Why Every Organization Needs a Renewal & Replacement (R&R) Forecast
- JD Solomon
- 43 minutes ago
- 9 min read
Most infrastructure and facilities organizations cannot answer a basic question: over the next 20 years, are we heading toward financial mountains or a manageable, flat stretch of road? That question is a guess without a renewal and replacement (R&R) forecast. And an organization that is guessing about its long-term costs cannot commit to being financially stable.
What is an R&R Forecast?
A renewal and replacement forecast turns a guess into a number. It pulls together remaining asset life, condition, replacement value, repair cost and frequency, O&M forecasting, and capital forecasting into one financial picture.
The forecast period is usually 20 years. Debt service usually runs 20 to 30 years, and most rotating equipment works through a full life cycle in roughly that span. Line those two facts up and you have a model that speaks the language your board and your finance team already use.
From the Real World: A Zoo, Two Old Mechanics, and a Legal Pad
Early in my career, I worked on a project at a zoo built up after World War II. The old-school maintenance director and a sidekick who reminded me of my grandfather, a head mechanic at a Dodge dealership, ran the show. Our project team was there to save the monkeys, lions, and tigers from pending death as a result of a failing run-to-failure maintenance approach. Change was needed, but we would have to use a pencil and legal pad and not a high-tech approach.
So we kept it simple. We built a list of every asset the zoo owned. We called vendors and asked how long the equipment was supposed to last and what it would cost to replace it. Where condition was clearly worse than average, we adjusted the useful life down through a quick tabletop exercise with the people who worked on the equipment every day. It was, by any modern standard, a dumb-down approach and a rough estimate of the financial mountain the organization was climbing.
Most importantly, it worked. Thirty-plus years of consulting later, that same basic model is still what I fall back on whenever a by-the-book asset management program stalls out.
Know what you own and what condition it’s in. Get the asset values. State your assumptions. Grip it and rip it.
The Four Things You Actually Need
You need four things in hand. Then everything is refinement that can be done in subsequent forecast updates.
1. Asset List with Install Dates and Condition
For rotating equipment, that means nameplate data. For most vertical assets like buildings and rotating equipment, you will be told the central database is about 60 to 80 percent accurate. The good news is that there is a lot of legacy data you can find, so that number is about right once available information is added and poor data is deleted.
For linear assets such as buried pipe, that usually lives in your geospatial information system (GIS). You will be told that’s 90 to 100 percent complete and accurate. In reality its closer to 70 to 80 percent.
Some digging will be needed, but remember the process is iterative rather than linear. Too many consultants push owners to perfect their data before running any model. Run the model first based on what you know. Determine the sensitivities. Refine and improve where needed.
2. Remaining Useful Life Estimates
Nobody knows exactly when an asset will fail. That is fine. Assign condition on a simple 1-to-5 scale, with 1 as brand new and 5 as failure could happen any moment, and let the people closest to the equipment fill it in through a tabletop session where formal assessments do not exist. Do not let missing condition data stop you from running a first forecast.
3. Replacement Asset Values
This is the financial half of the equation, and it deserves a team as capable as your technical asset team. Without replacement values, you cannot produce the forecast and you cannot know your true cost of ownership. Expect friction here too: your accounting figures, your insurance declarations, and your O&M-side replacement costs will all rest on different bases, and someone has to reconcile them.
4. Stated Assumptions
Are you using point estimates or ranges and probabilities? What does a “rebuild” actually mean in your organization, in terms of both frequency and cost? Many commercially available models skip rebuild cost entirely and only forecast full replacement. That is a defensible simplification, but only if you say so out loud.
Renewals and rebuilds typically sit in the maintenance budget even though they get capitalized, and there is an inverse relationship worth naming: assets in worse condition that are not renewed on schedule drive higher ongoing maintenance cost. A good model tracks both the capital side and the capitalized maintenance side so you can compare them.
With these four pieces, you are ready to produce something. Open Excel and build as much granularity in the systems, subsystems, and assets as you can.
Subsequent refining of the base R&R forecast can be prioritized based on which systems and subsystems have the most impact on financial needs.
Are You Really Doing Asset Management?
The Institute of Asset Management and ISO 55000 define an asset as anything of value, measured against total cost of ownership, the whole-life cost. You cannot know that number without a baseline, and you cannot manage risk, the effect of uncertainty on objectives, without a baseline to deviate from.
You will never maintain a high-reliability, low-risk operation if you don’t do this forecast.
The Side Benefits Are Almost as Valuable as the Number
The forecast itself is the point: a defensible answer to whether you are climbing a financial mountain or on a flat stretch of road. But building the forecast pays off in four other ways that rarely get mentioned up front.
The Forecast Tests Your Data
Check the ‘asset type’ and ‘asset install date’ fields against each other and you will find problems: PVC dated to 1921, when PVC pipe was not manufactured until the early 1970s, or cast iron installed in 2016, which was almost certainly ductile iron. Expect to lose real ground here.
You may believe your data is 80 percent complete, and once you cross-check install dates against material and condition, you will likely fall back to 65 or 70 percent. That is normal, and it is the whole point of doing this exercise: it is the best way to check and cleanse your data.
It isn’t just the money, or even the insight about the money. It’s the insights the data gives you.
The R&R forces you to clean your data, because nonsense install dates and missing values surface the moment you try to run the model.
Defining the Meaning of “Rebuild”
It forces your organization to agree on what “rebuild” actually means, in terms of both frequency and cost, instead of letting every department define it differently.
Where You Are Overbuilding and Overspending
The R&R forecast can reveal places where you have already been unintentionally conservative, quietly overbuilding and overspending on the engineering side because nobody had put a hard number on what “safe enough” actually costs.
Turning Around the Budget Conversation
An R&R forecast changes how you ask for money. A funding request set at the median of the distribution tells leadership, or a federal funding agency, that you have a 50 percent chance of success. If that is not enough certainty, show them what 75 or 80 percent confidence actually costs, and let them choose the risk they are willing to accept. That reframes the conversation from “give me this number” to “here is what each funding level buys you in probability of success.”
A tornado diagram, built from the same sensitivity analysis, ranks the subsystems that swing the total forecast the most. Whatever sits at the top deserves your condition assessment budget and your preventive maintenance attention. Effort spent lower on that list is largely wasted. It is also worth checking the forecast’s near-term needs against your existing capital improvement program. In most organizations, roughly 80 percent of what the model flags is already on the program. The missing 20 percent is where the real conversation with your capital planning team needs to happen.
Point Estimates Are a Trap
The next level of sophistication is moving from a single point estimate for the inputs to a probabilistic range. Enhancing the forecast with probabilistic estimates involves taking your Excel model, applying a Monte Carlo simulation add-in, and assigning distributions to your inputs. It's not simple but it's not hard, either. And Monte Carlo analysis is all about uncertainty analysis, and uncertainties are rampant in facilities and infrastructure.
Example 1: Cost and Useful Life
Here is a trap with point estimates. Ask a room of engineers what a submersible pump is worth or how long it will last. You’.ll get many answers, at best, and analysis paralysis, at worst.
In most cases, no one has bought that specific one in 20 years. It could cost $4,000 or $6,000. And it could last 15 years or 30. So use the range and apply a probability distribution to it.
Let a Monte Carlo simulation draw a random value from that range for every asset, across thousands of runs, and look at the distribution rather than fighting over a single number.
Example 2: Deferred Maintenance
A well-built probabilistic forecast tells you as much about your data as it tells you about your money. A hump early in the curve is usually the signature of deferred maintenance: the model is telling you assets are already behind schedule. Watch how wide the range is around that hump, too. A wide band that narrows once the initial surge clears usually means you have reached equilibrium.
Width itself is a risk signal. On one groundwater plant forecast, the low end of the range barely moved, meaning there was little chance of spending less than expected, but the high end carried real exposure to two or three times the baseline cost. That asymmetry is exactly where risk management effort belongs.
Example 3: Falsely Lulled to Sleep
The clearest example I have seen involved a set of century-old force mains. The probabilistic range of the forecast came back hugging zero. I naturally thought we had a programming error in the model. We confirmed all was ok. Then we checked the inputs.
Every pipe was listed as 100 years old and in good condition, because nobody had ever verified the data or performed a condition assessment. Yet the database was marked as “100 percent complete.”
Several replacement values were missing or entered as $10, with nobody sure if that meant $10 a foot, $10 a mile, or just $10. Essentially, with 100-year-old, good-condition pipe, there was very little that had issues in our 20-year window. And when there was an issue, it cost almost nothing.
The O&M team responsible for those force mains had been telling leadership there was no problem and to spend the money elsewhere, based entirely on data that had never been checked.
The forecast did not confirm the assets were fine. It confirmed the data was broken. The probabilistic analysis was unused to determine uncertainty where there was assumed to be none.
Example 4: Engineering Conservatism Equals Financial Challenges
I saw this play out with a small utility client. A prior consultant’s point-estimate forecast struck leadership as too high. Run probabilistically, leadership’s instinct proved correct: the original estimate sat near the 85th percentile of the distribution, not the middle.
To avoid running out of capacity, the consultant built in huge safety margins, turning an engineering precaution into a costly financial liability. That tension between conservative engineering and disciplined finance is worth naming explicitly in your own organization, and it is exactly the kind of unintentional overbuilding a probabilistic forecast is built to expose.
In the end, the utility decided that 50 percent certainty of not running out of capacity over the long term was something they could afford, not 85 percent certainty.
Build a Renewal & Replacement Forecast Now!
You cannot know whether you are about to climb a financial mountain or on a flat stretch of road without an R&R forecast telling you which one it is. You cannot optimize the life cycle value of your assets until you know what the baseline is. And you can’t really manage risk until you know the expectations.
Use your existing data and build the model in Excel if that is all you have. The forecast will not be perfect on the first pass, but those insights help you prioritize future activities to make it better. Every renewal and replacement forecast improves with each cycle, but only if a first generation gets built.
So show me the money! What is your facility or system going to cost over the next 20 years, and what is your data telling you about that number?
See a related video from IMC:
Need help getting started? JD Solomon Inc. provides practical solutions for developing long-term financial forecasts using the data that you already have, improving your asset management program, and applying both simple and advanced techniques.
JD Solomon is the founder of JD Solomon, Inc., the creator of the FINESSE Fishbone Diagram®, and the co-creator of the SOAP criticality method©. He is the author of Communicating Reliability, Risk & Resiliency to Decision Makers: How to Get Your Boss’s Boss to Understand and Facilitating with FINESSE: A Guide to Successful Business Solutions.
Comments