at once. A forecast for an estate built in one decade will show alarming peaks thirty years later. That is real information, not an error.

You can smooth the profile by bringing work forward or extending life where condition supports it. You cannot smooth it by moving numbers into later years because the peak is inconvenient, which is what usually happens.

Say what the funding gap is

Compare the forecast against the actual annual capital budget. A common way to express the result is the annual renewal need as a percentage of replacement value, often cited around 2 to 4 percent for a steady state. If you are funding well under that, the backlog is growing by arithmetic, whatever anybody does on the ground.

Present three scenarios, not one request

Current funding, steady state, and catch-up. For each one, show the FCI at year five and year ten. This turns a budget request into a choice with visible consequences, and it is remarkably harder to say no to a choice than to a request.

Refresh it every year, in the same format

A forecast produced once is a document. A forecast produced every year, in the same shape, becomes the thing everybody plans against, and small improvements accumulate. Keep the previous versions so you can show how well last year’s forecast held.