Sanjib Das: Planned vs Unplanned: Spare Strategies That Reduce Downtime (Part 3)

(Part 3 of our series with reliability engineer Sanjib Das)

In Part 1, Reliability Expert Sanjib Das showed how a simple, explainable scoring model removes emotion from stocking decisions.
In Part 2, we saw how “just in case” inventory quietly drains money through shelf-life failures, preservation tasks and carrying cost.

Now, in Part 3, we turn to the moment of truth for any spare strategy:
What happens when equipment actually fails, whether planned or unplanned?

This is where Sanjib is clearest:
Spare parts do not prevent failure.
They compress the consequence window.

Planned failures: when one wrong spare derails everything

Turnarounds are the best example.
In heavy industry, a turnaround is not a maintenance job, it is an organisational event.
Planned two years ahead, built on validated failure data, and executed with precision.

Which is why Sanjib’s story hits so hard.

“It extended the turnaround two weeks because of these little spares… and that is a lot of money for a refinery.”

The cause was trivial on paper:

“We found that couple of thou shortage. It’s not much but you just cannot fit a part that is too short.”

A shaft off by a few thousandths.
A spare that appeared correct but had never been checked.
And because the plant had not kept the right spare, despite having millions of dollars of stock sitting unused, the entire turnaround schedule collapsed.

Why it happened

Sanjib’s diagnosis is brutally simple:

• Spare parts were stocked emotionally
• No criticality-based stocking discipline
• No verification process on arrival
• No alignment between planned maintenance and spare availability

This is why planned downtime overruns are predictable without a structured spare strategy.

Unplanned failures: you cannot stop them, but you can shrink them

Sanjib is clear that unplanned failures will always happen:

“You have no control on the unplanned failure.”

But a spare strategy does control the consequence:

“You cannot resist failure, but you can reduce the consequences… instead of two weeks you can do it in three days.”

The logic is straightforward:

• right spares = faster restoration
• faster restoration = lower production loss
• lower loss = higher reliability, safer operations and calmer maintenance teams

The mistake many organisations make is assuming unplanned failures require a different approach. In reality, the same factors that drive planned maintenance stocking decisions (criticality, lead time, failure behaviour, consequence) also determine how much of an unplanned failure you can absorb.

Reorder points and minimums: the quiet backbone of reliability

One understated insight from Sanjib’s work is that a spare strategy is not just a list. It’s a policy.

His Excel tool, and the AI-based equivalent in SPARROW.Plan, calculates:

• minimum stocking levels
• reorder points
• consumption logic across identical assets
• differentiation between critical, medium and low consequence locations

As Sanjib explains:

“My tool… calculates also: I have 10 similar pumps… what should be your minimum, your reorder point…”

This eliminates the common pattern where plants:

• stock one part “just in case”
• consume it once
• forget to reorder
• discover the gap only during a failure

Reorder discipline is what turns a spare strategy into a reliability instrument, not just warehouse data.

Vendor-managed inventory vs in-house expertise: the cost of outsourcing thinking

One of the most practical parts of Sanjib’s experience is his critique of vendor-managed spares.

“Sometime we give that to a vendor because we do not have a tool… but that is costlier actually.”

Why?

1. Vendors optimise for coverage, ‍not your specific risk

They must satisfy every customer profile and don’t know where and how you’re going to use a spare part. So their lists tend to be long and expensive.

“They come out with lists which are just not optimised.”

2. Vendors cannot know your environment

The same part behaves differently in:

• aggressive media vs clean media
• continuous duty vs intermittent duty
• harsh vs controlled environment

Vendors cannot build failure modes into stocking guidance.

3. You lose organisational learning

Once stocking decisions leave the plant, so does the ability to link:

• real failures
• maintenance insights
• asset behaviour
• stocking adjustments

This is why Sanjib prefers an internal, mathematically consistent method:

“Using the tool will be the best option… that reduces the cost significantly.”

And this is precisely why SPARROW.Plan applies Sanjib’s logic in-house, not via a vendor’s generic catalogue.

Why SPARROW.Plan strengthens both planned and unplanned readiness

SPARROW.Plan mirrors the thinking Sanjib developed manually:

• criticality determines stocking priority
• lead time defines reorder points
• failure behaviour shapes consumption strategy
• consequence drives local vs regional stocking
• master data from SPARROW.Clean prevents ordering errors
• cross-site usage data identifies pooling opportunities

Driven by AI, SPARROW.Plan recalculates readiness every time:

• a failure occurs
• a part is consumed
• a supplier’s lead time changes
• assets or BOMs are updated
• a new spare is requested

This turns stocking into a living, dynamic reliability process, not a static drawing.

Where we go next

With a foundation in:

• Part 1 — structured decision-making
• Part 2 — the hidden cost of holding the wrong parts
• Part 3 — how spares shape both planned and unplanned outcomes

Part 4 now zooms out:
What if you didn’t need to stock every medium-critical item at every site?

Sanjib has seen global refineries in Singapore, Malaysia and Thailand save huge amounts by creating regional spare pools for medium-critical items, without increasing operational risk.

That’s where we go next.

👉 Part 4: When to Pool Inventory

‍