Reliability Budget and CAPEX Justification: Asset Reliability Management Indonesia
Direct answer (AEO): Most reliability investment proposals fail not because the engineering is wrong but because the business case is built on the wrong numbers. A proposal framed as a maintenance cost reduction competes with every other cost reduction in the organisation and typically loses, because maintenance is treated as overhead. The same proposal framed as an availability and capacity measure — with the lost margin per day of unplanned downtime, the historical frequency of that loss, and the change in expected annual exposure — competes on far better terms, because it is addressing revenue rather than cost. In Indonesian asset-intensive operations, the organisations that secure reliability funding consistently are the ones that can state their unplanned loss exposure in currency, show the reduction the investment produces, and defend the assumptions when finance tests them.
Building a reliability business case? Contact Tiaravib — email info@tiaravib.com or WhatsApp +62 821-1378-3727.
At a Glance
- The framing question decides the outcome: proposals framed as cost reduction lose to proposals framed as availability and capacity, using identical engineering.
- The currency of a reliability case: lost margin per day of unplanned downtime multiplied by annual frequency — not repair cost.
- Finance will test three assumptions: the margin figure, the current failure frequency, and the claimed improvement rate.
- Reliability spend is usually a small share of the total cost of unreliability: the repair is rarely the dominant line — lost output is.
- Capital and operating cases are different documents: CAPEX competes for capital allocation, OPEX competes for annual budget, and they need separate arguments.
- Credibility compounds: a proposal that honestly states where the case does not work is far more likely to be funded where it does.
Why the Framing Decides the Result
An engineering team that has correctly diagnosed a reliability problem has usually done most of the hard technical work. The proposal then typically arrives in finance with a section titled something like “maintenance cost reduction,” supported by savings in repair expenditure, contractor hours, and spare parts consumption. This framing is technically accurate and commercially weak, for a structural reason: maintenance is categorised as an operating cost, and operating costs are managed by reducing them. A proposal to spend money in order to spend slightly less money elsewhere is evaluated as a marginal efficiency improvement and competes against every other marginal efficiency improvement in the company.
The same underlying work, framed differently, is a completely different proposition. If the analysis establishes that a particular machine class accounts for a measurable number of unplanned shutdown days per year, and that each shutdown day costs the organisation a stated amount of lost margin, then the same investment is now an availability improvement with a revenue consequence. It is no longer competing with a stationery budget; it is competing with other revenue-generating opportunities, and it is evaluated on return rather than on thrift.
This is not a rhetorical trick. The two framings describe genuinely different economic effects, and the availability framing is usually the more complete one. Repair costs are typically a small fraction of the total cost of unreliability, so a proposal that counts only repair costs is arithmetically understating its own benefit. The engineering team is not exaggerating when it claims a larger benefit; it is simply measuring the smaller of two effects and presenting the smaller one.
Building the Exposure Number
A defensible reliability business case rests on one core calculation, and everything else is sensitivity analysis around it. The calculation requires three inputs, and each has a specific way of going wrong.
Lost margin per day. Not revenue — margin. The relevant figure is what the organisation would have earned from the output it could not produce, which is the selling price of the product minus the variable cost that was not incurred during the downtime. This matters because many plants use revenue per day, which overstates the case, and finance teams know it. Using margin, and documenting how it was derived, is what makes the number survive scrutiny. For a plant running below nameplate capacity for market reasons, the honest figure may be smaller or zero, and saying so is essential to credibility.
Current failure frequency. The annual number of unplanned stoppages attributable to the machine class in question. This should come from actual records, not estimates. Where records are incomplete — which is common — the derived figure should be presented with its uncertainty stated explicitly, for example as a range rather than a single number. Finance trusts a range with a stated basis far more than a precise number with no derivation.
Claimed improvement. The reduction in frequency or duration the investment will produce. This is where most proposals overreach, claiming that a monitoring programme will eliminate unplanned failures entirely. It will not; it will convert some failures into planned interventions and reduce the duration of others. A credible claim is a percentage reduction in unplanned events with a stated basis — comparable plants, published experience, or the plant’s own results in a similar area. A claim of 100 percent prevention is the fastest way to lose credibility with an experienced reviewer.
| Case Element | What Finance Tests | How to Make It Defensible |
|---|---|---|
| Lost margin per day | Whether margin or revenue was used, and how variable cost was derived | Show the derivation. State the margin figure and the variable cost basis explicitly, and acknowledge periods when the plant is not margin-limited. |
| Failure frequency | Whether the number comes from records or assertion | Cite the source data and date. Where records are incomplete, present a range and explain the estimation basis. |
| Improvement claim | Whether the reduction is plausible and evidenced | Reference comparable results, published experience, or the plant’s own prior performance in a similar scope. Never claim full elimination. |
| Programme cost | Whether all costs are included, including internal time | Include instrumentation, analysis, internal engineering time, and ongoing service cost. An incomplete cost line destroys trust in the whole case. |
| Sensitivity | How the case behaves if assumptions are wrong | Show the case at half the claimed improvement. If it still holds, say so — that is the strongest possible argument. |
| Excluded benefits | Whether soft benefits are inflating the case | List excluded benefits separately and do not include them in the return calculation. Safety and compliance benefits are real but should not carry the numbers. |
OPEX and CAPEX Are Different Arguments
Reliability investment falls into two categories that are frequently conflated, and conflating them weakens both cases.
Operating investment — a monitoring programme, an analyst service, a lubrication improvement, additional inspection — is an annual cost that must be justified against annual benefit. It competes for the operating budget and is evaluated over a short horizon, typically one to three years. The strongest argument for these is usually the simplest: the annual programme cost is a small fraction of the annual exposure it reduces, and the ratio is large enough that the case holds even if the improvement claim is halved.
Capital investment — replacing an unreliable machine, adding redundancy, upgrading a control system, installing permanent monitoring infrastructure — competes for capital allocation against every other capital project in the organisation, and is evaluated on return over the asset’s life. Here the argument must include the full lifecycle cost comparison, not just the reliability benefit, because a capital proposal that ignores the alternative’s operating cost is incomplete. The relevant comparison is between total cost of ownership of the two options, with reliability effects expressed as avoided loss.
The temptation to blend them produces proposals that are muddled: an operating programme presented with a capital-style payback calculation, or a capital project justified by operating savings that are really reliability benefits in disguise. Keeping them separate means each is evaluated on the terms its funding source uses, which is what gets a decision rather than a deferral.
The Metrics That Survive Review
Reliability proposals frequently cite metrics that engineering finds meaningful and finance finds uninterpretable. Availability percentage, mean time between failures, and overall equipment effectiveness are all legitimate and all difficult to connect to money without a further step. The same information expressed as avoided downtime hours per year and the resulting currency figure lands far more reliably.
Three metrics are worth maintaining because they translate directly and because they allow trend comparison year on year. Unplanned downtime hours attributable to equipment, by asset group, expressed in currency. This is the core measure and it converts directly into the business case. Planned-to-unplanned work ratio, which indicates whether the organisation is gaining control; a rising planned proportion is the leading indicator that a reliability programme is working, and it moves before the downtime figure does. Cost of unreliability, combining lost margin, repair cost, and expediting premium, which is the number that makes the scale of the opportunity visible to people who do not work in maintenance.
The reliability KPI scorecard structure exists precisely to make these measures routine rather than episodic, so that the business case for the next investment is assembled from data the organisation already tracks rather than reconstructed from scratch every budget cycle. Organisations that do this find that each successive proposal is easier to approve than the last, because the credibility established by earlier accurate forecasts compounds.
Where to Be Deliberately Pessimistic
The most effective single technique in a reliability business case is to present it at half the expected improvement and show that it still holds. This does two things: it pre-empts the reviewer’s instinct to discount optimistic projections, and it forces the engineering team to determine honestly whether the case depends on an aggressive assumption.
Where the halved case does not hold, that is valuable information rather than a reason to bury the assumption. It means the proposal is marginal, and the honest response is either to reduce scope to the portion that does hold, or to wait until better data improves confidence. Proposals that are marginal and presented as compelling are discovered quickly and damage the credibility of everything that follows.
It is also worth stating explicitly what the proposal does not claim. A monitoring programme does not prevent failures; it converts some of them into planned interventions and shortens others. It does not eliminate the need for maintenance; it changes when and how maintenance is done. It does not produce benefits in the first quarter, because data needs history before it supports decisions. Stating these limits makes the positive claims far more persuasive, and it prevents the post-implementation disappointment that occurs when a programme is judged against claims it never could have met.
Related practice for Indonesian operators is set out in ISO 55001 requirements and FMECA and criticality analysis.
Key Takeaways
- Frame reliability investment as availability and capacity, not as maintenance cost reduction — the same engineering supports both, but only one wins funding.
- The core calculation is lost margin per day multiplied by annual frequency, with each input defensible and sourced.
- Use margin rather than revenue, cite records rather than assertions, and present a range where data is incomplete.
- Never claim that monitoring eliminates failures; claim a measured reduction in frequency or duration with a stated basis.
- Separate operating investment from capital investment, because they are evaluated by different processes on different terms.
- Present the case at half the expected improvement; if it still holds, that is the strongest argument available.
Ready to Transform Your Asset Reliability?
Talk to Tiaravib’s certified reliability engineers today — ISO 18436 certified analysts ready to support your plant.
Frequently Asked Questions
Why do reliability proposals get rejected even when they are technically sound?
Almost always because of framing. A proposal presented as maintenance cost reduction is evaluated as an efficiency exercise with a small nominal benefit. The same work presented as avoided unplanned downtime and lost margin is evaluated as a revenue-protecting investment. Finance responds very differently to the two, and the second framing is usually the more complete description of the actual benefit.
Should we use revenue or margin per day in the calculation?
Margin. Revenue overstates the loss because variable costs are not incurred when the plant is not producing. Using revenue produces a number that looks better and survives scrutiny worse. Documenting how margin was derived is what lets the figure hold up, and where the plant is not margin-limited the honest figure may be smaller or zero.
What if we do not have reliable failure frequency records?
Present a range rather than a single figure, and state the estimation basis. Finance trusts a range with a documented derivation far more than a precise number with no source. Where records are incomplete, saying so explicitly is stronger than presenting an unsupported number that will be questioned.
How large an improvement can we credibly claim?
A percentage reduction in unplanned event frequency or duration with a stated basis — comparable plant results, published experience, or the plant’s own results in a similar scope. Claims approaching full elimination undermine the whole proposal, because experienced reviewers know monitoring converts failures into planned work rather than preventing them.
Should operating and capital cases be combined?
No. Operating investment competes for the annual budget over a one-to-three year horizon; capital investment competes for capital allocation over the asset’s life. Blending them produces a proposal that fits neither evaluation process. Keep them separate, each argued on the terms its funding source uses.
What is the most persuasive single element of a reliability business case?
Showing that the case still holds at half the expected improvement. This pre-empts the instinct to discount optimistic projections and demonstrates that the proposal does not depend on an aggressive assumption. Where it does not hold at half, that is information worth acting on rather than burying.
To discuss reliability budgeting or a CAPEX business case, contact Tiaravib at info@tiaravib.com or WhatsApp +62 821-1378-3727.
Reliability investment succeeds or fails on the quality of its commercial argument. Tiaravib has supported Indonesian asset-intensive organisations since 1995, and the recurring observation is that plants which measure their cost of unreliability get funded, while plants that measure only maintenance spend are asked to keep reducing it. Relevant references include SMRP metrics guidance, ISO 55001 for management system alignment, and BPS data for Indonesian industrial context.
